...... A SYSTEMATIC STUDY OF THE AMERICAN SPECIES OF ALNUS (BETULACEAE) Dissertation for the Degree of Ph... D MICHIGAN STATE UNIVERSIWL . i JOHN JACOB FURLOW < ' ' 1974 IIILIIIIIIHIIILIIIILIIIIIIIIIIIIIIIIIIIIIIHHI I LIBRARY * Iichégw hate I ABS TRACT A SYSTEMATIC STUDY OF THE AMERICAN SPECIES OF ALNUS (BETULACEAE) By John Jacob Furlow The genus Alnus consists of about 20 species of shrubs and trees, mainly of the Northern Hemisphere, but extending below the Equator into South America along the Andes. Although a wide array of studies of the genus during the past 70 years have produced an abundance of new data, and this period has seen the development of different philosophies concerning taxa in general, there has been no compre- hensiverevisionary effort in that time. Consequently, current treatments of $1.215 suffer from a lack of up-to-date information and unified taxonomic concepts. The species are often variable, and this has led to both widespread confusion in the treatment of taxa and a proliferation of infraspecific and Specific names. ' In addition, where the geographical ranges of taxa overlap, the plants appear to hybridize readily, adding to the problems of identification and classification. In this study the evolution, phytogeography, ecology, morphology, anatomy, reproductive biology, cytology, and economic importance of the genus are considered. Each subgenus, species, subspecies, and may I EISSEI. I t: EagicaI, tericar md an; II he I ":erIc these I iiiionz John Jacob Furlow varnny’is described, its range outlined, and its nomenclature dis— cussed. A traditional taxonomic investigation involving detailed morpho- loghxflq field, bibliographic, and nomenclatural study of all the American taxa of Alnus is supplemented with data from studies of wood anatomy, pollen structure, and phenolic constituents. Analyses of fluaumrphological and chemical data were accomplished using Inmmrical taxonomic procedures. In most cases, the results of flmme investigations are parallel to those obtained using the tra- cfiiional methods. In a few cases, however, they differ, and where flfis is so, reasons for the discrepancies are considered. In the numerical analyses, both phenetic and phyletic methods are employed to determine relationships among the taxa. Seventy Wflfls, representing each morphological variant in different parts ofiis geographical range, were scored for 150 characters. Separate muflyses were then performed for the vegetative, floral, and fruit characters in addition to the complete data set. A cluster analysis of'fluacharacters themselves, together with principal components muflyses, were used to determine traits having possible common gmufific control. Phyletic analyses were completed using the method <fl3Wagner (1962), in which a divergence index, based on the estimated advancement of the characters, is used to predict evolutionary rela- tionships. In general, the phenetic and phyletic analyses agree very (flosely, both with each other and with traditional classifications. FTom these studies it is shown that floral characters are often vaIuabIe from veg a consic slated shown I sinstar {agrapl aha mugs 1'. III are I John Jacob Fur low valuable in obtaining phylogenetic information not readily determined from vegetative features, which tend to be more plastic. In addition, a considerable amount of convergent evolution in the genus is demon— strated, and a relatively small number of complexes of taxa are shown to exist in the New World. The chemosystematic investigation includes an analysis of phenolic substances extracted from dried foliage and separated by paper chroma- tography. The data obtained were analyzed by means of the same taximetric procedures discussed above. In most cases, taxonomic groups cluster in the same ways seen for the morphological data. An interesting exception is Alnus maritime, which appears to be more closely related to the Latin American species in the chemical analysis. Possible implications of this fact in regard to the history of the genus in North America are discussed. From the results of the palynological, anatomical, chemosystematic, and numerical taxonomic studies, as well as ordinary observation, it is seen that the American taxa of Alnus group into three major clusters, which are treated as subgenera (Alnus, Alnobetula, and Clethropsis). Four taxa are regarded as conspecific with Eurasian species (Alnus incana subsp. rugosa, A. incana subsp. tenuifolia, A. viridis subsp. cris a, and A. viridis subsp. sinuata). The Latin American taxa are shown to comprise two species, Alnus acuminata and A. jorullensis, each of which is subdivided into two varieties (A. acuminata var. acuminata and A. acuminata var. glabrata on the one hand, and A. jorullensis var. jorullensis and A. jorullensis var. firmifolia on the other). \— John Jacob Furlow Subgenera Clethropsis and Alnobetula are represented by single Species in America (Alnus maritime and A. viridis, respectively). Subgenus Alnus is composed of two more or less distinct groups of taxa, one represented by the shrubby northern A. incana and A. ser- rulata, and the other by the Latin American species and the large tree Species of the northern and central sections of the western part of the continent (Alnus rubra, A. rhombifolia, and A. oblorflfolia). The latter group is regarded as the most primitive segment of the genus in the New World. _,__.. _i.. A SYSTEMATIC STUDY or .THE AMERICAN SPECIES OF ALNUS (BETULACEAE) By John Jacob Furlow A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Botany and Plant Pathology 1974 ACKNOWLEDGMENTS I am most grateful to Dr. John H. Beaman for his interest, en- couragement, and guidance throughout this study and for providing the principal facilities and materials with which it was conducted. I am especially indebted to him for a research assistantship funded through a grant from the National Science Foundation (No. GJ—573) under which he was Principal Investigator. This grant supported computer use and a portion of the field work. Dr. Beaman also supplied funds from a Ford Foundation Grant (No. 71—3511), administered through the Michigan State University Latin American Studies Center, to support travel in Mexico, and he accompanied me during a part of this field work. Sincere appreciation is also extended to Dr. James W. Hanover for his assistance and for the use of his laboratory and equipment in the chemosystematic investigations. I am grateful to Dr. Theodore J. Crovello for help with the numerical taxonomic phase of the study and for his generosity in supplying the computer programs used in that Work. Thanks go, in addition, to Mr. Larry E. Morse for providing me With his computer programs and for working with me in preparing the taxonomic data matrices to use with them. I am likewise indebted to Dr. William Tai for the use of his microscopes and photographic equip- ment and for his helpful advice, to Dr. Melinda E. Denton for her guidance during a portion of the study, and to Dr. Stephen Stephenson w. "" for his interest and assistance. Drs. Beaman, Hanover, Stephenson, and Tai read the entire manuscript; parts were also reviewed by Dr. Denton, Dr. Crovello, Dr. Ralph E. Taggert, and Miss Lynn E. Murray. The valuable suggestions made by all of these individuals are greatly appreciated. Specimens for study were collected by Dr. Beaman, Dr. Denton, Dr. James H. Anderson, Dr. Loran C. Anderson, Dr. C. Eugene Jones, and Mr. Warren Douglas Stevens. I wish to thank the curators of the herbaria I visited and from which specimens were borrowed for their helpfulness. These herbaria are listed at the beginning of the taxonomic treatment. Computer work was done at the Computer Laboratories of both Michigan State University and The University of Notre Dame. Scanning electron microscopy was carried out in the Electron Microscope Labora- tory of Michigan State University. Thanks are extended to the directors of each of these facilities. Much of the field work was financed by a Doctoral Dissertation Grant from the National Science Foundation (No. GB-28662), and this award is gratefully acknowledged. Support was also provided by the Department of Botany and Plant Pathology of Michigan State University in the form of both teaching assistantships and a graduate fellowship. Sincere thanks are extended to my parents, Mr. and Mrs. Wilbur S. Furlow, and to my parents-in-law, Mr. and Mrs. Donald H. Lotz, for both moral and financial support during this effort. Finally, without the devoted and unselfish support and assistance of my wife, Gretchen, this work would not have been possible. iii m mun—“l" “Wm—m mfg-duh." ' ‘n‘Ju' c -. ' :‘-—-_.-_..a.. - 'LD .1.- E" ' TABLE OF CONTENTS _P_agg LIST OF TABLES . . . . . . . . . . . . . . . . vii LIST OF FIGURES . . . . . . . . . . . . . . . ix LIST OF PLATES . . . . . . . . . . . . . . . . xii LISTOFPRINTOUTS............... xvii INTRODUCTION . . . . . . . . . . . . . . . . l HISTORICAL ACCOUNT . . . . . . . . . . . . . . 4 TAXONOMIC CONCEPTS . . . . . . . . . . . . . . 8 EVOLUTION AND PHYTOGEOGRAPHY . . . . . . . . . . . 11 Origin of the Genus and Subgenera . . . . . . . . 12 Dispersal in the New World . . . . . . . . . . 16 Historical Factors Affecting Distributions . . . . . l9 ECOLOGY . . . . . . . . . . . . . . . . . . 26 MORPHOLOGY AND ANATOMY . . . . . . . . . . . . . 32 Habit . . . . . . . . . . . . . . . . . 32 Stems . . . . . . . . . . . . . . . . . 32 Stipules . . . . . . . . . . . . . . . . 47 Winter Buds . . . . . . . . . . . . . . . 47 Leaves . . . . . . . . . . . . . . . . . 54 Inflorescences . . . . . . . . . . . . . . 66 Flowers . . . . . . . . . . . . . . . . 7O In IRUDU CRIHIIS IIIIIIII Pollen. . . . Fruits . . . . REPRODUCTIVE BIOLOGY . CHROMOS OME NUMBERS . NUMERICAL TAXONOMIC STUDIES Phenetic Studies Methods and Materials . Results . . Phyletic Studies Taxonomic Data Matrix Discussion . . CHEMOSYSTEMATIC STUDIES Methods and Materials Results . . . Discussion . . PHYLOGENETIC TRENDS . Studies Divergence and Convergence . Phylogeny . . . ECONOMIC IMPORTANCE . TAXONOMIC TREATMENT . Alnus . . . . Artificial Key to the Infraspecific Taxa . Alnus subgenus Alnus 1. A. rubra 2. A. oblogggifolia . Subgenera, Species 92 100 103 105 105 110 137 145 154 161 163 165 180 190 190 192 196 200 201 204 209 212 225 Alnus ‘ Inc EEK-DIX “In 6. A. incana . . . . . . . 6a. A. incana subsp. rugosa 6b. A. incana subsp. tenuifolia 7. A. serrulata . . . . . . 8. A. glutinosa . . . . . . Ami subgenus Alnobetula . . . . 9. A. viridis . . . . . . 9a. A. viridis subsp. crispa A. rhombifolia . . A. acuminata . . . 4a. A. acuminata var. acuminata 4b. A. acuminata var. glabrata A. jorullensis . . 5a. A. jorullensis var. jorullensis 5b. A. jorullensis var. firmifolia . 9b. A. viridis subsp. sinuata Alnus subgenus Clethropsis . 10. A. maritime . . . Uncertain and Excluded Names APPENDIX . LITERATURE CITED vi 0 255 286 295 298 311 320 324 345 360 382 391 393 397 418 430 432 442 445 481 Table 10 12 13 LIST OF TABLES Size of Alnus pollen grains (microns) . , . . Percent of Alnus pollen grains having various numbers of pores . . . . . . . . . . . Chromosome numbers reported for American species of Alnus and their Eurasian vicariants . . . . Operational taxonomic units (OTU's) used in phenetic analyses of Alnus . . . . . . . . Characters in the complete morphological data set for Alnus found to be highly correlated . . Average distance between each pair of taxa of Alnus based on analysis of the llS-character morphological data set . . . . . . . . . Morphological characters in Alnus highly cor- related with the first nine principal components Taxa of Alnus having correlations greater than a 0.500 with each of the first 19 principal components using morphological characters . . . Characters in Alnus having coefficients of variability of at least 0.5 among taxa . . . . Characters used for phyletic analysis of Alnus taxa . . . . . . . . . . . . . . . State of each character for each taxon in the phyletic analysis of Alnus species . . . . . Unique phenolic compounds found in taxa of Alnus Number of distinctive compounds (always present in one taxon and always absent in the other) for each pair of taxa of Alnus . . . . . . o 86 101 108 111 121 124 125 139 141 168 169 Table 14 15 16 17 18 19 20 21 22 23 Chemical similarities between each pair of taxa of Alnus using the paired affinity index . . . Chemical similarities between each pair of taxa of Alnus using the coefficient of association . Chemical similarities between each pair of taxa of Alnus using the distance coefficient . . . Phenolic compounds in Alnus having a 0.600 or greater correlations with the first seven principal components . . . . . . . . . . Mean chemical affinities among taxa of Alnus using the paired affinity index, the coefficient of association, and the distance coefficient . . Herbarium specimens from which pollen was obtained for study . . . . . . . . . . Characters used in numerical analyses . . . . Specimens of Eurasian taxa of Alnus used in the collection of data for numerical analysis . . . Sources of Specimens from which foliage samples for chromatographic analysis were obtained . . R values and color reactions of butanol-soluble compounds in Alnus foliage . . . . . . . . viii 172 173 181 188 445 448 457 459 462 071 H m LIST OF FIGURES Figure Page 1 Outlines of leaves of Alnus showing extremes in the patterns of shape and margin variation: 1-10, Alnus acuminata var. acuminata (South America); 11— 20, A. acuminata var. acuminata (Mexico and Centralm erica); 21- 27, A. acu- minata var. glabrata . . . . . . . . 56 2 Outlines of leaves of Alnus showing extremes in the patterns of shape and margin variation: lw4, Alnus oblon ifolia; 5-9, A, incana subsp. rugosa; 10—14, A. incana subsp. tenuifolia; 15 21, A. 'orullensis var. firmifolia; 22- 30, A. jorullensis var. jorullensis; 31- 35, A. maritime . . . . . . . . . . . 58 3 Outlines of leaves of Alnus showing extremes in the patterns of shape and margin variation: 1- 7, Alnus rhombifolia; 8- 15, A. rubra, 16- 20, A. serrulata; 21- 28, A. viridis subsp. crispa; 29- 35, A. viridis subsp. sinuata . . . . . . 6O 4 Range of variation in diameter of Alnus pollen grains . . . . . . . .I . . . . . 85 5 Phenogram showing relationships among 28 taxa of Alnus based on cluster analysis of 70 OTU's and 150 morphological characters . . . . . . . 114 6 Phenogram showing relationships among 70 OTU's of Alnus based on cluster analysis of 70 OTU's and 115 morphological characters . . . . . . . 116 7 Phenogram showing relationships among 28 taxa of Alnus based on cluster analysis of 70 OTU's and 115 morphological characters . . . . . . . 118 8 Projection of taxa of Alnus according to factor scores for principal components I and II using morphological data . . . . . . . . . . . 128 ix Prc sec 110! Figure 9 10 l3 l4 17 18 19 20 21 22 23 acuminata in Mexico and Central America Projection of taxa of Alnus according to factor scores for principal components I and III using morphological data a o o o o o o o n n Projection of taxa of Alnus according to factor scores for principal components I and IV using morphological data . . . . . . . . . . Phenogram showing correlations among morpho- logical characters of Alnus based on cluster analysis of 70 OTU's and 150 characters O n O Presumed phyletic relationships among Alnus taxa based on the weighted divergence index Composite chromatogram of phenolic compounds from Alnus foliage Phenogram showing chemical relationships among taxa of Alnus using dichotomous data and the paired affinity index . . . . . . . . . Phenogram showing chemical relationships among taxa of Alnus using dichotomous data and the coefficient of association . o o o a e n o Phenogram showing chemical relationships among taxa of Alnus using frequency of compound occurrence and the distance coefficient Projection of taxa of Alnus according to factor scores for principal components I and II using chemical data 0 o I a o o o o Projection of taxa of Alnus according to factor scores for principal components I and III using chemical data Phylogenetic tree of the New World taxa of Alnus and their Eurasian vicariants . . . . . , Distribution of Alnus rubra Bong. q o 0 Distribution of Alnus oblongifolia Torr. . Distribution of Alnus rhombifolia Nutt. Distribution of Alnus acuminata H.B.K. var. Page 130 132 135 144 167 175 177 183 185 194 220 232 245 272 26 27 30 31 Distribution of Alnus acuminata H.B.K. var. acuminata in South America . o o o o o p o 0 Distribution of Alnus acuminata var. glabrata (Fern.) Furlow . . o o n c o a o o o o u Distribution of Alnus jorullensis H. B.K. var. jorullensis . . . . . . . Distribution of Alnus jorullensis var. firmi- folia (Fern. ) Furlow . Distribution of Alnus incana subsp. rugosa (DuRoi) Clausen . . . . . . . . . . Distribution of Alnus incana subsp. tenuifolia (Nutt.) Breitung in North America . . . . . . Distribution of Alnus serrulata (Ait.) Willd. . . Distribution of Alnus glutinosa (L ) Gaertn. in North America— . . . . . . . . . . Distribution of Alnus viridis subsp. crisps (Ait. ) Turrill in North America . . . . Distribution of Alnus viridis subsp. sinuata (Regel) Love & Love in North America o o 0 Q Distribution of Alnus maritime (Marsh.) Nutt. xi Page 274 292 307 317 333 352 371 389 425 439 W Hate Plate LIST OF PLATES Nodules of nitrogen—fixing endophytes on roots of Alnus serrulata. A, X l. B, X 2.2 . . . . . 28 Habits and habitats of representative species of Alnus. A, Alnus incana subsp. rugosa, large shrub on the bank of the Pigeon River in northern Indiana. B, A, rubra, medium sized trees forming a dense stand in west-central Washington. C, A. serrulata, medium-sized shrubs growing on the banks of a small stream in the Blue Ridge Moun- tains of southern Virginia. D, A, viridis subsp. crispa, medium-sized shrubs forming an understory in pine woods in central Alberta . . . . . . . 34 Habits and habitats of representative species of Alnus. A, Alnus oblongifolia, large trees growing in a stream on Mt. Graham, Arizona. B, A, jorul— lensis var. firmifolia, large tree on Volcan Popocatépetl, Mexico. C, A. rhombifolia, medium- sized trees on the banks of the Trinity River in northern California. D, A. viridis subsp. sinuata, large shrubs covering a subalpine mountainside in Glacier National Park, Montana . . . . . . 36 Bark of representative species of Alnus. A, Alnus oblongifolia, X 0.05. B, A, incana subsp. rugosa, X 0.3. C, A. jorullensis var. firmifolia, X 0.25. D, A, serrulata, X 0.5 . . . . . . . . . . 38 Photomicrographs of Alnus wood. A, Alnus acumi- nata var. acuminata, radial section showing aggre- gate rays and large vessels, X 160. B, A, 3227 minata var. acuminata, longitudinal section ‘ showing rays and details of the vessel wall, X 395. C, A, acuminata var. glabrata, longitudinal sec- tion showing uniseriate and biseriate rays and details of the perforation plate of a vessel, X 395. D, A. incana subsp. rugosa, radial section showing little tendency for aggregation in the rays, X 160. E, A, incana subsp. tenuifolia, Plate 10 11 12 radial section showing aggregation of rays and moderate- sized vessels,X 160. F, A. jorul- lensis var. jorullensis, radial section showing multiseriate rays, X 160 . . . . . . . Photomicrographs of Alnus wood. A, Alnus jorul- lensis var. jorullensis, radial section showing ”wax chambers," X 160. B, A, maritima, radial section showing small vessels and a small multi- seriate ray, X 160. C, A. rhombifolia, radial section showing a well—developed multiseriate ray, X 160. D!.é' viridis subsp. crispa, radial section showing small vessels and uniseriate rays, X 160. E, A. viridis subsp. crisps, longitudinal section showing the vertical nature of the rays, X 395. F, A, viridis subsp. Eifl' uata, radial section showing slight aggregation of the rays, X 395 . . . . . . . . . . Stipules and expanding leaves of representative species of Alnus. A, Alnus incana subsp. rugosa, X 1.5. B, A. maritima, X l. 5. C, A. rhombifolia, X 1.5. D, A. viridis subsp. crispa, X 1.5 . . Twigs with winter buds of representative species of Alnus. A, Alnus oblongifolia. B, A, maritima. C, A. rubra. D, A. serrulata . . . . .1 . . Twigs with winter buds of representative species of Alnus. A, Alnus incana subsp. rugosa. B, A, incana subsp. tenuifolia. C, A. rhombifolia. D, A, viridis subsp. crispa . . . . . . . Microscopic features of Alnus leaves. A, Alnus viridis subSp. crispa, pubescent extreme (A. crispa var. mollis Fern. ), X 3. B, same as A, X 46. C, éf* maritima, glandular teeth. D, A, rubra, revolute margin, X 3 . . . . . . . Epi-illuminated microscopic views of abaxial leaf surfaces of Alnus showing veins, resinous deposits, and glands, all X 46. A, Alnus acuminata var. acuminata. B, A, jorullensis var. firmifolia. C, A. 'orullensis var. jorullensis. D, A, viridis subsp. sinuata . . . . . . . . . . . . Abaxial leaf cuticle imprints of Alnus made on thin acetate sheets showing stomata, epidermal cells, hairs, and glands, all X 160. A, Alnus xiii Page 43 45 49 51 53 63 65 Plate , Page acuminata var. acuminata. Ba.é' acuminata var. glabrata. C, A. oblongifolia. D, A. incana subsp. tenuifolia. E:.é' jorullensis var. jorullensis. F, .A' maritima . . . . . . . . 68 13 Mature infructescences of representative species of Alnus. A, Alnus acuminata var. acuminata (South America). B, A. acuminata var. acuminata (Mexico). C’.é‘ acuminata var. glabrata. D, A. incana subsp. rugosa . . . . . . . . . 72 14 Mature infructescences of representative species of Alnus. A, Alnus incana subsp. tenuifolia. B, A, jorullensis var. firmifolia. C, A, jorul- lensis var. jorullensis. D, A, maritima . . . . 74 15 Mature infructescences of representative species of Alnus. A, Alnus rhombifolia. B, A, rubra. C, A. serrulata. D, A? viridis subsp. sinuata . . 76 l6 Pollen grains of representative species of Alnus, all X 1300. A, Alnus acuminata var. acuminata. B, A. oblongifolia. C, A. acuminata var. glabrata. D, A. incana subsp. incana. E, A. incana subsp. tenuifolia. F, A. jorullensis var. firmifolia. G’.é' jorullensis var. jorullensis. H, .A' mari— tima. I, A. rhombifolia. J, A. rubra. K, A. serrulata.— L, A. viridis subsp. crispa. M,TA. viridis subsp. sinuata. N, .A' viridig subsp. sinuata. 0’.é' viridis subsp. viridig . . . . . 81 17 Scanning electron micrographs of pollen grains of representative species of Alnus. A, Alnus incana subsp. rugosa, X 2500. B, A. incana subsp. ten- uifolia, X 1500. C, A. igrullensis var. igrullen- 313, X 2250. D, A. maritima, X 2250. E, A, rubra, X 2250. F, A, viridis subsp. crispa, X 2500. G, A. viridis subsp. sinuata, X 2250. H, A? viri— dis subsp. sinuata, X 2250 . . . . . . . . . 83 13 Fruits of representative species of Alnus, all X 12. A, Alnus acuminata var. acuminata. B, A. oblongifolia. C, A. incana subsp. rug— osa. D, A. maritima. E, A. serrulata. F, A, viridis subsp. sinuata . . . . . . . . 90 19 A, pistillate and staminate inflorescences of Alnus serrulata. B, syrphid flies visiting staminate catkins of Alnus viridis subsp. nlJ lu’ Pl: u 3.1!- Plate Page cris a in Terra Nova National Park, Newfound- land zphotograph courtesy of Garrett E. Crow) . . 94 20 Representative Specimen of Alnus rubra Bong. . . . 218 21 Right: holotype of Alnus oblongifolia Torr. Left: specimen of A, acuminata H.B.K. . . . . . 230 22 Specimen of Alnus rhombifolia Nutt. Isotype of Alnus rhombifolia var. bernardiana Munz & JOhnSton O I. O O O O O O O C O O O O 241 23 Representative specimen of Alnus rhombifolia Nutt. . 243 24 Specimen of Alnus acuminata H.B.K. var. acuminata. Type Specimen of Betula arguta Schlecht. . . . . 261 25 Specimen of Alnus Aguminata H.B.K. var. acuminata. Isotype of Alnus pringlei Fern. . . . . . . . 263 26 Specimen of Alnus acuminata H.B.K. var. acuminata. Isotype of Alnus guatemalensis Gandoger . . . . 265 27 Representative specimen of Alnus acuminata H.B.K. var. acuminata o o o o o o o o o o o o o 267 28 Specimen of Alnus castaneifolia Mirbel (= Alnus acuminata H.B.K.—var. acuminata) . . . . . . . 269 29 Representative specimen of Alnus acuminata var. glabrata (Fern.) Furlow . . . . . . . . . . 290 30 Holotype of Alnus jorullensis H.B.K. (photograph courtesy of John H. Beaman) . . . . . . . . 303 31 Representative specimen of Alnus jgrullensis H.B.K. var._19rullensis . . . . . . . . . . . . 305 32 Isotype of Alnus firmifolia Fern. (= Alnus jorul- lensis var. firmifolia (Fern.) Furlow; . . . . . 315 33 Representative specimen of Alnus incana subsp. rugosa (DuRoi) Clausen . . . . . . . . . . 329 34 Specimen of Alnus incana subsp. rugosa (DuRoi) Clausen. Holotype of Alnus incana var. glauca f. tomophzlla Fern. - . . . - - - - - - . 331 35 Lower left: isotype of Alnus tenuifolia Nutt. XV late .a Plate Page (= Alnus incana subsp. tenuifolia (Nutt.) Breitung). Upper right: Specimen of Alnus incana subsp. tenuifolia (Nutt.) Breitung O O O O O O O O O O C O O O O 350 36 Specimen of Alnus serrulata (Ait.) Willd. Holotype of Alnus serrulata var. subellip— tica Fern. . . . . . . . . . . . . . . 367 37 Representative Specimen of Alnus serrulata (Ait.)Willd. . . . . . . . . . . . . . 369 38 Representative specimen of Alnus glutinosa (Lo) Gaertn. o o o o o o o ‘0 o o o o o 387 39 Specimen of Alnus viridis subsp. crisps (Ait.) Turrill. Lectotype of Alnus mollis Fern. . . O O O O O O O O O O C 0 O O 404 4O Specimen of Alnus viridis subSp. crispa (Ait.) Turrill. . O C O O O O O O O O O 406 41 Representative specimen of Alnus viridis subsp. sinuata (Regel) L6ve & Lave . . . . . . 423 42 Representative specimen of Alnus maritima (MarSh.) Nutt. o o o O O O 0 O 0 O 0 o o 436 xvi hintout Printout LIST OF PRINTOUTS 1 Sample taxon comparisons using taxonomic data matrix ALNUSM with program CMPARE . . . . . 2 Sample descriptions for Alnus taxa using taxo- nomic data matrix ALNUSM with program DSCRBE . . 3 Computer-generated key to Alnus taxa using taxonomic data matrix ALNUSM with program KEYZ - o O o a o o o o o o o o o . 4 Listing of taxonomic data matrix ALNUSM for the American taxa of Alnus and their Eurasian vicariants . . . . . . . . . . . . . 5 Listing of ALNUSD, the character list for taxo- nomic data matrix ALNUSM . . . . . . . . xvii i I l l I IN TRODUCT ION A survey of the labels of the herbarium specimens used in this study showed that much of the material was misidentified. After working with these specimens and seeing the genus in the field throughout much of the United States, Canada, and Mexico, I have come to the conclusion that the elders are, indeed, distinguishable, but that most of the taxa are quite variable, if not polymorphic, in their vegetative features (which are the principal characters used in their identification), and that accurate determinations often require the concurrent use of several different characters (polythetic identification in the sense of Sokal and Sneath, 1963). Many of the species and infraspecific taxa of M43 appear to be neither morphologically nor genetically well-differentiated. When these taxa are geographically isolated, they usually pose no problems, but when their ranges intersect or overlap, they may become difficult. A study of the literature dealing with the genus in North America re- vealed much confusion in the circumscriptions, ranks, and classifica— tion of taxa at all infrageneric levels. Repeatedly, single—character variants have been treated as distinct species, subspecies, or varie- ties, and there has been no unified concept of these taxa. In order to deal successfully with these problems, I have employed several newer systematic techniques, especially a number of numerical :oonodic and chem allcede to 831“ i jght not be appar at this work to zestal studies. ezion has much t that it is still to become intimat field study, and genetically-accui tterested in th. 'E‘te resulted in gtiate methods f '-‘:erical and ch 33310089. the be it present stuc Tiihlts from a . 2551:“ a (hope E3 generated iiisred among {eyeloled in at The genus Eezsive an inv Eli Gunning if the taXa ir Eleilal . Ill taxonomic and chemosystematic procedures. These, I hoped, would allow me to gain insights into the evolution of the genus which might not be apparent from gross observation alone. But I did not want this work to consist entirely, or even primarily of such experi— mental studies. I believe that the traditional method of classifi- cation has much to offer even in this day of sophisticated techniques: that it is still useful to collect data from a wide variety of sources, to become intimately familiar with the group through herbarium and field study, and to synthesize the most practical as well as phylo- genetically-accurate system. In the experimental work I was more interested in the comparison of taxa than techniques, and this may have resulted in some error if I have inadvertently chosen inappro- priate methods from the numerous available. Most of the techniques of numerical and chemical taxonomy are relatively new, and it is difficult to choose the best for a given problem. But the techniques used in the present studies are rather general and well-accepted, and the results from a variety of experiments were compared and used to syn- thesize a (hopefully) more accurate classification than might have been generated by the use of any method used alone. Where differences appeared among the results, these were examined and new hypotheses were developed in an attempt to explain them. The genus Al_r1£§ is not large, but in order to perform as compre- hensive an investigation as possible, I limited this work to those taxa occurring in the New World. By so doing, I was able to see all of the taxa in the field and to obtain an ample supply of herbarium material. In several instances where species occur in both the New allldllorlds, as no dialled was studied, Although I attei ruins to be done, Illcl herbarium mate significant improven the system proposed tentative, pending : lean and the accum lithe species and individual populati standing of this gr Metic and cytogen Mexistent. It i: mllnhle basis, in Nth future work. and Old Worlds, as much material from the Old World as could be readily obtained was studied. Although I attempted to be as complete as possible, much work remains to be done, especially with the Latin American species (for which herbarium material is not abundant). I feel that I have made significant improvements in the classification of these taxa, but the system proposed here must nevertheless be regarded as somewhat tentative, pending additional exploration of the areas in which they occur and the accumulation of a great deal more material. In all of the species and infraspecific taxa, more detailed study on an individual population basis will have to be made before a true under- standing of this group can be achieved. Especially important will be genetic and cytogenetic investigations, which today are practically nonexistent. It is hoped that the present work will provide a suitable basis, in terms of both classification and nomenclature, for such future work. lltl lllllOl an species 1 his iron man u lltlart. lllS‘l) in all plac Plants, and tens Inuit of .5 HISTORICAL ACCOUNT Although Alni was regarded as a distinct genus by Tournefort (1700) and most others before him, Linnaeus included it as a single species of Betula in the first edition of Species Plantarum (1753). I This treatment was not widely accepted, however, and the separate genera continued to be used by such authors as Miller, Hill, and Ehrhart. Miller, in the 7th edition of The Gardner's Dictionary (1759) wrote: "Dr. Linnaeus has joined this Genus to the Betula, and places it in the fourth Division of his twenty-first Class of Plants, intitled, Monaecia tetrandria, from the Plants having male and female flowers, and the male having four Stamina. But as the Fruit of the Alder is of a conical Form, and that of the Birch cylin- drical, so we shall keep them separate, whereby the Work will be rendered more intelligible to the Generality of Readers.” The compound name Betula—Alnus was used for the alders in 1770 by Weston and in 1785 by Marshall (the genus Betula being retained in both cases for the birches). Above the level of genus, ilfllfi was included with such gymnospermous cone-bearing trees as A4112, Firms, and Larix by Ray (1682) and Tournefort (1700). The develop- ment of suprageneric concepts applying to the alders is reviewed by Stearn (1973). The genus was first divided into segregate genera in the union of Spach (ii @ these segments he present trea’onen‘ i it“ and 9- en ruining species. into one section (3 lost. mothgsus: 1M 1- am llne next revis Ilaviewed the genu corresponding to S; M was fin £flu_lat_§_ was treat {1ng was cousin mp- £033 were lilS Regel comple mlgenera instead Eh“filial many name nltlon of A1315 ' Wifl- The tre u lMore. Alnus revision of Spach (1841) who separated Alnaster and Clethropsis from Alnus, these segments being equivalent to the subgenera recognized in the present treatment. Alnaster included only A. viridis; Clethropsis, _C_. nitida and g. nepalensis, two eastern Asian species; and Alnus the remaining Species. Spach separated the South American species of Alnus into one section (Phyllothyrsus) and the remaining species into another (sect. menothyrsus), failing to note, however, the great similarity between A. acuminata and A. arguta. The next revision of the genus was the monograph of Regel (1861), who viewed the genus as a Single unit, but divided it into sections corresponding to Spach's genera and sections. In this work A_11_1u_s jorullensis was first treated as a variety of A. acuminata. Alnus serrulata was treated as a variety of the European A. glutinosa, and A. Ewes considered a variety of A. incana. Elements of A. incana subsp. rugosa were included in both A. incana and A. glutinosa. In 1865 Regel completed a second revision of the genus in which he used subgenera instead of sections as his major infrageneric divisions and changed many names and circumscriptions as well, including the recog- nition of Alnus iorullensis, A. rubra, and A. serrulata as separate species. The treatment of A. incana subsp. rugosa remained confused as before. Alnus maritima appeared in the second of Regel's works, but in his subgenus Gymnothyrsus rather than subgenus Clethropsis. A third treatment by Regel appeared in de Candolle‘s ________Prodromus (1868)’ this closely following his 1865 system except for a return to the use of sections rather than subgenera. Winkler revised the genus in has Pflanzenreich in 1904. In this when he recognized only nut to Spach' s genus Alnae Sun‘s genus Clethropsis. dude the use of the names Al Anosa, and all of the Lat of g. joxullensie . Although not monographi Mallier appeared about th these publications are a lar Tue most recent revisio (1955) and Murai (1964). C of genera, sections, subsec filemny real changes in t “185m quite artificial, : [hexane section as A. _r_1}_o_m] Min another section. lila, such as the recogniti Mpecies. In a few cases tluding the placement of A Met in the same series. 1°! the genus, but unfortu hetican libraries. In an l“Senate, Alnaster and g mun two subgenera am 1011045 popular current u: treatment he recognized only two sections, sect. Alnobetula (equi- valent to Spach's genus Alnaster) and sect. gymnothyrsus, including Spach's genus Clethropsis. Major changes introduced in this work in— clude the use of the names Alnus alnobetula (for A. viridis), A. ru osa, and A. tenuifolia. Alnus serrulata was treated as a variety of A. rugosa, and all of the Latin American taxa were considered varieties of A. jorullensis. Although not monographic, several important revisionary works by Callier appeared about this time (1892, 1904, 1911, 1918). In these publications are a large number of new names and combinations. The most recent revisions of the genus are those of Czerepanov (1955) and Murai (1964). Czerepanov erected an intricate structure of genera, sections, subsections, series, and Species, but he did not make many real changes in the previous systems. Several of his group- ings are quite artificial, such as the placing of Alnus serrulata in the same section as A. rhombifolia while putting A. incana subsp. rugosa in another section. Some show a lack of understanding of the taxa, such as the recognition of both A. tenuifolia and A. densiflora as species. In a few cases, however, his choices seem to be wise, in— eluding the placement of A. firullensis and A. firmifolia close to- gether in the same series. Murai's treatment (1964) is the most recent for the genus, but unfortunately it is practically unobtainable in American libraries. In an earlier version, Murai (1963) recognized two genera, Alnaster and A1331, but in 1964 he combined these into one with two subgenera and six sections. That treatment fairly closely follows popular current usage, but it differs in several important insects. First, liurai treat species of a single species, nary close phylogenetic rel: oieastem Asian species by ' inlet-us (flothzrsus (pro hithird, he recognizes the rliithe Asian species of his in area in the taxonomy attention is the complex of Erica. Besides the origin e: less superficial treatmen h'sred above, the only study lanald (1904b), Bartlett (1 in} limited herbarium mate respects. First, Murai treats A_1rfl crisps and A. sinuata as sub— species of a single species, A. crispa (the treatment used by Hulte’n, 1944, in wag Alaska fl flog). Second, he attempts to show a very close phylogenetic relationship between the Latin American and eastern Asian species by placing them all in the section Japonicae of subgenus Gymnothyrsus (probably not a correct interpretation). And third, he recognizes the close relationship of Alfi maritima with the Asian species of his section Clethropsis. An area in the taxonomy of Alfl which has received very little attention is the complex of species and infraspecific taxa in Latin America. Besides the original descriptions of these taxa, and more or less superficial treatments in the monographs and revisions listed above, the only study of this group has been the work of Fernald (1904b), Bartlett (1909), and Standley (1920), all based on very limited herbarium material. TAXONC Although the alders are A separate genera, strong overs that they should be considers widely accepted today. The 8 WM; by its woody pi llestructure of its buds, ar line are three rather distie lithe present treatment. '11 Meat venetion, exposure 0 “metre of the buds, seaso httesubgenera, still smal hithese are not as distinc “Ville them into sections. Many of the species of F055 morphology, especiall‘ tImus separate species, “the basis of variation i Characteristics of the hem Without the ranges of 81 featen-es almost always are “the other without disco TAXONOMIC CONCEPTS Although the alders are sometimes segregated into a number of separate genera, strong overall similarities in morphology suggest that they should be considered a single genus, Alnus. This view is widely accepted today. The genus is distinguished from its closest ally, Betula, by its woody pistillate catkins with persistent scales, the structure of its buds, and the number of stamens. Within the genus there are three rather distinct lines, these being considered subgenera in the present treatment. The subgenera differ strikingly, especially in leaf venation, exposure of the pistillate catkins during the winter, structure of the buds, season of blooming, and flower structure. With- in the subgenera, still smaller natural groups of species are evident, but these are not as distinct, and it does not seem useful to sub- divide them into sections. Many of the Species of Alnus are quite variable in terms of their gross morphology, especially in vegetative characters. In the past numerous separate species, varieties, and forms have been described on the basis of variation in the shape, margin, pubescence, and gland characteristics of the leaves. When a large number of specimens from throughout the ranges of such species are examined, however, these features almost always are seen to vary continuously from one extreme to the other without discontinuities. A number of taxa that were heedn'mly on extremes in pt AIA, A. ferruginea, A. prir Ami. emersoniana, A. _1 (hargpta var. subsericea, A MA, A. serrulata var. sube clingy) are no longer recog Inseveral cases, wide-r into separate species largely in: the similarity in their pear to hybridize where the ten are actually conspecifii M. M in eastern Nor 5st parts of a single speci Audra, A. sinuata in weste has relationships have bee tests the tendency has been “spite of the similaritie: hrsntiation is evident in lEthical localization, the “d E- _V1_n_d_1_s- respectively 5% in Mexico and nor chindered separate specie: “it they are treated as 0‘ 1eSituation. Subspecies, as ‘emploi of“hides having relative based mainly on extremes in pubescence characters (Alnus crispa var. mollis, A. ferruginea, A. pringlei, A. rhombifolia var. bernardina, A. rugosa f. emersoniana, A. serrulata f. noveboracensis), leaf shape (A. arguta var. subsericea, A. crispa var. harricanensis, A. w- foli, A. serrulata var. subelliptica) and cone shape (A. crispa var. elongata) are no longer recognized. In several cases, wide-ranging taxa have previously been split into separate species largely on the basis of geographical separation. From the similarity in their morphology and the ease with which they appear to hybridize where their ranges overlap, it seems that these taxa are actually conspecific. Alnus tenuifolia in western North Ameri- ca, A- rugosa in eastern North America, and A. incana in Eurasia repre- sent parts of a single species, as do Alnus crispa in eastern North America, A. sinuata in western North America, and A. viridis in Eurasia. These relationships have been recognized for a long time, but in recent years the tendency has been to treat the units as separate species. In Spite of the similarities among the above taxa, however, some dif- ferentiation is evident in each case. Because of this and their geo- graphical localization, they are considered subspecies of Alnus incana and 5' m respectively. Alnus acuminata in South America and A' ELM in Mexico and northern Central America are also usually CODSidered separate species. Their differences are so slight, however, that they are treated as one species (A. acuminata) without subsPeCific de$ignation. SUbSpecieS, as "employed in the present treatment, are segments of species having relatively large geographical ranges and which are distinct in morphology, distr' hhttotlcf. DuRietz, 1930; 1 category PM is also used modes (which may occur wi moevertheless ecologicall Sometimes closely-relat storal hybrids where their shop. m and A. serrula mtphologically distinct and ecologically isolated as wel isolated, as shown by succe tint and geographically is “flags are therefore prob biologically distinct in mo Mines incompletely isolo 10 distinct in morphology, distribution, and, to at least some degree, i habitat (cf. Du Rietz, 1930; Camp and Gilly, 1943; Mayr, 1947). The category variety is also used here, but for more local facies of species (which may occur without major geographical gaps, but which are nevertheless ecologically and morphologically distinct). Sometimes closely—related species in M produce putative natural hybrids where their ranges overlap (for example, A. incana subsp. rugosa and A. serrulata). In such cases, the taxa are usually morphologically distinct and, for the most part, geographically and ecologically isolated as well. Many species of Alnus are extrinsically isolated, as shown by successful artificial crosses between very dis- tinct and geographically isolated taxa (cf. Ljunger, 1959). The species in £1223 are therefore probably best thought of as entities which are biologically distinct in morphology, geography, and ecology, but sometimes incompletely isolated genetically. ‘ EVOLUTIO Ania is distributed th inhelav the Equator along iron the Antilles in both mo tally related genera (F3 5, existed there in the Tertiar teeters can be recognized: ta. 2) in southern and sent northern South America, and ta from northern Mexico to [units of Llnlii have been ; a . ‘ linens does not occur to ‘ his identity by some pale: this is represented 5M8, appearing in the u Vith other early woody gel “1%, and M133 (cf. l1Eyellen“ in North Ar whountains, and the Wish Columbia (Wolfe. 3“ W1 is represented ‘Jlnoos tossil florss EVOLUTION AND PHYTOGEOGRAPHY Alnus is distributed throughout the Northern Hemisphere, extend- ing below the Equator along the Andes in South America. It is absent from the Antilles in both modern and fossil floras, though floristi- cally related genera (Fagus, Salix, Myrica, Liguidambar, and Nyssa) existed there in the Tertiary (Graham, 1972b). Four distributional centers can be recognized: I) in circumpolar Eurasia and North Ameri- ca, 2) in southern and southeastern Asia, 3) from central Mexico to northern South America, and 4) along the western coast of North Ameri- ca from northern Mexico to southern Alaska. In addition, alleged fossils of Alni have been reported from Australia and Tasmania, where the genus does not occur today, but these are regarded as of question— able identity by some paleobotanists (cf. Berry, 1923). M is represented in the early fossil record of the angio- sperms, appearing in the upper Cretaceous, where it was associated with other early Woody genera, including Betula, Acer, Corylus, Cercis, M, and Seguoia (cf. Takhtajan, 1969). The earliest "definite M pollen" in North America appears in the Maestrichtian of the Rocky Mountains, and the first megafossils are of the same age in British Columbia (Wolfe, 1973). During the Tertiary it became promi- nent and is represented by Wood, leaves, fruits, cones, and pollen in numerous fossil floras. Over 40 Cenozoic fossil species are recognized fl tylaibtte (1952) in North Ame homely formd in bog depoe star of climatic and vegetati Islamic fossil alders bear a notice, and it may be poseib' oitheoe taxa. eater th According to Takhtaj an lantern Asia. lie shows that thrust primitive segment 0 restricted to the Himalayas, thelieves was important in Wuodhelou, however, subge thrust primitive aspect of tiliooo that it is the most it“ this group has not aim all. although the genus may “8101!, recent information ' “its a site of early eVO glilxelrod, 1972). hirai (1961+) concludes may Specifically an area milmother of species an illtrihutions occur there ,fiftroatus of subgenus A}: 12 by LaMotte (1952) in North America alone. Fossil pollen of A_lnls_ is commonly found in bog deposits of Pleistocene age and is an indi- cator of climatic and vegetational changes during that period. Many Cenozoic fossil alders bear a striking resemblance to present-day species, and it may be possible to trace the migrations and evolution of these taxa . Origin 2f the Genus and Subgenera According to Takhtajan (1969), m may have evolved in south- western Asia. He shows that subgenus Clethropsis (which he considers the most primitive segment of the genus) is presently more or less restricted to the Himalayas, Assam, and southwestern China, a region he believes was important in early angiosperm evolution. As dis- cussed below, however, subgenus Clethropsis may not actually represent the most primitive aspect of Alnus; in fact, there are some indi- cations that it is the most specialized. The fossil record shows that this group has not always been so largely restricted to Asia, and, although the genus may, indeed, have originated in the Himalayan region, recent information has greatly reduced the importance of this area as a site of early evolution in the angiosperms as a whole (Raven and Axelrod, 1972). Murai (1964) concludes that Alnus must have originated in eastern Asia, specifically an area including or near Japan, because the largest total number of species and the largest number of supposedly relict distributions occur there today. He regards members of his section Bifurcatus of subgenus Alnobetula (today completely restricted to " W I n— fl Joya) as the most primitive ‘ It is impossible to pred yussAlflon the basis of p inner one can make a few ge oithegroup. Axelrod and Ra theory and evolutionary bioge its primary evolution and re hope and America were sep conclude that "migration acr minterrupted during much of smeof the disjunctions be his result from extinction [in across Beringia." Thi Eyhsve lasted until well i ‘ hussia as a whole, the As: if present species of Alpgg ifintipal that the greatest ofits origin, it might be Wisinste in that part of t “that that subgenus SL212} ”hears Alnobetula in east intheastern America. But aidhetause extinction has ~ “element of present ran Mclusions. 0n the basis of wood 13 Japan) as the most primitive taxa of the genus. It is impossible tolpredict the exact site of origin of the genus A_l£1£s_ on the basis of present information about its history. However one can make a few generalizations concerning the development of the group. Axelrod and Raven (1972), discussing plate tectonic theory and evolutionary biogeography, state that the Amentiferae had its primary evolution and radiation on the Laurasian land mass before Europe and America were separated by the North Atlantic Ocean. They conclude that "migration across Laurasia was direct and essentially ' uninterrupted during much of the Cretaceous..." and that "no doubt some of the disjunctions between eastern North America and eastern Asia result from extinction in western Eurasia rather than by migra- tion across Beringia." This connection between EurOpe and America ’may have lasted until well into the Eocene (cf. Raven, 1972). Viewing Laurasia as a whole, the Asian sector shows a much greater diversity of present Species of Alnus than other regions. From Vavilov's (1951) principal that the greatest diversity in a group occurs in the area of its origin, it might be suspected that the genus did, in fact, originate in that part of the world. Present distributional patterns Suggest that subgenus Clethropsis evolved in the Himalayan region, subgenus Alnobetula in eastern Asia, and subgenus £292 in Europe and northeastern America. But because Alnus is so ancient and widespread, and because extinction has apparently played an important role in the development of present ranges, these must be considered tentative conclusions . On the basis of wood anatomy, it appears that the Betulaceae nosefron an ancient Hamame ooestors (Tippo, 1938). Ho patterns of leaf venation in noting: "in their consiste the various members of the B hogtype and could well b training, he shows that oattern) is entirely absent he,and is rare in the R09 men in the Fagaceae and hthe history of the famil formed one line while Cor l indforned another. Betula features, have been shown t hippo, 1938; Hall, 1952). little, as seen from their to early hypothesis 01 evolution of thug is outl his theory, the early Ter three principal "floristic (heeoland Province) was to mate deciduous trees Sht aMowers rare in this 1 "tion (lhrgai Province) ‘ aIdlloeka, and it was co % Occupied an imports 14 arose from an ancient Hamamelidaceous stock derived from Magnoliaceous ancestors (Tippo, 1938). Wolfe (1973), in discussing the origin of patterns of leaf venation in the Amentiferae, supports this View, stating: "in their consistent crasPedrodromy and intercostal venation, the various members of the Betulaceae are highly similar to the 923'.“ _1_op_§i_§ type and could well be the end products of this hamamelid line." Continuing, he shows that the leaf venation of the Betulaceae (Corylus pattern) is entirely absent from the Magnoliidae, Dilleniidae, and Asteri- dae, and is rare in the Rosidae. In the Hamamelidae, however, it is common in the Fagaceae and Ulmaceae as well as the Betulaceae. Early in the history of the family the genera Betula and Alain—s apparently formed one line while Corylus, Ostrya, Ostryopsis, and Carpinus diverged and formed another. Betula and _Al_nu_s_, on the basis of anatomical features, have been shown to be the most primitive members of the family (Tippo, 1938; Hall, 1952). These two genera have diverged relatively little, as seen from their similar morphology and anatomy. An early hypothesis of the events which may have followed the evolution of w is outlined by Kryshtofovich (1921). According to this theory, the early Tertiary flora of Europe became divided into three principal "floristic provinces." The northwestern division (Greenland Province) was made up of Species of Populus and other tem- perate deciduous trees showing no Asiatic relationship. A1233 and Mala were rare in this region during the early Tertiary. The second region (Turgai Province) occupied the middle zone of Siberia, America, and Alaska, and it was covered by deciduous forest in which Betula and Alnus occupied an important place. To the south was the third region (loltm Province) which incl non was characterized b olregetation. As the clima toyed, spreading into the Gr t'elossil record was interp leg originated in the son moss Siberia and western N replete the circumpolar dis lottery. The present distr stiller with the known fossil regent such an early intro are recent migration from The species comprising liens by llurai (1964), one lantern Asian taxa as well . other (sect. Glutinosae) in allies. Murai concludes th Vicinity of Japan, section ll! sis, and section 3139 M (subgenus 91-22131 intense, however, suggest “phyletic and distinct, IEttluns in this manner in ‘ileats, rather, to be der tea cooler climate. Clo: his and northern Latin N 15 (Poltava Province) which included the Ukraine and southern Russia, and which was characterized by remnants of the older evergreen tropi- cal vegetation. As the climate cooled, the Turgaian vegetation en— larged, spreading into the Greenlandian and Poltavan regions. Thus, the fossil record was interpreted by Kryshtofovich as showing that A1313 originated in the southern Asian highlands, spreading east across Siberia and western North America, but not moving west to complete the circumpolar distribution seen today until later in the Tertiary. The present distribution of Alnus in North America, to- gether with the known fossil record there, does, in fact, seem to suggest such an early introduction from the west followed by a much more recent migration from the east. The species comprising subgenus Alnus are divided into two sec- tions by Murai (1964), one of these (sect. Japonicae) including several eastern Asian taxa as well as all of the Latin American taxa, and the other (sect. Glutinosae) including Alnus incana and its various allies. Murai concludes that both of the sections originated in the vicinity of Japan, section Japonicae directly from his section fle— thropsis, and section Glutinosae by a second route from his section Bifurcatus (subgenus Alnobetula). All of the morphological and other evidence, however, suggests that subgenera Alnus and Clethropsis are monophyletic and distinct, which would make the derivation of Murai's sections in this manner impossible. Murai's section Glutinosae appears, rather, to be derived from his section Japonicae and adapted to a cooler climate. Close relationships between taxa of eastern Asia and northern Latin America were first noted by Hara (1948) and F. - ' hindiscussed briefly by Na then said to resemble that tleearly Oligocene of Colors demotion does not substan haibetveen the alders of rill be needed to firmly est he most widespread vie inllorld is that it repres hetaceous and Arcto-Tertia Equestioned the validity flora of this type in the e iota, noting that "no known More nixed mesophytic ge Interested in the tropical lligoeene that it became a Ia“periods in the early 1 3the and other "temperate Wand regions. Cooling tillfigetstion, but a true hill the Miocene. In die hold and MncGinitie (1‘. “Sim of North America he stamina, but that by [’“lllcal American aspect. 16 later discussed briefly by Matuda (1953) and Sharp (1953a, 1972). Pollen said to resemble that of Alia; japonica has been reported from the early Oligocene of Colorado by Van Alstine (1969). Nevertheless, observation does not substantiate the close relationship suggested by Murai between the alders of Mexico and eastern Asia. Additional study will be needed to firmly establish the true relationship. Dispersal B the New World The most widespread View of the origin of the genus firm; in the New World is that it represents a remnant of the widespread Arcto- Cretaceous and Arcto-Tertiary Geofloras (Axelrod, 1952). Wolfe (1972) has questioned the validity of the concept of a uniform mesophytic flora of this type in the early Tertiary on the basis of paleobotanical data, noting that "no known Cretaceous or Paleocene flora contains 20 or more mixed mesophytic genera." He concludes that flog was first represented in the tropical vegetation and that it was not until the Oligocene that it became a component of the temperate forest. During warm periods in the early to middle Oligocene, the alders, along with maples and other ”temperate" woody genera are seen as again occupying lowland regions. Cooling that followed this period removed the tropi- cal vegetation, but a true mixed mesophytic vegetation did not appear until the Miocene. In discussing the history of the Rocky Mountains, Leopold and MacGinitie (1972) show that early Eocene floras in this region of North America have affinities with modern plants of south- eastern Asia, but that by the mid-Eocene the flora had taken on a tropical American aspect. Chaney (1947) shows that during Eocene "W Liedeneotropical flora ex mien edge of the continent atinof Asiatic affinities uhs, lasting until mid—Mice forest had already developed nzie,1972). Modern interpretation linearized by Graham (197 thesouthem United States leperate to tropical eleme {minding £13191)! while a iileotropical components ex ialaeogene, the Rocky Mount fluent, which was graduall‘ forest in the mid-Tertiary. hanging climates, eliminat 01 the western United State lhe ancestors of both caMpecies were probably ‘ cussed above. Although f0 “Wish exists to show that “511! time. Based on P011 thus that 51‘1“}. (along W5 a Lie. £1.62 a; Mid-Miocene but is “0 flan from the Paraje $01 17 time the neotropical flora extended north to about 49 degrees on the western edge of the continent. In far northwestern America the termi- nation of Asiatic affinities was more gradual than in the Rocky Moun- tains, lasting until mid-Miocene, by which time the modern coniferous forest had already developed in the Rocky Mountains (Leopold and Mac- Ginitie, 1972). A modern interpretation of the Tertiary floras of North America is summarized by Graham (1972a). The lower Tertiary vegetation of the southern United States probably consisted of neotropical warm- temperate to tropical elements with some broad-leaved deciduous species (including A_1nls_), while a broad-leaved deciduous community with paleotropical components existed in the Pacific Northwest. After the Palaeogene, the Rocky Mountain and western floras lost their Asiatic element, which was gradually replaced by an expanding mixed mesophytic forest in the mid-Tertiary. Later, mountain building resulted in changing climates, eliminating the mixed mesophytic element in much of the western United States, Canada, and western Europe. The ancestors of both _A_lr_1_1§ maritima and the present Latin Ameri- can species were probably part of the early Tertiary vegetation dis- cussed above. Although fossil evidence from Latin America is scanty, enough exists to show that E reached this region at a relatively early time. Based on pollen from Veracruz, Graham (1972b, 1973) Shows that _AlLus (along with _A_b_i_e_s_, Juglans, w, Liguidambar, Ulmus, Celtis, Picea, Myrica, and Populus) reached southern Mexico by the mid-Miocene but is not present in older sediments there. This flora from the Paraje Solo formation, furthermore, closely parallels, hgeheric makeup, the modern he has, and meal hing the late Miocene, bein sells of work by Van der hemthem South America to age, although rare, isolated meaedimeuts (Van der Ham lllll) published an often-oi [Ill Bolivia, but this dete pollen appears in South Ame MM is not found un lstlett and Barghoorn (197. lltlstlmus of Panama, statl lilprodueed islands betwee wdid not migrate south lithe isthmus (Talamanca ll seals: and a. .i hence until after the M11 mentigin of A. serrulata earlier than the introduc E”remnant of an early I1Siesta a later origin, QlLale most likely a re "l; as are A. rhombifoli lleee last three species tan cumulex. The origin l l l————’ W 18 in generic makeup, the modern flora of that area. Three of these genera, (Alnus, Juglans, and Myrica) are found in pollen deposits in Panama during the late Miocene, being absent from earlier floras there. Re- sults of work by Van der Hammen show that the earliest Alw- pollen in northern South America to occur in abundance is mid-Pleistocene in age, although rare, isolated grains are also found in lower Pleisto- cene sediments (Van der Hammen, 1972; Graham, 1972b, 1973). Berry (1922) published an often—cited Pliocene species (mi preacuminata) from Bolivia, but this determination is not generally accepted. Myrica pollen appears in South America before Aln—us, in the lower Pliocene, but Juglans is not found until the lower Quaternary (Graham, 1972b). Bartlett and Barghoorn (1973), in reviewing the geological history of the Isthmus of Panama, state that by the mid to late Jurassic, orogenies had produced islands between North and South America, but presumably m did not migrate south, at least in numbers, until the last fold of the isthmus (Talamanca Ridge) became fully emerged in the Miocene. £12133 viridis and A. incana may not have been introduced to North America until after the Miocene, when the northern climate became cooler. The origin of A. serrulata is more obscure, but it was undoubtedly earlier than the introduction of A. incana. Like A. maritima, it could be a remnant of an early Tertiary flora, but its closeness to _A_. incana Suggests a later origin, perhaps in the Miocene or Pliocene. £9113. £11333 is most likely a remnant of the western mesophytic Miocene for— est, as are A. rhombifolia and A. oblongifolia or their ancestors. These last three species share their affinities with the Latin Ameri- can complex. The original stock which formed this group moved south htollexico, Central America, serially in response to cli retains, to form the taxa 3 Historical Fac 0f the alders occurring Ma. appears to be the he its apparent progenitor pelminautly coastal distr' elite and Ozark highlands a (1955) points out that ther upland areas (the Appalachia lllantlc Coastal Plain and t Tertiary, followed by disse united in the loss of sui the glacial periods, the r lllhe north, especially 1! llaclation did not strongl‘ Southeastern United States illthem Ohio and Kentuck mvived the Pleistocene ' “£5. LIME. subSp. Eliot Europe more recent tall divergence of these ltlantic floras are very toniitions around the No serrulata appears to be the most ancient, having specialized greatly 19 into Mexico, Central America, and South America, where it diverged, especially in response to climates at different altitudes in the mountains, to form the taxa seen in those regions today. Historical Factors Affecting Distributions Of the alders occurring in northern North America, Alnus from its apparent progenitor, A. incana. The species today has a predominantly coastal distribution, although it occurs in the Appala- chian and Ozark highlands as well, where it may have evolved. Braun (1955) points out that there are many disjunctions of vegetation from upland areas (the Appalachian and Interior Highlands) on the younger Atlantic Coastal Plain and explains that these occurred during the Tertiary, followed by dissection of the intermediate peneplane, which resulted in the loss of suitable habitats there. During the Pleisto- cene glacial periods, the range of A. serrulata may have been decreased in the north, especially in the mountains. Braun (1951) feels that glaciation did not strongly disrupt the deciduous forest zone in the southeastern United States as far north as the Appalachian Plateau in southern Ohio and Kentucky. If this is so, A. serrulata might have survived the Pleistocene in much of its present range in North America. Ping viridis subsp. crisps. must have been separated from A. v_j_£i__ gig of Europe more recently, as shown by the relatively very small divergence of these taxa. Hulte’n (1963) argues that the North Atlantic floras are very ancient, and that "the phytogeographical conditions around the North Atlantic...give poor support for a land [ridge that could have existe iallowa, however, that "cit ligated by wind over the fro mld not he difficult to im hreletively recent times. advances in several refugia, date it occurs as a relict live (1967) provide evidence svia, Iceland, and the Can blend, and Ives (1963) con Ilse subsp. m occurs extent of Hisconsin glaciat thesevere conditions of th M by Thorarinsson (1963 tenevolcanic sediments in liarshberger, as early Wations before advancing hpalachlans as a glacial 1 ll! Scandinavian element, lithe late Pleistocene, r Mining subsp. r_u39_5l_§ 59% of southwestern ll 80, the populations of 5 an have displaced the P‘ llulster time since the Mot subsp. incana a 20 bridge that could have existed in Quaternary or late Tertiary times." He allows, however, that "circumpolar plants of the far North" probably migrated by wind over the frozen polar sea or on floating ice. It would not be difficult to imagine 5.11% being dispersed in this manner in relatively recent times. This taxon probably survived glacial advances in several refugia, including the southern Appalachians, where it occurs as a relict today. A. Love (1959, 1967) and Love and Love (1967) provide evidence that refugia existed in parts of Scandi- navia, Iceland, and the Canadian arctic archipelago west of Baffin Island, and Ives (1963) concludes that large parts of northern Labrador, Where subSp. crisEa occurs today, remained ice-free during the maximum extent of Wisconsin glaciation. That the species could have withstood the severe conditions of the Pleistocene in such northern refugia is shown by Thorarinsson (1963), who cites fossil alder leaves in Pleisto- cene volcanic sediments in Iceland. Harshberger, as early as 1903, rejected the idea of massive migrations before advancing Pleistocene glaciers and accepted the Appalachians as a glacial refugium in the eastern United States. The Scandinavian element, according to him, invaded the New World in the late Pleistocene, remaining after the final glacial retreat. £11123 incana subsp. rugosa, which is scarcely differentiated from A. incana of southwestern Europe, could have arrived in this way, but if so, the papulations of Alnus incana present in Scandinavia today must have displaced the Pleistocene forms by migrating from the east at a later time since these are quite different from either subSp. rugosa or subsp. incana of southwestern Europe, as shown below. It hpsaible that subsp. pried, that subsp. M th ndthat the present norther Scandinavia from the east, a Malpmhahly survived in mat of the Appalachian h hell). lave, 1959). In the Northwest, snap. 3% apparently e tithe mixed mesophytic for first fossils are in the hi Mes corresponding to th @(5. carpinoides), res hited States and Canada du m they have today (Chane lhal’leistocene _A_. incana s was $119321 5 ‘lide gap in the continuitj l3311p, 1947). Subspecies _: advances in refugia in ung 1955) and points southward its present habitats in t1" Wprobably at lower e lMilitias is supported b] Hidtnce (Heusser, 1960) - Iunived in its present 11 21 is possible that subsp. rugosa entered the New World at an earlier period, that subsp. incana then became extinct in northern Europe, and that the present northern European subsp. incana stock invaded Scandinavia from the east, all before the Pleistocene. Subspecies mprobably survived in refugia at low elevations, either east or west of the Appalachian highlands and immediately south of the ice (D. L6ve, 1959). In the Northwest, fl incana subsp. tenuifolia and A. viridis subsp. sinuata apparently entered from the north after the extinction of the mixed mesophytic forest in the middle Miocene, because their first fossils are in the Miocene of Alaska (Wolfe, 1969). Fossil species corresponding to these present taxa, A. harneyana and A. fos- silis (A. carpinoides), respectively, were present in the western United States and Canada during the Miocene in habitats similar to the ones they have today (Chaney, 1959; Chaney and Axelrod, 1959). During the Pleistocene A. incana subsp. rugosa was separated from subsp. tenuifolia and A. viridis subsp. crispa from subsp. sinuata by a "wide gap in the continuity of the northern coniferous forest” (Raup, 1947). Subspecies sinuata may have survived the glacial advances in refugia in unglaciated Alaska (Hultén, 1937b; Livingstone, 1955) and points southward to Washington (Heusser, 1960) as well as in its present habitats in the mountains of the northwestern United States, though probably at lower elevations. Its Pleistocene presence in these localities is supported by geological, paleobotanical, and botanical evidence (Heusser, 1960). ABE. incana subsp. tenuifolia probably survived in its present habitats in the Sierra Nevada and Rocky botains as far south as his (£12,5- M suhSp. ru es sinetsuhsp. tenuifolia in hearse time, the range of no that of subsp. sinuata W M subsp. ru sunny lltis (1966) in wh across the glacial boundary uglaciated land. Iltis i nltinate western origin of hast, it seems more likely glaciated soil in the east unsodhy climate. In the iitlthof the glacial bound: iiinthe for North. The < gin not exist south of utter of fact, a consider illhe southern limits of eastern North America. It is probable that .1 fidenpread in North Ameri aMound only in Asia, f W) genera of eastern l lihrnenhers of subgenu: oflheveetern United St distinctive leaf feature 22 Mountains as far south as New Mexico and Arizona. After the Pleisto- cene, A. incana subsp. rugosa again extended across northern Canada and met subsp. tenuifolia in northern Alberta or Saskatchewan. At the same time, the range of A. viridis subsp. crispa became continuous with that of subsp. sinuata in Alaska. £1313. incana subsp. rugosa and subsp. tenuifolia fit the pattern shown by Iltis (1966) in which western species range north and south across the glacial boundary while their eastern vicariants are restricted to glaciated land. Iltis interprets this pattern as evidence for the ultimate western origin of these eastern taxa, but in this case at least, it seems more likely that the restriction of subsp. rugosa to glaciated soil in the eastern United States and Canada is an artifact caused by climate. In the mountainous West, subsp. tenuifolia occurs south of the glacial boundary only at high elevations in the mountains or in the far North. The cool climatic conditions required by A. i_n_- 9% do not exist south of the glacial boundary in the East. As a matter of fact, a considerable amount of glaciated land exists south of the southern limits of the range of A. incana subsp. rugosa in eastern North America. It is probable that Alnus maritima was formerly much more widespread in North America than it is today. Its closest relatives are found only in Asia, following a pattern shared by many other woody genera of eastern North America and that region (Li, 1971, 1972). Other members of subgenus Clethropsis are known in the fossil record 0f the western United States and Canada. These taxa have the same distinctive leaf features (margin and venation) of modern npuentatives of this grou wed in Idaho and Oregon liliiar to those of _A_. marit idly (Chaney, 1959). Anoth «his, described from the E differs in certain respects , lectures of this group (cf. The presence of w hstberegarded as of rela Species is not hardy furthe survived the Pleistocene t resent location on the co following the Pleistocene hearth. The cause of t blame and Oklahoma is n1 Shun in the numerical tax iohave diverged somewhat, if the disjunction (cf. F1 “ii the indications of 81 nnrigin of the dis junct 1aliens following range 1‘ The Latin American ”l1. oblon ifolia, hav fhavery long time, as “in! restricted to the "lint of the moist mix 23 representatives of this group. One such species, Alnus relatus, occurred in Idaho and Oregon during the Miocene in habitats similar to those of _A_. maritima in the eastern United States today (Chaney, 1959). Another fossil species, A. cremastogyn- Edd—9.5., described from the Eocene or Oligocene of British Columbia, differs in certain respects, but also shows the characteristic leaf features of this group (cf. Berry, 1926; Brown, 1937). The presence of Alnus maritima on the Delaware Peninsula might best be regarded as of relatively recent achievement since this species is not hardy further north and probably could not have survived the Pleistocene there. It most likely migrated to its present location on the coastal plain during the climatic moderation following the Pleistocene glaciation from some location further to the south. The cause of the wide disjunction of this species in Delaware and Oklahoma is not known, but the two populations are shown in the numerical taxonomic and chemosystematic studies (below) to have diverged somewhat, pointing away from a recent achievement of the disjunction (cf. Fryxell, 1967). This evidence, together with the indications of great antiquity discussed above, indicate an origin of the disjunct distribution by isolation of the popu- lations following range restriction. The Latin American taxa, along with Alnus rubra, A. rhombifolia, and A. oblongifolia, have apparently been in their present locations for a very long time, as .suggested by McVaugh (1952). Alnus rubra, today restricted to the Pacific coastal fog belt, appears to be a remnant of the moist mixed forest of the Northwest in the Miocene. llflrhonblfolia may have al adapted to sumer-dry condit‘ lime fossils of _Aln_ua hol nrphologically similar to Intern lowland and slope to n'tl such mesophytic genera all. oblongifolia as vice Woodlands and Chaparral mum Oak Hoodlands and C nllfled from the Oligocene lest two regions were appa lullaby the Pliocene. The antiquity of the films been a subject of di twin of this problem was “It "the present temperaU “411th during the late Plei 'lherewas little continue Elumerate vegetation . . . " after reviewing the paleot lllles that the temperate ofNorth America as early t“Pliocene, the eastern ml Mexico until the ear mmvenent of temperat 24 M rhombifolia may have also been a part of this forest, having adapted to summer-dry conditions since the Pleistocene (Wolfe, 1969). Miocene fossils of Ewandiana (A. corrollina), a species morphologically similar to modern A. rhombifolia, are described from western lowland and slope forests by Chaney (1959), in association with such mesophytic genera as £e_r, Fraxinus, £52523, Ill-mi, Liguid- ambi, Ostrya, Juglans, and Ema; Benson (1962) lists A. rhombifolia and A. oblongifolia as vicariants in the California and Southwestern Oak Woodlands and Chaparral, respectively. He feels that the South- western Oak Woodlands and Chaparral have been ”relatively little modified from the Oligocene prototype, which also had summer rain." These two regions were apparently separated by intervening arid lands by the Pliocene. The antiquity of the temperate flora of Mexico and Central Ameri— ca has been a subject of dispute among various workers. An excellent review of this problem was made by Rzedowski in 1965. One View holds that "the present temperate flora of Mexico must have come from the north during the late Pleistocene" because prior to the Pliocene "there was little continuous area with sufficient elevation to support a temperate vegetation..." (Sharp, 1953b). Dressler (1954), however, after reviewing the paleobotanical and physiographic evidence, con- cludes that the temperate types were introduced from the western part of North America as early as the Miocene and reached Guatemala by the Pliocene, the eastern North American plants probably not migrating into Mexico until the early Pleistocene. Chaney (1936), in discussing the movement of temperate forests southward during the cooling period of the Cenozoic states that ' IlBtllEd as far south as sou eluding such genera as 51% ling, and _S_e_tli_x_. Axelrod Ella, and M in Mexico am and thus having "no fl title same genera." McVau adjacent Central America si litmus time. . . " From the morphological Mexican alders, it seems him in such remote time WEE all show strong Manta of stocks which e' lleflrst place, or which Tertiary Flora during or 1: little is known of tl AlIfltlltmselves due to tl temy closely related, indicating relatively rec 25 of the Cenozoic states that "a modified type of Miocene forest had migrated as far south as southern Guatemala..." by this time, in- cluding such genera as Alnus, Arbutus, Crataegus, Ostrya, Pinus, Quercus, and £935. Axelrod (1939) lists the genera Quercus, Cornus, M, and Juglans in Mexico as possibly having differentiated in that area and thus having "no floristic relationship to the northern species of the same genera." McVaugh (1952) suggests "that such genera as Alnus, Arbutus, Carpinus, and others have persisted in Mexico and adjacent Central America since early Tertiary or even since Cre- taceous time..." From the morphological and biochemical uniqueness of the Latin American alders, it seems likely that their ancestors did reach southern Mexico in such remote times. Alnus rubra, A. rhombifolia, and A. 51): longifolia all show strong affinities to these species and may be remnants of stocks which either did not migrate further south in the first place, or which originated from a northward-moving Madro- Tertiary Flora during or before the Pliocene. Little is known of the history of the Latin American species of A333 themselves due to the scanty fossil record. All of the taxa are very closely related, however, and are not strongly differentiated, indicating relatively recent speciation. Host of the alders are to occur in standing water, loss and muekegs, or on moi: mhat drier conditions. is sometimes found in appar ocuuxsonmountain slopes a Both of these species, have undated with some source asuhvious as in the other “alight, an exception aga' flyoccur as an understory i! the case, the forest is 5:] actually represent a d lletaxa are isolated by l hlions overlap, although mutually occur sympat‘ All species of Alfllj c(twining nitrogen-fixin hump, 1954). The “Rain since they were : “950) reviews the ear 1y I'l‘mr ECOLOGY Most of the alders are associated with wet habitats. The plants may occur in standing water, on stream banks, on wet floodplains, in bogs and muskegs, or on moist mountain slopes. A few have adapted to somewhat drier conditions. Alnus viridis subsp. crisEa, for example, is sometimes found in apparently dry woods, and A. jorullensis often occurs on mountain slopes away from standing or running water. In both of these species, however, the plants are still usually closely associated with some source of moisture, even though it may not be as obvious as in the other species. Most alders grow only in full sunlight, an exception again being A. viridis subsp. cris a, which may occur as an understory component in conifer woods. Where this is the case, the forest is usually very open, and the alder stratum may actually represent a declining successional stage. Nearly all of the taxa are isolated by habitat where their geographical distri- butions overlap, although some have fairly wide tolerances and may actually occur sympatrically. All species of ATE-g are mycotrophic, with root nodules (Plate 1) containing nitrogen-fixing micro-organisms (cf. Bond, 1956, 1963; Bond 31.2.3.1." 1954). The identity of these endophytes has been un- certain since they were first discovered in the early 1800's. Kelley (1950) reviews the early controversy surrounding this question. Recent 26 ytes on roots of Alnus . ‘ ‘ doph nitrogen—f1X1ng en X l. B, X 2'2' 28 Plate 1. identifications 0f the organ liasnodiophorales (l’hycomyce MW by Neal _e_t duels evidence (Bond, 19‘ that the nitrogen fixing en< species upon which they are Synhionts forming myco lhsui, 1926; KleXka and Vu dill. These organisms app nd the minimization of roc nil understood. The role of m in 1 :tier of investigators, e: Southern Alaska. In studi 1,1939) showed that A_l 9 recently deglaciated la ”f mosses, horsetail hronghathicket stage 01 must climax of spruCe hjnrl . 1955), Crocker alid ill ' “hilitchell (1968) ’ ll )1 and others Rein . S in the Successiona WW: Salix in him \Ufls \% sub itihtnsis \ 50 to 070 yea] , ‘.'-.._.,,..,,,_a 29 identifications of the organisms causing the nodules include Plasmodiophorales (Phycomycetes) by Hawker and Fraymouth (1951) and Strejtonlzces by Neal 31131. (1968). Whatever their identity, there is evidence (Bond, 1963; Rodriquez-Barrueco and Bond, 1968) that the nitrogen fixing endophytes of M are specific for the Species upon which they are growing. Symbionts forming mycorrhizae on £13133 have also been reported (Masui, 1926; KleXka and Vukalov, 1935; Trappe, 1964; Neal EEELH 1968). These organisms apparently have a role in nutrient absorption and the minimization of root diseases, but their biology is not yet well understood. The role of _A_ltlli in plant succession has been studied by a number of investigators, eSpecially in areas recently deglaciated in southern Alaska. In studies at Glacier Bay, Cooper (1923a, 1923b, 1931, 1939) showed that .A_l_n_ug plays an important part in the succession of recently deglaciated land. He demonstrated that from a pioneer stage of mosses, horsetails, Epilobium, and m, the community moves through a thicket stage of willows and alders before reaching the forest climax of spruce. This work has been continued by Crocker and Major (1955), Crocker and Dickson (1957), Lawrence (1958), Heilman (1966), Mitchell (1968), Ugolina (1968), Hurd (1971), Reiners _e_t;_a_l_. (1971), and others. Reiners $511. (1971) estimate the length of stages in the successional sere at Glacier Bay as follows: horsetails, sedges, Epilobium, Salix, and P02111113, O to 5 years; Dryas, 5 to 20 Years; Alnus viridis subsp. sinuata, 20 to 40 years; Papulus and Picea Sitchensis, 50 to 70 years; Picea sitchensis forest, 75 to 100 years; ndlsugaforest, over 100 y n addition of nitrogen to nitrogen enters the soil fr< rather than from the nodule: foliage of fling m 5“. nonhined nitrogen and estim ndiive feet tall adds 140 nenyean from the fallen 1e that the soil in alder thic nitrogen per acre by the en addition of fixed nitrogen found that the soil in aid us less than 5.0 within 3 letter (with oil 4.2 to 4 E ULnlated below the plantE filotest floor combined titanic carbon per square Other species of A1“ include w w in th intongfl” 1968) and ill in Guatemala (Standle iithe labels of herbarit iitlirtually ever l y spen on “it or burned-0' 399C195 . _. inc ana subsp not are regarded as 30 and 2113.33. forest, over 100 years. One of the roles of Aln_u§_may be the addition of nitrogen to the soil. Lawrence (1958) found that nitrogen enters the soil from alders mainly from the dropped leaves rather than from the nodules themselves. He showed that the autumn foliage of Ale viridis subsp. sinuata contains up to 3% (dry weight) combined nitrogen and estimated that an alder .thicket five years old and five feet tall adds 140 pounds of nitrogen to the soil per acre per year from the fallen leaves. Crocker and Dickson (1957) found that the soil in alder thickets had accumulated over one ton of nitrogen per acre by the end of seventy years. In addition to the addition of fixed nitrogen to the soil, Crocker and Major (1955) found that the soil in alder thickets changed from a pH over 8.0 to one less than 5.0 within 35 to 50 years. In this period, organic matter (with pH 4.2 to 4.6) six to seven centimeters deep had ac- cumulated below the plants. the l8-inch-deep mineral soil profile and forest floor combined had accumulated almost 4.0 kilograms of organic carbon per square meter. Other species of £1213 cited by authors as pioneers in succession include fly in the Pacific Northwest (Worthington,,_e_t_ 5%., 1962; Newton e_t'_ai., 1968) and A. acuminata var. acuminata in southern Mexico and in Guatemala (Standley and Steyermark, 1952). From information on the labels of herbarium specimens used in this study, it appears that virtually every species plays this role as well, occurring regu- larly on cut or burned—over land and in abandoned fields. Many Species (_A_. incana subsp. rugosa, A. serrulata, A. viridis subsp. crisps) are regarded as weeds, growing in ditches, along embankments, til pond shot one not desi in sonn pioneer in out that al occurring 0 hotly llounl favorable : :etessary '5 the tol thea 1“- Eliglam nlsenvatin the the 31 on pond shores, in wet pastures, and other such places where they are not desired. In some cases where A_lnln_s_ might be expected as an important pioneer in succession, it is absent. Tisdale flifl' (1966) point out that alders are not present in the successional vegetation occurring on recently glaciated land on Mt. Robson in the Canadian Rocky Mountains even though the environmental conditions seem to be favorable for its development there. Additional study will be necessary for a more complete understanding of this problem as well as the role in general of these plants in succession. McVean, in a series of papers on the ecology of A133 glutinosa in England (1953a, 1953b, 1955, 1956a, 1956b) makes a number of observations that appear to hold for the American species as well. Among these is the fact that the root system has both surface and deep branches, enabling the plant to survive in either waterlogged soil or where the water table is deeper. Only the surface roots bear nodules, however. He also shows that seed germination is in- dependent of light, normal temperature variation, and pH within the range of 3.5 to 8.0, but is sensitive to low oxygen and moisture levels. He (1953a) lists a large number of animal feeders, parasites, insect pests, and destructive fungi and discusses the effects of each on plants in various stages of their life history. The alder 11 general, th tees of cool-t 5511i. lifew W) an item. The 5' old climates Stocks, as di file SPQCies llatme liter to We illitger tr 3: 9"“ he on inaddllion inpnate 9. tether Shel humane“) Presence oi MORPHOLOGY AND ANATOMY Habit The alders are woody, ranging from small shrubs to large trees. In general, the species of warm latitudes are arborescent while the ones of cool-temperate, boreal, or montane regions are fruticose in habit. A few shrubby species (Alnus maritima and A. incana subsp. tenuifolia) approach tree stature, but nearly always have several stems. The shrub habit is regarded as an adaptation to cool or cold climates. It has apparently arisen independently in a number of stocks, as discussed below. Photographs of the habits of representa- tive species are provided in Plates 2 and 3. Stems Mature stems of Alnus range in diameter from about one centi— meter to over two meters. In most species the bark is smooth, though in larger trees it may be broken into flat plates (A. oblongifolia) or can be corky (A. jorullensis). Most of the Latin American species, in addition, develop transverse constrictions in the bark, as shown in Plate 4. 0n smooth bark, the lenticels are visable and range from rather small, inconspicuous, round spots (A. serrulata) to large, prominent, transverse markings (A. incana subsp. rugosa). The presence of smooth bark in the shrubby species, as well as the shrub 32 Plate 2. .33 Habits and habitats of representative species of Alnus. A, Alnus incana subsp. rugosa, large shrub on the bank of the Pigeon River in northern Indiana. B, A, rubra, medium- sized trees forming a dense stand on a river floodplain in west—central Washington. C, A, serrulata, medium-sized shrubs growing on the banks of a small stream in the Blue Ridge Mountains of southern Virginia. D, A, viridis subsp. crispa, medium-sized shrubs forming an understory in pine woods in central Alberta. - mug.-. _- 34 7" ,..v' Y, AV .\ “do .4 ‘1 w Plate 3. 35 Habits and habitats of representative species of Alnus. A, Alnus oblongifolia, large trees growing in a stream on Mt. Graham, Arizona. B, A, jorullensis var. firmi- folia, large tree on Volcéh Popocat petl, Mexico. C, A, rhombifolia, medium-sized trees on the banks of the Trinity River in northern California. D, A, viridis subsp. sinuata, large shrubs covering a subalpine mountain- side in Glacier National Park, Montana. Plate 3. Plate 4. 37 Bark of representative species of Alnus. folia, X 0.05. B, A, incana subsp. rugosa, X 0.3. C, A, jorullensis var. firmifolia, X 0.25. X 0.5. A, Pessim- serrulata, D, A- Plate 4. habit itself, teaching "mate The twig: pith in the o symetrical (1 indie to near end 1. m 5&0“le conc ends aw D“he suriav llfl- In tr The epi ilndular’ a densltd, col Slines. Tl “than along , lidtlcels a. lalitally e tn “laugh Stall) the heed tinged than those militia, In in 39 habit itself, may be a result of neoteny, these species never reaching "maturity" in an anatomical sense (Hall, 1952). The twigs, in cross section, have a triangular-shaped region of pith in the center. The shape of the pith varies from that of a symmetrical (equilateral) triangle in Aflg maritima and A. rhombi- folia to nearly linear in A. viridis subsp. crispa. In A. incana . and A. serrulata, the pith is equilaterally~triangular, but with strongly concave sides. The color ranges from a light buff in A. xiii- d_is_ and A. rhombifolia to a dark brown in A. incana, A. serrulata, and A. acuminata. Sometimes pronounced longitudinal ridges are evident on the surface of the twigs, especially in Alli—ug acuminata and A. gar—i- i“: In transection these ridges are seen to contain air chambers. The epidermis of young shoots is usually somewhat pubescent, glandular, and resin—coated. The pubescence and glands vary in density, color, and size, but are otherwise similar from species to species. The glands are invariably denser and larger at the nodes than along the internodes of the stems. On the youngest twigs, the lenticels are always oriented longitudinally, but these rapidly become laterally elongate with secondary growth of the stem. The leaf scars are triangular and about as high as broad. They bear five bundle scars, the upper two much larger than the lower three, which may be fused together. In Alnus rhombifolia, A. acuminata, A. rubra, and A. jorullensis the lower two lateral bundle scars are more elongate than those of the other species, although this varies from specimen to specimen. In Alnus viridis one can usually see a differentiation in the length of thd growing date: are borne di‘ the short sh The woo (nu) first of large nu] there is no and 1. “£9 adiiegnte 0 discusses ‘1 roman in SPdcies, in 915 phenon hierson an Egg”Bate ' “the nn 0 d. . as having no abSem E of s( itch Case Planes, a of “00d 0 in an of all on no 40 length of the twigs. The leaves are borne on short spur shoots growing laterally on longer branches. In other species the leaves are borne directly on the long shoots. After one to several seasons, the short shoots may elongate and become long branches. The wood of _Al_n1§_ has long interested plant anatomists. Bailey (1911) first showed a phylogenetic series involving the building up of large multiseriate rays from uniseriate ones. He noted that there is no inclination of any aggregation in A. incana, A. £92, and A. maritima. Alicis rhombifolia was shown to have completely fused aggregate or compound rays. In a second paper (Bailey, 1912) he discusses in detail the evolution of the compound ray and shows a reversal in the trend (from multiseriate to uniseriate) in several species, including Aln_us_ viridis subsp. crispa and "A. acuminata.” This phenomenon has also been discussed by Hoar (1916), Forsaith (1920), Anderson and Abbe (1934), and Hall (1952). The latter states that aggregate rays are absent in A. oblongifolia, A. iorullensis, A. ferruginea, and A. mirbelii. All of the above authors describe the wood of "AJE acuminata” as having uniseriate rays, and several of them note that growth rings are absent in this species. It is not certain, however, that A. E91111}..- nata of South America is what these workers actually had in mind. In each case, anatomical material was apparently obtained from cultivated plants, although the authors are not clear on this point. Examination 0f wood of all the species showed that multiseriate rays are present in all of them except A. viridis, which does, in fact, have uniseri- ate or partly biseriate rays. Some species, such as A. incana subsp. nosa, i' nitiseria senate re Then although i enter par idChallt ring of s in cork Hall Tel its 1 List con 39 Show Ell-iii in Perform in 50,, tissue, 11 the 41 rugosa, A. serrulata, and A. maritima have rather poorly-developed multiseriate rays, however (Plates 5 and 6). All species have uni- seriate rays between the multiseriate ones. The wood is made up primarily of vessels and fiber tracheids, although some true tracheids may be present in certain species. The outer part of the primary cortex is always collenchyrnatous (Metcalf and Chalk, 1950). The pericycle contains a composite and continuous ring of sclerenchyma, and the secondary phloem contains stone cells. The cork cells are almost tabular in shape. Hall (1952), in characterizing the family Betulaceae, its tribes, and its genera on the basis of wood anatomical features, provides the most complete description available of Alnus wood. Species differences are shown to occur with respect to vessel length, diameter, frequency, shape in cross—section, and the number of bars in the scalariform perforation plates. Both tracheids and fiber tracheids are present in some species, while only fiber tracheids occur in others. Inter- vascular pitting may be opposite, alternate, or transitional, depending on the species. Specialized features, according to Hall (1952), include alter- nate intervascular pitting, as opposed to opposite; a large or small number of bars in the perforation plates rather than an intermediate number; the absence of true tracheids; and uniseriate rays by reduction from aggregate rays. Other features in an advanced state include a reduced frequency of vessels, angular vessel shape in cross section, and a large number of vessels per cluster. Unfortunately, Hall did not examine fossil alder wood in the collection of his data. Such Plate 5. 42 Photomicrographs of Alnus wood. A, Alnus acuminata var. acuminata, radial section showing aggregate rays and large vessels, X 160. B, A, acuminata var. acuminata, longitudinal section showing rays and details of the vessel wall, X 395. C, A, acuminata var. glabrata, longi— tudinal section showing uniseriate and biseriate rays and details of the perforation plate of a vessel, X 395. D, A, incana subsp. rugosa, radial section showing little tendency for aggregation of the rays, X 160. E, A, incana subsp. tenuifolia, radial section showing aggregation of rays and moderate-sized vessels, X 160. F, A, jorullensis var. jorullensis, radial section showing multiseriate rays, X 160. WW fiai " i filly W 43 0" . div. .1. _ 3a\\m Q. ' 0.5.: :0- .9 a 0" Plate 5. Plate 6. 44 Photomicrographs of Alnus wood. A, Alnus iEEEllSEEii var. jorullensis, radial section showing "wax chambers, " X 160. B, A. maritima, radial section showing small vessels and a small multiseriate ray, X 160. C, A. rhombifolia, radial section showing a well- developed multiseriate ray, X 160- D, A. viridis subsp. crisps, radial section showing small vessels and uniseriate rays, X 160. E, A. viridis subsp. crispa, longitudinal section showing the—vertical nature of the rays, X 395. F, A, viridis subsp. sinuata, radial section showing slight aggregation of the rays, X 395- ‘3 '31:; a; If 1-2 ‘21" n», nstudy nigl advanced con ldany o cation, and haddition the partied ii the spot however, an lolloving The n uniseriate Hill a rel; tine, hove ll! voode it losses Elternate lih node in losses “the: Sp: Wits its [Hunt it ind m m thtin h m“ llllits ’ 46 a study might prove useful in the determination of primitive and advanced conditions in wood traits. Many of the species names used by Hall are of uncertain appli- cation, and he does not include documentation of the specimens studied. In addition, in many cases he does not name the species showing the particular features discussed, so his work is of limited value at the species level. From the information that is available there, however, and from my own observations, it is possible to make the following general statements. The wood of Alnus viridis is specialized in that it possesses uniseriate rays apparently derived from multiseriate ones by reduction and a relatively large number of vessels per unit area. It is primi- tive, however, in its small vessels and opposite intervascular pitting. The wood of Alnus incana and A. serrulata is somewhat advanced in that it possesses moderate—sized vessels of intermediate frequency and alternate intervascular pitting, but it is primitive in that it has only moderately well—developed multiseriate rays. Wood of w maritima is advanced in its small, infrequent vessels, but primitive in possessing aggregate rays. Alnus jorullensis is distinct from the other species in having very angular vessels of medium size. This species is primitive with respect to its high frequency of vessels and its multiseriate rays. Alnus rhombifolia, A. oblongifolia, and A. acuminata are generally primitive in their aggregate rays, but advanced in their numerous, large, circular vessels. Thus, each species can be seen to have become specialized in some wood features and not in others, probably in response to the particular environmental conditions present in The st in length a Stipules, ‘ Eeing part Vith acute on the ab; Ms thaI 47 present in its habitat. Stipules The stipules vary greatly in size, ranging from about 4 to 15 mm in length and from about 0.5 to 6 mm in width, as shown in Plate 7. Agng§_viridis, A, serrulata, and A, acuminata have relatively large stipules, while A, maritima and A, oblongifolia have much smaller ones, being particularly reduced in width. Mostly, they are ovate in shape with acute tips. Like the leaves, they are pubescent and glandular on the abaxial surface. The margins are lined with longer and coarser hairs than those borne on the surface. Winter £393 The buds of A1225 are stalked (though nearly sessile in A, viridis). In all of the species except A? viridis they are protected by two Slightly modified stipules and sometimes one or more of the stipules of the next level of the bud (Plates 8 and 9). These may cover the entire bud or be reduced in size or entirely absent (A, maritima, A. oblongifolia). The stipular scales are equal in size, and when large enough, valvate. Buds of élEEE viridis are covered by several unequal bud scales apparently derived from stipules. In all species, flm stalks and scales are pubescent, glandular, and covered with resinous secretions. Within the buds the leaves are conduplicate and usually more or less plicate within their stipules. The degree 0f development of the young leaves, however, varies considerably in different species. Plate 7. 48 Stipules and expanding leaves of representative species of Alnus. A, Alnus incana subsp. rugosa, X 1.5. B, A, maritima, X 1.5. C, A, rhombifolia, X 1.5. D, A, viridis subsp. cris a, X 1.5. Plate 7. 50 Plate 8. Twigs with winter buds of representative species of Alnus. A, Alnus oblongifolia. B, A. maritima. C, A. rubra. D, A. serrulata. _’ Swizz::zgzsaz:Sarga— m. w A 5. :E:553:3: ) F _ . 7312C 2:: Plate 8- Plate 9. 52 Twigs with winter buds of representative species of Alnus. A, Alnus incana subsp. rugosa. B, A. incana subsp. tenui- folia. C, A, rhombifolia. D, A, viridis subsp. crispa. .IIII7IITIIIIIII111‘I. 53 ’LL'}! If bbbbbbbbbb u o o ”are:_:_:::._:_ .fyi _‘f 3. Plate 9. The 1e; petioles are The leaves M an leaves with of seedling species in North Amer hisuhSp. Species, '1 lbevate 1e sion of t} for Each p llttured [he later form Subs Venetian ihile in 1913). p veins, in 54 Leaves The leaves of Alnus are borne alternately along the twigs. The petioles are not inflated and are usually rather deeply grooved above. The leaves are typically ovate and double—serrate. However, Alnus maritima and other members of subgenus Clethropsis have much narrower leaves with low, distant, upturned, single teeth (though the leaves of seedlings of A, maritima may be ovate and double-serrate). Many species in the genus have laciniate forms (cf. Hylander, 1957). In NorfliAmerica, local populations of A, incana subsp. rugosa, A, viri- dis subsp. sinuata, and A, rubra demonstrate this condition. Several fimcies, including A, maritima, A, serrulata, and A, jorullensis, have obovate leaves, apparently derived from the ovate form through compres- sion of the apical region. The range of variation in shape and margin for each American taxon is shown in the series of leaf outlines pictured in Figures 1, 2, and 3. The leaves are pinnately veined, the lateral veins sometimes branching one or two additional times to form subsecondaries near the margins, especially toward the base. Venation in Alnus maritima is semicraspedodromous to eucamptodromous, while in all the other species it is simply craspedodromous (cf. Hickey, 1973). The size of the leaf has little effect on the number of lateral veins, but it does on their distance from each other. The lateral veins Of all species form about the same angle with the midrib (between 30 and 40 degrees). Between the lateral secondary veins run tertiary cross- Veins. These are well—developed, forming ladder-like reticulations Hisome species, but they are incomplete in others. Within these reticulations occur well-developed oreoles containing simple to Figure l. 55 Outlines of leaves of Alnus showing extremes in the patterns of shape and margin variation: l—lO, Alnus acuminata var. acuminata (South America); 11-20, A: acuminata var. acuminata (Mexico and Central America); 21-27, A, acuminata var. glabrata. h 57 Figure 2. Outlines of leaves of Alnus showing extremes in the patterns of shape and margin variation: 1—4, Alnus oblongifolia; 5-9, A. incana subsp. rugosa; lO-l4, A. incana subsp. tenuifolia; 15-21, A. jorullen81s var. firmifolia; 22-30, A. jorullensis var. m3 31-35, A. maritima. Figure 3. rhombifolia; 8-15, A, 59 Outlines of leaves of Alnus showing extremes in the patterns of shape and margin variation: 1—7, Alnus rubra; 16-20, a- seamless; 21-28, A, viridis subsp. crispa; 29-35, A, viridis subsp. sinuata. 7‘ l h l l }\ flfiii?‘ he hhheee he h he e h h e e ‘ he hi hi! ewes? : as branching \ Leaf I especially tipped app the margin truncate, apex is us truncate r The ‘ najor vei Sentiallp 0“ the V( tarp in ‘ liarance fWild p hands m hep 0 61 branching veinlets. Leaf margins are somewhat thicker than other parts of the blade, especially at the apices of major teeth, giving them a glandular— tipped appearance. In some species (best illustrated by élEE£.£EE£i) the margin is revolute (Plate 10). The base is usually rounded, truncate, or cordate, but it may be extended or even attenuate. The apex is usually acute, but its shape ranges from long-acuminate to truncate to deeply notched in some species (Figures 1, 2, and 3). The foliage of all the alders is pubescent, especially along the major veins on the abaxial surface, although a few taxa such as fléflfifi viridis subsp. sinuata and A, acuminata var. glabrata are es- sentially glabrous. The hairs are simple, straight, and borne mainly on the veins and veinlets, even in densely-pubescent forms. They vary in length, color, and density, but not in general shape or ap- pearance. The leaves of all species of A1323 are glandular as well as pubes- cent (Plate ll). The glands, which are—more frequent on the lower sur- face and petiole than above, are described by Metcalf and Chalk (1950) as "consisting of a short but broad stalk composed of low, suberized cells and a shield—like head made up of cells appearing to be polygonal in surface view, but resembling a palisade in sections of the gland.” The development of the glands is described by Dorman (1924), who found that they secrete a high molecular weight polyterpene. The glands may be large, yellow, and dense, as in Alngg jorullensis var. jorullensis, or small, dark, and sparse, as in A, incana subsp. rugosa. They often darken with age, especially on the adaxial surface. The Plate 10. 62 Macroscopic features of Alnus leaves. A, Alnus viridis subsp. crispa, pubescent extreme (A, crispa var. mollis Fern.), X 3. B, same as A, X 46. C, A, maritima, glandular teeth, X 3. D, A, rubra, revolute margin, X 3. 63 Plate 10. Plate 11. 64 Epi-illuminated microscopic views of abaxial leaf sur- faces of Alnus showing veins, resinous deposits, and glands, all X 46. A, Alnus acuminata var. acuminata. B, A. jorullensis var. firmifolia. C, A. jorullensis var. jorullensis. D, A. viridis subsp. sinuata. -——-———-— --_-._ . M...._.._. ._ _ /- -._ a. Plate 11. leaves, like t posit of resir species. 0f 1 glutinous. The smart treated by B0 in its leaf a histological] cnlaceous (Sr and present r Present just MW] in s dermis is f, M starrinate T late “their arranged Sp braces, on cases ' Th b“ air it W 1 “19) whul “fl “its Qatk 66 leaves, like the twigs and buds, are often covered with a thick de- posit of resinous material, but the amount varies from species to species. 0f the American species, A, viridis is by far the most glutinous. The anatomy of leaves of the Betulaceae was comprehensively treated by Boubier (1896), who shows that AAEEE varies considerably in its leaf structure and describes the blade, veins, and petiole histologically (cf. also Solereder, 1908). The Stomata are ranun— culaceous (surrounded by several irregularly-arranged epidermal cells) and present only on the abaxial surface (Plate 12). A hypodermis is present just below the epidermis in some species, being better de- veloped in some of these than in others (Boubier, 1896). The hypo- dermis is found in species of all the American subgenera. Inflorescences Alflgg is monoecious with the flowers borne in catkins, the Staminate pendent and the pistillate erect (Plate 19). The pistil- late catkin was shown by Abbe (1935) to be composed of numerous cymules arranged Spirally on a primary axis. Each cymule has three levels of bracts, only the two tertiary florets being present except in rare cases. The staminate catkins are likewise made up of tiny cymules, bUt all three florets are usually present. Alnus viridis and A, EEEEEEEA lack one of the two tertiary bracts of the staminate cym- Ule, While both are usually present in all of the other species. Alnus maritima and its allies in eastern Asia bear the stami- ——___________ Hate catkins singly in the axils of leaves at the tips of branches Plate 12. 67 Photomicrographs of abaxial leaf cuticle imprints of Aiggg made on thin acetate sheets showing stomata, epidermal cells, hairs, and glands, all X 160. A, Alnus Agggiflgfig var. acuminata. B, A, acuminata var. glabrata. C, A; 227 longifolia. D, A, incana subsp. tenuifolia. E, A; jQEElf lensis var. jorullensis. F, A, maritima. 0‘s oo , \Iw' . "Fist ‘ "it” e A I, z a. 7 1 ’. l??? ,9 k =. 7:; .iffi‘r ‘>-” .g‘ _,. ‘ . xx. - , if‘fgtlygffl' Plate 12- 9..., r ., l l f I : c I" I". . / . , . V v r": l ‘ .Q. v .1 w, , V‘ \\ “H gr}; ., ' « a1... is J-§’ 5 , J and the pistil just below the late catkins i larp staminate appear in brar subgenus Am retain the su‘ while in the the internode are seen on t 0r staminate above the pig Staminate f1, 30; the Stain The pig “‘1“ reduncr littlunclesj a hem-mg nea pistillate < lass 0f the lip. Murai lsnerpc Ser be“ 0f sub 0 n the "Pet HIIIIIIIIIIIIIF_______________________________i 69 and the pistillate catkins singly in leaf axils on the same branch just below the staminate. All of the other species bear the pistil- late catkins in racemose clusters just below the solitary and axil- lary staminate catkins, which occasionally on vigorous plants also appear in branched clusters. In élEEi viridis and other members of subgenus Alnobetula, the staminate and lower pistillate catkins retain the subtending leaves, which may be somewhat reduced in size, while in the other species, all or most of the leaves are lost and the internodes appreciably shortened. Sometimes pistillate catkins are seen on the staminate branches just below the staminate catkins, or staminate catkins may appear at the tips of the pistillate branches above the pistillate catkins. Occasionally, both pistillate and staminate flowers may appear in the same catkin, and when this is so, the staminate are usually apical. The pistillate catkins of Alnus viridis are pendent on long, thin peduncles. Alnus maritima has somewhat stouter and shorter peduncles, and the remainder of the species have still shorter ones, becoming nearly sessile in A, incana and A, serrulata. Where the pistillate catkins occur in racemose clusters, the catkins near the base of the cluster always have longer peduncles than those at the tip. Murai (1964) regards the single, axillary pistillate catkins of Alflg§_maritima as advanced over the clustered type and shows a phylo_ genetic series of reduction leading to this state in the Asian mem_ bers of subgenus Clethropsis. In Alnus maritima nodes are evident on the "peduncle" of each solitary PiStillate catkin, suggesting this reduction subgenera m condition in w is shown by ti on the same bi kins are deri' appearance of At natur “WT and con ““816 persis CODES are oft Shape and si: “Shane and 0f the cone Shape, degre lobe. These features of lenus B . a c The f1, “i“d‘eollin formed Dell in Si“Sic have fOUr T ,2 h c: S / w TIIIIIIIIF___________________________________——__7 70 this reduction. That the short axillary pistillate branches in the subgenera A1231 and Alnobetula are already reduced from a primitive condition in which both pistillate and staminate catkins were present is shown by the occasional presence of both kinds of inflorescences p on the same branch. Similarly, that simple axillary staminate cat- kins are derived from branched systems is shown by the occasional appearance of such systems in vigorous specimens. At maturity, the pistillate catkins of all the alders become woody and cone-like, the five bracts of each cymule fusing into a single persistent scale bearing the two fruits. These woody mature~ cones are often useful in identification, varying considerably in shape and size (Plates l3, l4, and 15). The cone scales also vary in shape and size, but this is mostly correlated with the dimensions of the cone itself. Other variation in the cone scales involves the shape, degree of thickening, and amount of reflexing of the terminal lobe. These woody infructescences are one of the most distinctive features of A133: and the most useful in distinguishing it from the genus Betula. Flowers The flowers of Alnus are unisexual and minute, as in most other wind—pollinated plants. The staminate flowers have a tiny but well- formed perianth with from two to six parts, usually appearing to be in a single whorl. Most species of the subgenera Alnus and Clethropsis have four perianth parts and an equal number of stamens opposite them. Alnus rhombifolia commonly has only two perianth segments and two —___—_ 71 Plate 13. Mature infructescences of representative species of Alnus- A, Alnus acuminata var. acuminata (South America). acuminata var. acuminata (Mexico). glabrata. D, A, incana subsp. rugosa. C, A. acuminata var. B, A- o i h 72 l l VE'HeC Ir 1 lll’llplrr‘li"‘ p a ,1 l ‘IT‘OTfrrwhofi‘ Plate 13. 73 Plate 14. Mature infructescences of representative species of Alnus. A, Alnus incana subsp. tenuifolia. B, A. jorullensis var. M. C, A. jorullensis var. jorullensis. D, A. mari- tima. 74 Plate 14. 75 Plate 15. Mature infructescences of representative species of Aggie A, Alnus rhombifolia. B, A. rubra. C, A. M. D, A. viridis subsp. sinuata. nunscl Z 3 ‘ 5 p p ‘1 I’ll il‘ “1 Plate l5. stamens, and A. tepals and two four or five te six perianth pz libhe, 1935). The stamer Parts. They a: the catkin. I “(find the len ”Jill intermedia the filament 1 The PiSti “mi of a Sp one, together lild). Abbe found in the ‘ has m" Poorl in each gem his (1935) l 3 bills) 1966) The War who f°llnd the ring 0n the ( sometimes am he found trur 77 stamens, and A, oblongifolia usually has four, sometimes with two tepals and two stamens smaller than the other two. In Alggg viridis, four or five tepals and stamens are the rule, but occasionally up to six perianth parts and six stamens may be present, each in two whorls (Abbe, 1935). The stamens vary in length and degree of fusion to the perianth parts. They are tetrasporangiate and never exserted very far from the catkin. In élEEi incana, the filaments are subsessile, never ex- ceding the length of the tepals. They are long in A. viridis, A, magi— tima, A. oblongifolia, A. rhombifolia, A. rubra, and A. serrulata, and intermediate in the remaining species. In Alnus incana of Europe, the filament length is variable. The pistillate flowers are normally without a perianth, consisting mainly of a single bicarpellate ovary derived from a tricarpellate one, together with a simple two—branched style (Abbe, 1935; Hjelmqvist, 1948). Abbe (1938) noted that single tricarpellate pistils are often found in the axils of foliage leaves below the catkins. The ovary has two poorly—defined locules, becoming one above the single ovule in each section. The placentation is basically axile according to Abbe (1935); the ovules are anatropous, unitegmic, and crassinucellar (Davis, 1966). I The ovary in A1223 was shown to be inferior by Abbe (1935, 1938), Who found that three or four small elongate glands of the type occur- ring on the edges of the perianth parts of the staminate flowers sometimes appear at the summit. In several specimens of Alnus rubra he found true tepals in this position, with the glands now at their spices. Four sr two were noted llorld. In addi the glands in _A On the has the most primit W is t regard to inflc this character Pollen gr in abundance . 3% prod ll llercus rib Em“ is about and 37.0 X 10. that alder pol Posed Severn windhorn P0111 mscspraine is The Srai hoping in di tench, an d th occasionally hie aspidate 78 apices. Four such glands were found in A, subcordata of Asia, and two were noted in A. maritima, A. incana, and A. rubra of the New World. In addition, vestigial vascular traces were seen running to the glands in A. incana. On the basis of staminate floral morphology, Alnus viridis is the most primitive species, having lost the fewest parts, and A, 22: longifolia is the most advanced, having lost the most. Likewise, with regard to inflorescences, A, viridis is the most primitive, but for this character A, maritima is the most advanced. Pollen Pollen grains of 5&5551 are small, light in weight, and produced in abundance. Erdtman (1969) estimates that a single catkin of élEEE glutinosa produces as many as 4,500,000 grains, compared with 1,250,000 in Quercus £222; and 175,000 in £3522 sylvatica. The weight of a single 9 grain is about 6.5 X 10- gm, compared with 3.8 X 10"9 gm in Juniperus and 37.0 x 10'9 gm in F2533 (Erdtman, 1969). Wodehouse (1935) states that alder pollen is "caught in great abundance on pollen slides ex- posed several miles from any trees.” Where Alggg is abundant, the windborn pollen is sometimes responsible for causing hayfever in susceptable individuals (Chamberlain, 1927). The grains are isopolar, radiosymmetric, stephanoporate monads ranging in diameter from 17 to 35 microns. They are somewhat flat- tened, and they typically have from three to six apertures, though occasionally more than six pores may be present. The apertures are aspidate and often somewhat elliptical, ranging from 2.5 to 4.5 microns long. EaCh all or slightly granular but can be seen to be spinules (nanoverruca low ridges when viewe ll). The sexine is p l969). Thickened bar run in pairs from ape appearance to shrunk: these bands are not . material. Below the t° form large onci. To find norphol grains were examined as discussed by Ridg h erbarium specimens ihccrine jelly perp stained light red W' rl - ESQ grains were S tersion at a nagnif Seven grains on e30 to determine diamet per species) were e per grain. For see her . harm's Specimens adh - ssrwe cellophane 79 microns long. Each aperture has an annulus. The surface is smooth or slightly granular in appearance under the compound microscope but can be shen to be covered with very minute verrucae, gemmae, or spinules (nanoverrucae, nanogemmae, or nanospinules) positioned on low ridges when viewed with the scanning electron microscope (Plate 17). The sexine is pertectate and thicker than the nexine (Erdtman, 1969). Thickened bands (arcus) in the exine (tectum and nexine) run in pairs from aperture to aperture, giving a very characteristic appearance to shrunken or collapsed grains (Plate 17), although these bands are not always evident in fully expanded and unstained nmterial.n Below the pores, the intine is thickened considerably to form large onci. To find morphological differences in the pollen of filflflfi Species, grains were examined using both light and scanning electron microscopy, as discussed by Ridgway and Skvarla (1969). Pollen from three to seven herbarium specimens of each Species was collected and mounted in glycerine jelly perpared by the formula of Johannsen (1940) and stained light red with safranin-O. Within one week after preparation these grains were studied, measured, and photographed under oil im- nwrsion at a magnification of 630 X with a Zeiss Photomicroscope 11. Seven grains on each slide (21 to 49 grains per species) were measured to determine diameters, and 100 grains per slide (300 to 900 grains per species) were examined to establish the average number of pores FBI grain. For scanning electron microscopy, pollen collected from herbarium specimens was attached to metal stubs by means of double- adhesive cellophane tape, coated with gold, and examined with an Plate 16. 80 Pollen grains of representative species of Alnus,_all X 1300. A, Alnus acuminata var. acuminataé i, #3 22:0:;bsp gifolia. C A. acuminata var. glabrata. , _3 inc . - incana. E,’AT incana subsp. tenuifolia. F, A, jorullenSIS var. firmifolia. G, A: jorullensis var. jorullen31s. r— H, A. maritima. I, A. rhombifolia. J, A. rubra: .K, Ab Ee— rulata. L, A. viridis subsp. 55:5353. M, g, VlrldlS'Su Sg- sinuata. N,:é. viridis subsp. §}nuata. 0, g, v1r1dls mlspo viridis. l 81 Plate 16. Hu— 0 ~m -~~ A-‘-.-—W- ~ _w-. ,._. . a “mm-“JIM . - J ‘Mfl u~ nu.— -‘h m A, .; Plate 17. 82 Scanning electron micrographs of pollen grains of repre' sentative species of Alnus. A, Alnus incana subsp. Eggggy X 2500. B, A. incana subsp. tenuifolia, X 1500. C, A; jorullensis var. jorullensis, X 2250. D, A, maritimaa. X 2250. E, A. rubra, X 2250. F, A, viridis subsp. 25332? X 2500. G, _A_. viridis subsp. sinuata, X 2250. H, A. £33“ £33 subsp. sinuata, X 2250. 83 Plate 17. Advanced Metals 1 Michigan State U1 lection of polle Few tangibl diameter, were n ation in the co] and the size of size ranges of 1 but the mean va‘ small grains, a: Three principal lending rOUghly W) has t m. Wong, SUbger >I'5I-I_ o- - H r: ,_. 7° .5; r—‘ H H. r. H. S a, [’15:- l?— 19>]?va [3’] <. <. ,5- r‘é’ g S" "f "I 0" o . . . r—v r-1 5 0 Q— C ED 0" S . . ,_.. ,_. H. m D, )—h H' y-r 0 Ph (I) n: r—' O c: ,_.. H .— m e... 84 Advanced Metals Research Model 900 scanning electron microscope at Michigan State University. Herbarium specimens used for the col- lection of pollen grains are listed in Table 19 (appendix). Few tangible differences, other than pore number and grain diameter, were noted in the material studied, although slight vari- ation in the color of the grains, the sculpturing of the surface, and the size of both the aspides and the onci was detectable. The size ranges of the grains of all species overlap (Table 1, Figure 4), but the mean values of some, such as élEEE maritima, which has very small grains, are useful in distinguishing them from the other taxa. Three principal groups based on size can be recognized, these corres- ponding roughly to the three subgenera. thropsis) has the smallest grains, with average diameters of only 19.7 microns, subgenus Alnobetula, represented by A. viridis, has larger grains, Table 1. Size of Alnus pollen grains (microns). Species Mean Diameter Range ' A. acuminata var. acuminata 23.83 18.6 - 29.5 A. acuminata var. glabrata 23-78 20.2 — 26.4 A. incana subsp. incana 24-49 18-6 - 29-5 A. incana subsp. rugosa 22-80 19-4 - 25.6 A. incana subsp. tenuifolia 23.60 13-5 - 27.9 A. 'orullensis var. firmifolia 23.14 20.2 - 26.4 A. jorullensis var. jorullensis 24.34 21.7 - 27.9 A. maritima 19.67 17.1 - 23.3 A. oblongifolia 24.34 20.2 - 27.9 A. rhombifolia 24.51 21.7 - 28.7 A. rubra 25.02 20.0 - 27.9 A. serrulata 21.06 18.6 — 24.8 A. VLrLcis subsp. crispa 21.06 18.6 - 23.3 A. vnricis subsp. sinuata 21-83 17-1 - 26-4 A- VlrlCiS subsp. viridis 22-08 18-6 - 26-4 Alnus maritima (subgenus Cle- viridis A.virb sinuata A.viri crispa A. serr _— b r A. A. rhon ‘ A.obh — § ‘ - mar l as, I1 A\ Em C Q in\6 A. viridis subsp. viridis A. viridis subsp. sinuata A. viridis subsp. crispa A. serrulata A. rubra A. rhombifolia A. oblongifolia '5“ maritima IK> . jorullensis Var. jorullensis A. jorullensis var. firmifolia A. incana subsp. tenuifolia A. incana subsp. rugosa A. incana subsp. incana A. acuminata var. glabrata A. acuminata var. acuminata 85 ins o . Range of variation in diameter of Alnus pollen gra Figure 4 Table 2. Percent of Species i. acuminata var. :3 i. M subsp. m a Les subsp. re e- m subsp- a i. jorullensis var. i. jorullensis var. i. maritima - \ Miridis subsp. c 5- Viridis subsp. E é-‘Iiridis subsp. \_I \ with mean diameters naming species, me diameters, ranging Pore number is than size, as showr Nita) tl’Pically 1 d3 subsp. $92. A. ' . mullensiS Var AI 7 %) Usually subspp‘ I. w and these scum-mg in folia \) A. rhomb. \ lfc \ at llltl %) and A. let ed trains in eh" - my in Pore n1 86 Percent of Alnus pollen grains having various numbers of pores. Table 2. Number of Pores 1 Species 3 4 5 6 A. acuminata var. acuminata 0.0 25.6 73.3 1.1 A. acuminata var. glabrata 0.0 31.6 67.8 0.6 A._}ncana subsp. incana 0.0 38.0 58.6 3.4 A. incana subsp. rugosa 0.0 46.8 51.4 1.8 A. incana subsp. tenuifolia 0.0 70.3 29.8 0.0 A. _orullensis var. firmifolia 0.0 14.3 82.0 3.7 A. orullensis var. jorullensis 0.0 14.4 82.0 3.4 A. maritifl 6.0 86.0 8.0 0.0 A. oblongifolia 0.0 33.2 64.6 2.2 A. rhombifolia 0.0 23.4 76.0 0.6 A. rubra 0.0 7.8 76.0 16.2 A. serrulata 11.0 80.6 8.4 0.0 A. virrcns subsp. crispa 0.4 10.2 82.0 7.4 A. viricis subsp. sinuata 0.0 9.6 82.4 0.0 A, viric1s subsp. viridis 0.0 7.4 77.2 15.4 with mean diameters ranging from 21.1 to 22.1 microns, and the re- maining species, members of subgenus Alnus, have still larger mean diameters, ranging from 21.1 to 25.0 microns. Pore number is often a better criterion for distinguishing species than size, as shown in Table 2. Two species QA. maritima and A, ser- rulata) typically have grains with four pores, while six taxa (A, viri- dis subsp. crispa, A, viridis subsp. sinuata, A, viridis subsp. viridis, A. 'orullensis var. firmifolia, A. jorullensis var. jorullensis, and A. rubra) usually Show a five-pored configuration. Two taxa QA. incana subspp. rugosa and incana) have either four- or five—pored grains, these occurring in about a one to one ratio, and five taxa QA. oblongi- folia, A. rhombifolia, A. incana subsp. tenuifolia, A, acuminata var acuminata, and A. acuminata var. glabrata) have both four- and five— pored grains in ratios of two or three to one. In most cases, vari— ability in pore number is high, with up to 35% of the grains of a Particular SPeCime“ 1' 513% M13 sul iour-pored condition night be considered apertures is probabl by Doyle (1969). re is indicated by the otherwise quite sper taxa having five or more generalized. iron the Tertiary 0 “1th a tendency ton be interpreted to 5 Four groups 0: sUriace sculpturin 17). The first of mm, shows game Positioned in . subsp. 51ml\atai i. 1 ‘ % subsp. 1 dens e v ETTUQae’ tl ridges. The Verr and h se % , . hird grOUp’ cont V er, %s A. mederateh‘long 1 87 particular Specimen lying outside the "typical" number. Alnus incana subsp. tenuifolia demonstrates a tendency toward the four-pored condition shown by A? serrulata. The four-pored condition nfightbe considered advanced in.él22§: even though a small number of apertures is probably the ultimate primitive condition, as suggested by Doyle (1969). That four pores is an advanced condition in.élfl§§ is indicated by the fact that it occurs mainly in species that are otherwise quite specialized (A, serrulata and A, maritima) while those taxa having five or six pores (A, viridis subspecies) are otherwise more generalized. In addition, the fact that "in fossil alder pollen from the Tertiary of west Scotland, increased aperture number is linked with a tendency towards a panotreme condition" (Erdtman, 1969) could be interpreted to support this idea. Four groups of Species can be distinguished on the basis of surface sculpturing, as seen in scanning electron micrographs (Plate 17). The first of these, represented by Alnus viridis subspp. crisEa and sinuata, shows grain surfaces densely covered with prominent gemmae positioned along short low ridges, the elements being larger in subsp. sinuata. A second group, represented by A? serrulata, A. incana subsp. ru osa, and A. incana subsp. tenuifolia, has fairly dense verrucae, these positioned along longer and more prominent ridges. The verrucae are very dense in A; incana subsp. tenuifolia and A. serrulata and somewhat less so in A, incana subsp. rugosa. A third group, containing A. acuminata var. acuminata, A. acuminata -————-_ ——_——-—.—.__.___ Var. glabrata, A. iorullensis, A. rubra, and A. rhombifolia, has low, Hmderately-long ridges with only moderately dense verrucae. Finally, p. M shows quent elements . With the ligl may be distinguisl that it is genera ranin) and has sm rounder or less a appearing grains, and they bear mec m and my 5. w a, “115, While A. 3 but of intermedi Small aSpides, m AS noted by on Pollen morphc ologT. The sub: is ~with While relatEd t, Olin as well. A to be c103,,t a The fruits l . I I they have been 88 A. maritima shows pronounced long ridges with very small and infre- quent elements. With the light microscope, pollen from subgenus Alnobetula may be distinguished from that of species in the other subgenera in that it is generally lighter in color (not as deeply stained by saf- ranin) and has smaller onci and aspides, which give the grains a rounder or less angular appearance. Alnus acuminata has rather light- appearing grains, but not as light as those of subgenus Alnobetula, and they bear medium to large aSpides and onci. Alnus incana subspp. rugosa and tenuifolia, as well as A, serrulata, A, oblongifolia, and A. rhombifolia all have medium to large aspides and onci and dark walls, while A, maritima and A, EEEEE have grains of similar shape but of intermediate shade. Finally, A, jorullensis bears relatively small aspides, medium—sized onci, and is of intermediate darkness. As noted by Stern (1962, 1970) in Dicentra, groups in élEEE based on pollen morphology cluster in the same ways they do with gross morph- ology. The subspecies of A, viridis form one group; A, incana and A, serrulata another; and A, acuminata a third. AAAAA jorullensis, while related to the A, acuminata complex, retains an identity of its mmtas well. Alnus rubra, A, rhombifolia, and A, oblongifolia appear to be closest allied to the A, acuminata group. Fruits The fruits of Alnus are small, smooth-surfaced, light, dry, indehiscent, and laterally-winged (Plate 18). In the literature they have been termed both nutlets and samaras. In Alnus viridis, Plate 18. 89 Fruits of representative species of Alnus, all X 12. A, A. acuminata subsp. acuminata. B, A. oblongifolla. C, A. incana subsp. rugosa. D, A. maritima. E, A. ser- rulata. F, A. viridis subsp. sinuata. llllllll each wing is as wide while other species are essentially wing otherwise occur. in rather than wind as the fruit is crowner usually elliptic in wider at the summit and base, than the l 91 each wing is as wide as or wider than the body of the fruit itself, while other species have narrower wings. A few taxa (A, maritima) are essentially Wingless, having only low ridges where the wings would In these species, dispersal is apparently by water otherwise occur. rather than wind as in the species with winged fruits. The summit of the fruit is crowned by two persistent styles. The fruit body is usually elliptic in shape, and the wings, when present, are often wider at the summit than below and extend further, both at the apex and base, than the body. The alders are attracted to the st genera Alni and Al perate zone, from 1‘ America, and as lai subgenus We in Alfl maritima) catkins are produc are exposed during staminate catkins exposed during the with the new folia hoth staminate ant “littering. AnthES gen“ Aw. whrr. the bud in Silbgeni out in SUbgenus C Staininate me sinner (September 5 W: and in Silbgenus m T—I— REPRODUCTIVE BIOLOGY The alders are anemOphilous, although insects are sometimes attracted to the staminate catkins (Plate 19). Species of sub- genera Alnus and Alnobetula flower early in the spring in the tem- perate zone, from November to February in southern Mexico and Central America, and as late as July in northern Canada, while those of subgenus Clethrppsis bloom in the early autumn (September or October in Alnus maritima). In subgenus Alnus, both staminate and pistillate catkins are produced during the growing season prior to blooming and are exposed during the winter. In subgenus Alnobetula, only the staminate catkins are produced during the prior growing season and eXposed during the winter, the pistillate catkins appearing along with the new foliage. Members of the subgenus Clethropsis produce both staminate and pistillate catkins during the same season as flowering. Anthesis takes place before the leaves appear in sub- genus Alppg, while the leaves are still small or just emerging from the bud in subgenus Alnobetula, and when the plants are fully leafed out in subgenus Clethropsis. Staminate meiosis was found by Wetzel (1929) to occur in late summer (September) in members of subgenus Alnus (A, rubra, A. glutinosa, and A, cordata) and somewhat earlier (July and August) in subgenus Alnobetula (A, viridis). Meiosis occurs in August 92 Plate 19. 93 A, pistillate and staminate inflorescences of Alnus . ————— ' serrulata. B, syrphid flies visiting staminate catkins of Alnus viridis subsp. crispa in Terra Nova National Park, Newfoundland (photograph courtesy of Garrett E- Crow). Plate 19. and September (immec _ (e- .2)- tlcVean (1953a) protogynous, althou Casual observation pattern may be true available. The age of fir hilt it probably oc< Species. Lawrence in Alaska reaches of p w gro ten reached reprod ternined from annu l. in _ cana subsp. 5 all ( reproduced fir e- e 5' M in general not 56‘ large. To plants 11' ing season and a subalpine climate an earlier reprod reduction under 5 with anatomical i v w by Forsait} 95 and September (immediately before anthesis) in subgenus Clethrop- S_i_§ (A. maritima). McVean (1953a) reports that élEEE glutinosa is nearly always protogynous, although up to 12 percent protandry has been observed. Casual observation of the American species indicates that this general pattern may be true for them as well, although no definite data are available. The age of first floWering of most Species of A1235 is not known, but it probably occurs early in the life of at least the shrubby species. Lawrence (1958) reports that A1233 viridis subsp. sinuata in Alaska reaches flowering age in less than seven years. A. specimen of A. maritima grown in an eXperimental garden in East Lansing, Michi- gan reached reproductive maturity at the age of three years, as de- termined from annual rings. Individuals of A. viridis subsp. crispa A. incana subsp. rugosa, and A, serrulata grown in the same garden all reproduced first at the age of five years. The tree-sized species (A. m, A. rhombifolia, A. acuminata, A. jorullensis, g.) were in general not seen reproducing in the field until they were relatively large. To plants living in an environment with a relatively short grow- ing season and a severe winter (such as that of far northern and subalpine climates) the parallel development of the shrub habit and an earlier reproductive maturity would be advantageous. Structural reduction under such environmental conditions seems to be associated with anatomical immaturity in otherwise adult plants (neoteny), as noted by Forsaith (1920) and Hall (1952). Lh‘i’i frequent1 such groups 0f P13“t field, small plants occasionally found 1 transplanting. Sprs ilfl m subsp. oblongifolia. Both propagating alders, observed with adven abundant seed (frui plants to develop i Steele (i961), judy “USPS 0f A. incan; \ these were not do The fruits 0f Mill in most cas however, the Wings Serv'mg as floatat found that the in water for a perm no Similar Study , assumed that Such W are like Scam, (196 m ental stage of t 96 Alppg frequently occurs in thickets, but it is not known whether such groups of plants always (or ever) represent clones. In the field, small plants of A, serrulata, A, incana subspp. rugosa and tenuifolia, A, viridis subsp. sinuata, and A, oblon ifolia were occasionally found to be interconnected by their roots when dug for transplanting. Sprouting from exposed roots in streams was noted in 59322 incana subsp. tenuifolia, A, viridis subsp. sinuata, and A, oblongifolia. Both layering and cuttings are effective methods of propagating alders, and submerged branches in nature are sometimes observed with adventitious roots. On the other hand, A1233 produces abundant seed (fruits), and one would expect thickets of individual plants to develop in suitable habitats from natural seeding alone. Steele (1961), judging from the amount of variation observed in clumps of A, incana subsp. rugosa and A, serrulata, concluded that these were not clones. The fruits of A1223 are small and usually winged. Their dis- PETSal in most cases is probably by the wind. In some species, however, the wings are much reduced and somewhat corky, apparently serving as floatation mechanisms for water dispersal. McVean (1955) found that the fruits of élEEi glutinosa would float in still water for a period of over 12 months without decomposing. Although no similar study has been made of any of the American species, it is assumed that such essentially wingless-fruited species as A1225 maritima are likewise dispersed by water in most cases. Schalin (1968), using X—ray techniques to determine the develOp- mental stage of the embryo and endosperm in seeds of Alnus incana and 5- 311212938. fo‘ contained incomplett to nearly 1007. of v This system presuma germinate immediate nation in m is of normal variatior tive to low oxygen Many Species n hYbrids have been 0f the New World 3 of naturally hybri and SUbsp, m intergrade morphoT W and subsp. Alaska, an eXtremn as does one in th rnl _ hybrids "normally Intermediate, am and concludes th; this area. No { various Latin Am t he species Stro bet“ Sen W 3 v at. acuminata r C 97 V and A, glutinosa found that approximately half of the viable seed V contained incompletely-developed embryos. Germination was increased to nearly 100% of viable seed by stratification and cold treatment. This system presumably insures that part of a plant's seed crop will germinate immediately while a portion is delayed until later. Germi- nation in Alggg is epigeal and appears to be relatively independent of normal variation in light, temperature, and pH, but it is sensi- tive to low oxygen and moisture levels (McVean, 1953a). Many Species of Alnus hybridize, and numerous putative natural hybrids have been named (see Winkler, 1904; Murai, 1964, 1968). Of the New World species, three pairs of taxa have long been suspected of naturally hybridizing. The ranges of Alnus incana subsp. rugosa and subsp. tenuifolia overlap in north-central Canada, and the two intergrade morphologically in this region. Where Alnus viridis subsp. -—-!--—-—-r-.—— crispa and subsp. sinuata occur sympatrically in central and northern Alaska, an extremely variable hybrid swarm exists (Hultéh, 1944), as does one in the region where Alnus incana subsp. rugosa and A, ser- rulata come in contact. In the latter case, Steele (1961) found that hybrids "normally occur in places that may be regarded as somewhat intermediate, and that have almost invariably been disturbed by man” and concludes that A. serrulata is introgressing into A, incana in this area. No conclusive evidence exists for hybridization among the various Latin American taxa, but the range of variation in some of the species strongly suggests that it does occur. Intermediate forms between Alnus acuminata var. acuminata and var. glabrata, A, acuminata var. acuminata and A._joru11ensis var. jorullensis, and Ag jorullensis var. jorullensis and occur in areas where Because putatii one must conclude tl trinsic in nature. has not been report lated from all the species of the subg hut they do not prc Slecies or infraspe 0r altitudinally ar Aponixis was r loodWorth (1929, 1 $wa the a late Catkins were of pollen release, morphologically nc hle seed. Embryo the pollen Was Sh‘ one to four embryr maturing in each the diploid egg a (1931) showQ‘d the Perfect Pollen an in the New Englat ll. ' ‘ % SUbSP. 98 var. jorullensis and var. firmifolia are common, and these usually occur in areas where hybridization could be expected to take place. Because putative hybridization appears so frequent in the genus, one must conclude that reproductive isolation in Alngs is mainly ex- trinsic in nature. Successful crossing among the three subgenera has not been reported, however. Alng§_maritima is effectively iso- lated from all the other species by its flowering season. Several species of the subgenera Aiggs and Alnobetula grow sympatrically, but they do not produce hybrid offspring. Within each subgenus, the species or infraspecific taxa are all separated either geographically or altitudinally and climatically. Apomixis was described in Alnus serrulata from New England by Woodworth (1929, 1930). Meiosis in the plants studies (called A. rugosa by the author) was found to be extremely abnormal. Pistil- late catkins were found to have no embryo sac development at the time of pollen release, and the pollen was judged to be less than 5% morphologically normal, yet the plants produced an abundance of via- ble seed. Embryo sac develOpment began about three months after the pollen was shed without a reduction of chromosome number. From one to four embryo sacs developed in each ovule, one to five embryos maturing in each embryo sac. The embryos were seen developing from the diploid egg and by nucellar budding. In a later paper, Woodworth (1931) showed that AiflEi serrulata from Virginia produced almost Perfect pollen and concluded that the cytological irregularities seen in the New England populations were the result of hybridization with A . _, incana subsp. rugosa. Pollen from spe each taxon was exam? abnormalities were 1 part of the range 0 that aponixis is ra the genus. Further needed for a better 99 Pollen from specimens selected from throughout the range of each taxon was examined (as described above), but such morphological abnormalities were not found in any other species or in any other part of the range of élflfli serrulata. From these results it appears that apomixis is rare or absent in most of the American species of the genus. Further work, especially on a population basis, is needed for a better understanding of this problem. All of the Ame information is avai tunately, there are our for any of. the In 1934, Hans the Betulaceae is . Wand 3 = 14 found a diploid m, thS\Uta var. m which may be consc A~'hir5\“ta, the in some number in Q Four levels is by far the mos Ported by Chiba ( several European 2t= 56. W i Miles or both : l- Su % am 2!: 56 (Gram Et 3» CHROMOS OME NUMBERS All of the American species of w about which chromosome information is available have numbers of 2_n_ = 28 (Table 3). Unfor- tunately, there are no data for A_1nu_s_ rhombifolia, A. oblongifolia, nor for any of the Latin American taxa. In 1934, Wanscher predicted that the basic chromosome number in the Betulaceae is _x = 7, based on numbers of x = 8 in Carpinus and Ostrya and Pi = 14 or higher in the other genera. In 1962, Chiba found a diploid number of 14 in root tip cells of Japanese fl hirsuta var. microphylla (A. inokumae Murai and Kusaka), a taxon which may be conspecific with A, incana. In the typical variety of A. hirsuta, the number was found to be 2_1_'1_ = 28. The basic chromo— some number in A1233 is therefore taken to be 5 = 7. Four levels of polyploidy occur in A323, although 21 = 28 is by far the most common condition. In addition to the number re- ported by Chiba (22 = 14) and the many species having 22 = 28, several European and Asian species have numbers of 23 = 42 and 23 = 56. A_lfl1_s_ glutinosa has been reported with somatic chromosome numbers of both 22 = 28 and 2_r_1_ = 56 (Woodworth, 1929, 1931), while 39;.- subcordata and A. japonica may have either 23 = 28, 2_r_1_ = 42, or 2}; = 56 (Gram 3311;” 1941; Jaretzky, 1930; Woodworth, 1929, 1931). Alnus cordata shows either 23 = 28 or 23 = 42 (Jaretzky, 1930. 100 Table 3. Chromosor and their Soecies l. incana subsp. _i_ l. incana subSp. r - _ _ l' lncana subsp. _l l mar' ‘ .' ltlma l1. rubra \ l- serr ~ \ulaa l. ' - . ‘ % SubSp, Al I'. ‘WSUbSp l. '-. ‘WSUbSF \ 101 Table 3. Chromosome numbers reported for American species of Alnus and their Eurasian vicariants. Species Number (2n) Reference A. incana subsp. incana A. incana subsp. rugosa A. maritima A. rubra A. serrulata —*_ A. viridis subsp. crispa A. viridis subsp. sinuata ——-* A. Viridis subsp. viridis A. incana subsp. tenuifolia 28 28 28 28 28 28 28 28 28 28 28 28 Gram g£_gl., 1941 L6ve, 1954 Wetzel, 1928 Love, 1954 Woodworth, 1929 Woodworth, 1931 Gram E£.E£" 1941 Wetzel, 1929 Woodworth, 1929 Woodworth, 1931 Calder and Taylor, 1968 Gram 33.2l" 1941 Jaretzky, 1930 Wetzel, 1927 Wetzel, 1928 Woodworth, 1929 Woodworth, 1930 Woodworth, 1931 L6ve, 1954 L6ve and vae, 1965 L6ve and Love, 1966 Pouques, 1949 Woodworth, 1929 Woodworth, 1931 Calder and Taylor, 1968 L6ve, 1954 Contandriopoulos, 1964 Jaretzky, 1930 Pouques, 1949 Wetzel, 1928 Wetzel, 1929 In addition, Kodam: from root nodule 1; whether other part Jaretzky (193 irregular and the cones (M w having 23 = 56 chr the MM is wel‘. in which 21 = 28 5' m subSp. Q 1931) likewise in It seems cle and 23: 56 Tepre mosomes may have ZR = 56 types. 102 Alnus orientalis has been found with 23 = 42 (Gram 233;” 1941). In addition, Kodama (1967) reported a chromosome number of 23 = 112 from root nodule tissue of Alnus firma in Japan, but it is not known whether other parts of the plant had this number. Jaretzky (1930) and Gram _e_1_:__a_l_.,(1941) report that meiosis is irregular and the pollen quality poor in the species having 42 chromo- somes (Alnus cordata, A. subcordata, and A. orientalis). In plants having 2_r_1_ = 56 chromosomes, meiosis is nearly regular and almost all. the pollen is well-formed (Gram 53$ng 1941). All reports for species in which 22. = 28 (except those involving putative hybrids between A. incana subSp. rugosa and A. serrulata by Woodworth, 1929, 1930, 1931) likewise indicate a normal meiosis. It seems clear that the alders with chromosome numbers of 23 = 28 and 23 = 56 represent a simple polyploid series. Those with 42 chro- mosomes may have resulted from hybridization between 22 = 28 and _2_r_1_‘= 56 types. "‘--«t...." ‘ :dfi. ‘5‘ M ._,..p 7' The analysis r electronic compute The field of numer it gained widespre Sneath's W come one of the m taxonomic researcl viewpoints: the ; canT-generated c fell but should r l°ssible (Sokal, 5°“ and Crovello, school, in which foal (Cain and H; Farris, 1967; We; includes StUdies latz and Torres Tailor (1966), u billiikinson et studies have prj and Princip al C( NUMERICAL TAXONOMIC STUDIES The analysis of large amounts of comparative data by the use of electronic computers can be a very powerful technique in systematics. The field of numerical taxonomy originated during the 1950's, but it gained widespread attention only with the publication of Sokal and Sneath's Principles 2f_Numerical Taxonomy in 1963. It has since be- come one of the most active (and sometimes controversial) aspects of taxonomic research. Today the field is represented by two principal viewpoints: the phenetic school, in which it is held that numeri- cally-generated classifications should make no assumptions of phylo- geny but should reflect relationships based on as many characters as possible (Sokal, 1962; Sokal and Sneath, 1963; Crovello, 1970; Jack- son and Crovello, 1971; Sneath and Sokal, 1973) and the phyletic school, in which quantitative evolutionary schemes are the primary goal (Cain and Harrison, 1960; Camin and Sokal, 1965; Mayr, 1965; Farris, 1967; Wagner, 1969). Recent work utilizing phenetic methods includes studies of 93133 by Morishima and Oka (1960), Zinnia by Katz and Torres (1965), Zlgli by Pettet (1965), Lithophragma by Taylor (1966), Haplopappus by Jackson and Crovello (1971), 23222 by Wilkinson et a1. (1971), and Platanus by Hsiao (1973). These Studies have primarily utilized the techniques of cluster analysis and principal components analysis (factor analysis), which are 103 reviewed by Sokal ( (1968), Jackson ant Phyletic methods, 1‘ been used in syster (1957), Clefl by Scora (l966), anur Solbrig (1970), an study of m, M0 for his conclusion the same data. Several works Km and Torres, '. ThQDEtic numerica lrtsumably indica studies, the phen lnolrn relatioushi similar. The aut “Win can be , informatio“, inc] lldgement in some Another nap, will very likely use of computers quantitatimy‘C lccially “new“ llorse 3 a1. (19 TIIIIIIIIIIIIIIII________________________________—_—7 ‘ “7""""‘?' 104 reviewed by Sokal (1961, 1962), Sokal and Sneath (1963), Crovello (1968), Jackson and Crovello (1971), and Sneath and Sokal (1973). Phyletic methods, in particular those of Wagner (1962, 1969), have been used in systematic studies of the Hippocastanaceae by Hardin (1957), Cleome by Iltis (1959), Dicentra by Stern (1961), Monarda by Scora (1966), anurans by Kluge and Farris (1969), Gutierrezia by Solbrig (1970), and Catostomus by Smith and Koehn (1971). In his study of 95153, Morishima (1969), seeking a more comprehensive basis for his conclusions, undertook both phenetic and phyletic analyses of the same data. Several workers (Soria and Heiser, 1961; Heiser £2 31., 1965; Katz and Torres, 1965; Rhoades SE 31., 1968) have attempted to ”test” phenetic numerical methods by using taxa for which cytological data presumably indicate true phylogenetic relationships. In each of these studies, the phenetic classification was found to disagree with the known relationships in some respect, although in general they were similar. The authors concluded in each case that while numerical taxonomy can be useful, it should be used in conjunction with other information, including a certain amount of subjective taxonomic judgement in some cases. Another aspect of numerical taxonomy, in a broad sense, which will very likely become increasingly important in the future, is the use of computers for the storage and retrieval of large amounts of quantitatively-coded systematic information (cf. Morse, l974a). Es- PSCially noteworthy is the system of Morse (1968, 1971, l97fib) and Morse g; al (1971), which includes a generalized data format and aseries of programs tasks, including spe the determination of in the present components analyses thellagner method ( prepare a taxonomic accompanying charac land 5). Several for key-constructi< were used in organ though they were n descriptions in th “ethic“ fl Mater Cluster anal sistem developed ill/155 °°mputer calculated as a ( Sokal, 1961), em case the Correla by the ““WEighte filial CommentE (1957) 0“ the C] i “Gnome data 1 105 a series of programs for accomplishing a variety of taxonomic tasks, including specimen identification, key—construction, and the determination of diagnostic characters. In the present work, phenetic studies using cluster and principal components analyses were complemented with a phyletic study based on the Wagner method (1962, 1969). In addition, the data were used to prepare a taxonomic data matrix. The data matrix, along with an accompanying character list, is included in the appendix (Printouts 4 and 5). Several of the Morse computer routines, including those for key-construction and testing the diagnostic power of characters, were used in organizing the data for other parts of this study, al- though they were not used directly in the construction of the key or descriptions in the taxonomic treatment. Phenetic Studies Methods 23g_Materials: Cluster analyses were performed with programs of the NT-SYS system developed by F. James Rohlf and associates using the IBM— 370/155 computer at the University of Notre Dame. Similarity was calculated as a distance coefficient (Average Taxonomic Distance of Sokal, 1961), except where the characters were clustered, in which case the correlation coefficient was used. Clustering was performed by the unweighted pair-group method using arithmetic averages. Prin- cipal components analyses were made using program FACTOR of Veldman (1967) on the CDC—3600 computer at Michigan State University. The taxonomic data matrix was processed using the Morse MSU-SYS system with the General 13' with a remote Tele Taxa for none herbarium methods; continuous pattern among groups. The smallest discernal explained below, I plied to most of might not be clea 18)," which occur vhich differs frc pubescent leaves'1 Which Occurs in ( mfirican 5. “A“ fofi (GUatemala ters from typica taxa represent j fit and subs} and p. % with the New No “1%) and 5 item the r “Siting, these features (Such lands) and the 106 with the General Electric Mark II Network GE—635 computer accessed with a remote Teletype terminal at Michigan State University. Taxa for numerical analysis were first selected by conventional herbarium methods; that is, specimens were sorted into sets showing continuous patterns of variation within and discontinuous patterns among groups. The taxa thus chosen were selected to provide the smallest discernable units which appeared to be natural; later, as explained below, many of these groups were combined. The names ap- plied to most of the taxa are self-explanatory, but several which might not be clear include ”Alnus acuminata var. acuminata (Guatema- 1a),” which occurs in Guatemala, Honduras, and southern Mexico, and which differs from Mexican var. acuminata by its larger, darker, more pubescent leaves; '23. acuminata var. acuminata (Central America)," which occurs in Costa Rica and Panama, and which differs from South American 5, acuminata in leaf shape; and ”A, jorullensis var. £2323: £2133 (Guatemala)," which occurs in Guatemala and El Salvador and dif- fers from typical var. firmifolia in leaf shape. In three caseS, the taxa represent intermediates between other species: A, viridis subsp. crispa and subsp. sinuata; A, incana subsp. rugosa and A, serrulata; and A, acuminata Var. acuminata and Var. glabrata. For comparison with the New World subspecies, Eurasian A, viridis (A3 viridis subsp. “ viridis) and A. incana (A. incana subsp. incana) were also included. Next the range of each taxon was divided into a number of regions, these usually corresponding to obvious physiographic features (such as the Atlantic Coastal Plain, the Appalachian High- lands, and the Ozark Plateau in Alnus serrulata) or to more or less discrete geographic of a taxon was cons the various analyse selected (Table 4) A tentative l ture, and addition mens from all taxa measured on each c in all, 488 vegetg for the collectiov fOY such characte: ore or degrees. alogicaily ascen acters were found to measure accurg ters for analeis were Obtained an isles) or are 118 chISter ana data (150 Charac characters, and data set of 70 C so t, was anaIYZe Character 8 them itatuI-ES . ll d8 llSQd t0 det 107 discrete geographical parts of the range itself. Each such segment of a taxon was considered an Operational Taxonomic Unit (OTU) in the various analyses. In all, 70 OTU's representing 30 taxa were selected (Table 4). A tentative list of characters was abstracted from the litera- ture, and additional features were added after examination of speci- mens from all taxa. The 319 characters thus obtained were then measured on each of 5 to 25 randomly—selected specimens for each OTU. In all, 488 vegetative specimens and 459 flowering specimens were used for the collection of data. The measurements were continuous in nature for such characters as length, and multistate for features such as col— ors or degrees. For multistate characters, the states were scaled in a logically ascending or descending manner. After coding, many char— acters were found to be invariant, excessively variable, or difficult to measure accurately, and these were eliminated, leaving 150 charac- ters for analysis (Table 20, appendix). Specimens from which data were obtained are cited in the taxonomic treatment (marked with aster- isks) or are listed in Table 21 (appendix) in the case of Eurasian taxa. Cluster analyses of the OTU's were carried out using all of the data (150 characters), the 96 vegetative characters, the 30 floral characters, and the 24 fruit characters. In addition, an abridged data set of 70 characters, selected randomly from the complete data set, was analyzed for comparison. A cluster analysis of the 150 characters themselves was made to determine relationships among the features. A principal components analysis of the complete data set was used to determine relationships among the characters and to Table 4. Operatiot 0TH No. l 2 11 ll 13 id ana 1y 5 e s Acron' ACUCA ACUGU ACUGU ACUME ACUME ACUME ACUME ACUSA ACUSA ACUSA GLABR GLABF GLABF ACUG] ACUGl CAST, FERR‘ FERR FERP FERP 108 Table 4. Operational taxonomic units (OTU's) used in phenetic analyses of Alnus. OTU No. Acronym Tag 1 ACUCA A. acuminata var. acuminata (Central America) 2 ACUGU (N) A. acuminata var. acuminata (Guatemala, north) 3 ACUGU (S) A. acuminata var. acuminata (Guatemala, south) 4 ACUME (N) A. acuminata var. acuminata (Mexico, north) 5 ACUME (E) A. acuminata var. acuminata (Mexico, east) 6 ACUME (S) A. acuminata var. acuminata (Mexico, south 7 ACUME (W) A. acuminata var. acuminata (Mexico, west) 8 ACUSA (N) A. acuminata var. acuminata (South America, north) 9 ACUSA (C) A. acuminata var. acuminata (South America, central) 10 ACUSA (S) A. acuminata var. acuminata (South America, south 11 GLABR (N) A. acuminata var. glabrata (north) 12 GLABR (C) A. acuminata var. glabrata (central) 13 GLABR (S) A. acuminata var. glabrata (south) 14 ACUGL (W) A. acuminata var. acuminata - var. glabrata intermediate (west) 15 ACUGL (E) A. acuminata var. acuminata - var. — glabrata intermediate (east) 16 CASTA A. _:astaneifolia 17 FERRU (N) A. ferrug._nea (north) l8 FERRU (C) A. _ferrug:_r_1_e'A (central) 19 FERRU (S) A. ferrug._nea (south) 20 FERPI (N) A. _ferruginea (pilose leaves, north) 21 FERPI (C) A. ferrugifla (pilose leaves, central) 22 FERPI (S) A. ferru inea (pilose leaves, south) 23 FEROB A. ferruginea (obovate leaves) 24 INCAN (N) A. incana subsp. incana (north) 25 INCAN (S) A. incana subsp. incana (south) 26 INCAN (E) A. incana subsp. incana (east) 27 RUGOS (N) A. incana subsp. rugosa (north) 28 RUGOS (S) A. incana subsp. rugosa (south) 29 RUGOS (W) A. incana subsp. rugosa_ (west) 30 TENUI (N) A. incana subsp. tenuifolia (north) 31 TENUI (C) A. incana subsp. tenuifolia (central) 32 TENUI (S) A. incana subsp. tenuifolia (south) 33 RGSER (E) A. incana subsp. rugosa - A. serrulata intermediate (east) Table 1+ (Continued OTU No. it U'Ul U'KJ'LJI-PJ—‘J—‘bP-l—‘J—‘D mm oomgo‘waN @MPWNH axocnox U1 a‘ulu' DWNP—‘OKDWN Acror RGSER FIRMI FIRMI FIRMI FIRGU JORUL J ORUL JORUl JORUI MARI] MARI] OBL01 OBLOl OBLOI OVALi PRINv PRIN runs RHOM RHOM RUBR RUBR RUBR SERE SERi SER} SERI CR1: CR1: CR1: CR1 SIN SIN CRS V11 V11 109 Table 4 (Continued). OTU No. Acronym Taxon 34 RGSER (W) A. incana subsp. ru osa - A. serrulata _ intermediate (west) 35 FIRMI (C) A. var. (central) 36 FIRMI (S) A. var. (south) 37 FIRMI (N) A. var. (north) 38 FIRGU A. var. (Guatemala) 39 JORUL (s) A. var. (south) 40 JORUL (C) A. var. (central) 41 JORUL (N) A. var. (north) 42 JORUL (w) A. var. (west) 43 MARIT (D) A. laware 44 MARIT (o) A- (Oklahoma) 45 OBLON (r) A. (typical) > 46 OBLON (R) A. (leaf tips rounded) 47 OBLON (a) A- (leaf tips acute) 48 OVALI A. l 49 PRING (N) 9: north) 50 PRING (C) A. (central) 51 PRING (s) A. (south) 52 RHOMB (N) A. (north) 53 RHOMB (s) A- (south) 54 RUBRA (N) A~ ) 55 RUBRA (c) A. (central) 56 RUBRA (s) A. (south) 57 SERRU (N) A- (north) 58 SERRU (c) A- (central) 59 SERRU (s) A- (south) 6O SERRU (W) A. (west) 61 CRISP (E) A. subsp. EEPE (east) 62 CRISP (C) A. subsp. ESE (central) 63 CRISP (W) A. subsp. gig (west) 64 CRISP (P) A. subsp. Efifl (pubescent extreme) 65 SINUA (N) A. subsp. sinuata (north) 66 SINUA (S) A. subsp. sinuata (south) 67 CRSIN (N) A. subsp. crispa - subsp. intermediate (north) 68 CRSIN (S) A. subsp. crispa - subsp. intermediate (south) 69 VIRID (E) A- subsp. viridis (east) 70 VIRID (W) A. subsp. viridis (west) determine clusteri rather than origir From the priv characters in the to be correlated nate redundancy ( were eliminated, and Sneath, 1963: it represented ti should have resu' reseInblance . u. 31:8. The pheno gr 70 on“ 8 (reduce character) data onVs PIOduCed . CUSSed abOVe, a 28 taXa fol. Cla COmpar 180:] and 70'Characte number of char though differs the 115 character an cl 8 Wine gr am ' 110 determine clustering patterns based on the principal components rather than original characters. From the principal components and cluster analyses of the 150 characters in the complete data set, many characters were found to be correlated at a very high level (Table 5). In order to elimi- nate redundancy (332:, use only ”unit characters"), 35 such traits were eliminated, resulting in a 115-character data set (cf. Sokal and Sneath, 1963). This data set was also analyzed, and, because it represented the largest number of non-superfluous characters, it should have resulted in the most accurate assessment of overall resemblance. Results: The phenogram in Figure 5 represents the cluster analysis of 70 OTU's (reduced to 28 taxa for clarity) for the complete (150- character) data set. Figure 6 shows the complete phenogram of 70 OTU's produced in the analysis of the 115-character data set dis- cussed above, and Figure 7 represents the same phenogram reduced to 28 taxa for clarity. Comparison of the phenograms from analysis of the 150-, 115-, and 70-character data sets showed these to vary somewhat with the number of characters used, as also found by Crovello (1969). Al— though differences between parts of the 70—character phenogram and the 115-character phenogram were considerable, those between the 115- character and ISO-character phenograms were much less striking. The phenogram based on analysis of only the vegetative characters hbh 5. Character found to hug 1 18 18 2 10 10 3 10 1C l a t 5 15 U 6 7 l l l l d l l Table 5. Group 1 10 11 13 Characters in the complete morphological data set for Alnus found to be highly correlated. 111 Characters 183 185 107 108 102 103 41 62 156 157 23 24 184 188 189 190 198 199 201 202 27 28 48 49 81 83 Length of mature cone Ratio of length to width of mature cone Color of leaf above Color of leaf below Length of leaf Length of leaf from base to widest point Density of glands on stem Density of glands on bud stalk Grouping of pistillate catkins Relation of staminate to pistillate catkins Length of lenticel Width of lenticel Diameter of mature cone Length of cone scale Width of cone scale at widest point Width of cone scale at narrowest point Length of fruit Diameter of fruit Number of bud scales Equality of size of bud scales Type of bud scales Relative position of bud scales Length of fruit wing Length of fruit wing from base to widest point Height of leaf scar Width of leaf scar Length of bud body Diameter of bud body Shape of leaf base Angle of leaf base Curvature of sides of leaf base Table 5 (Continued: Group 14 8. 10' 15 19 19 16 16 17 ll 2] 2] l9 1 l l l 1 20 1 l 21 22 23 24 ’7— .~ nasal...“ H» a 112 Table 5 (Continued). Group Characters 14 85 Compression of leaf apex 104 Ratio of leaf length to length from base to widest point 15 195 Shape of terminal lobe of cone scale 196 Extension of terminal lobe of cone scale 16 169 Length of staminate floral bract 170 Width of staminate floral bract 17 212 Length of perianth part 213 Width of perianth part 18 39 Length of pubescence on stem 58 Length of pubescence on bud stalk 19 38 Density of pubescence on stem 57 Density of pubescence on bud stalk 114 Density of pubescence on leaf blade, above 116 Density of pubescence on leaf blade, below 121 Density of pubescence on leaf veins, below 124 Density of pubescence in leaf vein axils, below 148 Density of pubescence on petiole 20 117 Length of pubescence on leaf blade, above 125 Length of pubescence in leaf vein axils, below 21 42 Size of glands on stem 63 Size of glands on bud stalk 22 99 Shape of leaf apex lOO Angle made by leaf apex 101 Curvature of sides of leaf apex 23 119 Regularity of leaf serrations 120 Regularity of leaf teeth or lobes 24 75 Length of pubescence on stipule 115 Length of pubescence on leaf blade, above 113 Figure 5. Phenogram showing relationships among 28 taxa of Alnus based on cluster analysis of 70 OTU's and 150 morphological characters. ////////'/E/Hf/d / / / / l / l l .‘3 .’ IA— . (J _— - 1‘ 0.2 L] MRIT CASTA F “160 JORUL FIRMI ACUCA GLABR OVAL! FEROB FERPI FERRU PRING ACUGL ACUGU ACUME ACUSA RUBRA SERRU IENIJI INCAN RGSER RUGOS RHDMB OBLDN SINUA CRS IN VIRID CRISP 0.0 Figure 5. Figure 6. 115 g g s. 115 morphological character 70 OTU's and mm 3"— m m m h We Wt) \ I ”(ll \ M 915 Wm Wm Mn mmEEE;:]‘ dd :4 “Elm \ 'fitm \ Wm“-__ “(It \ Mu) \ m"llld \ all“) \ "all h fim:::::: m it E I 9 I Q N 116 4,? Figure 6. RGSER (E) RUGOS (H) k euros csr' euros (to (3) W Mmcm-——— onion (A) onion m sum (3) SINUA cu) cram (s) CRSIN (u) '— vmm (u) " mm m CRISP (P) _L CRISP (u) .2 CMSP (c) \ CRISP (E) ‘— l l I l l “ a n I g _ N l r l l a N 1'; H H to 09 H H a N a, o 6 Q a o Figure 7. 117 Phenogram showing relationships among 28 taxa of Alnus based on cluster analysis of 70 OTU's and 115 morphological characters. MRIT (ASIA FIRGII JORUI. FIRHI ACUCA GLABR FEROB FERPI FERRU PRIIE AEIIGL ACUI‘E ACUSA RUBRA SERRU TENUI I "CAN RGSER RUGOS OBLON SINUA CRSIN VIRlIJ CRISP Figure 7. was very simila more so, in fac tus. Since tl while the flora this many, it : the complete dc ngmbme the one using data or the on gram based on vegetative-chg Differences be gram in Which involvmg OTU When ch data Phenogra oneseeS seve folia are mow moved closer Clusters wi t1 remote from I 1% The fru Var ‘ an, and 115 clusters Clo Incdna’ whi] 119 was very similar to the one involving the complete data set, much more so, in fact, than were those based on floral or fruit charac- ters. Since the vegetative-character phenogram represented 96 traits while the floral- and fruit-character analyses used less than a third this many, it is clear that the number of vegetative characters in the complete data set weights it strongly in that direction. The phenogram based on analysis of the floral characters was more like the one using vegetative features than either the one based on fruit data or the one in which the complete data set was used. The pheno- gram based on fruit characters was about equally similar to the vegetative-character, floral-character, and total-data results. Differences between the vegetative—character phenogram and the pheno- gram in which all of the data were used were rather minor, mostly involving OTU's which are parts of other taxa. When comparing the floral-character phenogram with the complete- data phenogram and the 115-character data set phenogram, however, one sees several major differences. A122§_oblongifolia and A, rhombi— folia are moved much farther from the A, incana group. A, rubra is o a a l ' moved closer to A. acuminata var. acuminata (Mex1co), which, in turn, clusters with var. glabrata. £1323 ferruginea becomes slightly more remote from A, acuminata var. acuminata (Mexico), and A, jorullensis var. jorullensis is widely separated from Var. firmifolia. The fruit-character data set phenogram differs from the complete- data and 115-character data set phenograms in that élEEi serrulata clusters closer to A, incana subsp. rugosa and further from subsp. incana, while 5, incana subsp. tenuifolia becomes much more remote from the rest of allied only to A acuminata (South in Central Ameri var. M mc M becon the h. acuminata uala) moves awaj From the a 0f the llS-char matim Provided Pittures of rel 0f Clusters wit @w While the phenc other Cases in 81°“ include M are cl with subsp. fl subsp. f . % are Clo: subsp. . IHCana \ and the fact ' in, . w Var . v er. acwininat \ 11a Q Var, aCu \ 120 from the rest of this group. Alnus rubra becomes more remote, being allied only to A, jorullensis var. jorullensis. Alnus acuminata var. acuminata (South America) moves closer to pOpulations of this species in Central America than in the complete data set tree, and A, acuminata var. glabrata moves closer to A, ferruginea. Alnus_iorullensis var. firmifolia becomes more distant from var. jorullensis and closer to the A, acuminata cluster, while A, jorullensis var. firmifolia (Guate- mala) moves away from all the other Latin American taxa. From the average distance matrix for the taxa, based on analysis of the 115-character data set, (Table 6), one can supplement the infor— mation provided by'the phenograms, which give somewhat more distorted pictures of relationships because they involve average similarities of clusters with OTU's. For example, the distance matrix shows that Alnu§_maritima is closer to A, rhombifolia than to any other taxon, while the phenogram (Figure 7) does not indicate this relationship. Other cases in which the relationship is not clear from the phenogram alone include the fact that both A, viridis subsp. crispa and subSp. sinuata are closer to subSp. viridis than they are to each other, but with subSp. viridis being somewhat closer to subsp. crispa than to subSp. sinuata; the fact that A, incana subsp. rugosa and subsp. 5222;, Eglig_are closer to each other than they are to‘subsp. incana, but with subsp. incana being closer to subsp. rugosa than to subsp. tenuifolia; and the fact that A. jorullensis var. jorullensis is closest to A, 'or- ullensis var. firmifolia, while the latter is closest to A, acuminata var. acuminata (Mexico). The closest neighbor of A, rubra is A, acumi- nata var. acuminata (Mexico), and that of A, serrulata is A, incana ,.-s_-.-x.¢,_‘.,. ---. . -.-‘ ,:>’. 1 1.13 —- <:11 51]: an <:‘t:(3 1: 1> ea :5 <2 <1 <> {1 ea {1 5) 1137’s; 1. £3 <> 1? t11‘1c2 1? l: slzcea <9 1? 18.1.r1L125 between each pair 0 (1 451 I; .51 :3 (52 l: - .aa.\r £2 27's; 2; e: (1 j. a; I: 53 11 <: (a (5 . Cl‘ga t> 17¢: n1c>2‘[911<) 1.C)ég i.<:za 1 Kean Dill. VIRID TEEDI SIXUA SEEIU RUGOS RUBRA [Boll RGSEI PlIflG WHI OILON HARIT JORUL INCAN GLABR FIRXI FIlGU FERRU FERFI FERoa cwsiw CRISP (ASIA ACUSA 1th: ACUGU ACUGL Acvcw 1.50 1. 36 ; as.a nu mo.a mH.H oo.a Hn.H sa.a oh.a ms.a as.H aw.a sn.H s~.~ Hm.a H0.H ma.~ ma.a Nm.a ms.H ms.a sa.a s~.H mo.a mo.~ ma.H Ns.H ma.H mm.H om.a aqus ss.H a- ~0.H s~.H oo.H os.a as.a oH.H m~.a ss.a m~.a sm.a as.H ao.a on.a as.a ma.a mm.H sm.fl Nn.H ms.a Ns.H ma.H Ns.a o~.H ~n.a wm.a as.H Hazms so.a I- oa.a as.a as.a AN.H nn.a os.a mo.a mm.a am.~ N¢.H as.a ms.a sw.H sa.a so.a am.a ne.a as.o OH.H ma.“ sa.a an.H so.H as.a sm.H <=sz os.H u: so.“ as.a oN.a aa.o sm.H om.a s~.a sm.~ ~n.a «H.H as.H mm.H an.H s~.a sm.a on.H as.a as.H Ha.a HQ.H o~.H so.a an.a sm.a aaamm nn.a u- os.H Hm.a om.o 0H.H m~.H m~.H ma.a sn.a mm.o o~.H ~m.a mn.a s~.a s~.H sm.a as., Hm.a ma.a m~.H OH.H H~.H mn.H sn.a moons ~s.a we ms.H mn.a o~.a nn.H om.H mo.~ os.H an.” RH.H a~.a mm.a m~.H ma.” o~.H om.H as.a aa.a a~.H oa.H wH.H n~.H m~.H «mane an.a .. nn.a as.a mn.a ~o.a sm.a an.H an.a os.H om.a He.a ms.a ms.” om.a Hm.H ma.a H¢.H as.a m~,a mm.a on.H ss.a nzomm «n.a n- ma.a m~.H -.H co.” s~.a sm.o ¢~.a a~.H as.a aa.a sa.a sm.a ~s.H as.a sa.a a~.H ma.a n~.H mm.H mm.a memos a~.a s- am.o s~.H so.~ m~.H ~H.H mo.H sH.H ms.a om.o am.o NH.H mo.a ms.H mm.a oa.~ am.o Nm.o oa.o mn.H ozHaa sm.a In an.a so.~ NH.H an.a mo.a H~.H om.a aa.a mo.H mo.H aa.a as.a mm.H mN.H oo.a sH.H MH.H 0N.H Haa>o an.“ a- am.a sa.a H~.a NH.H mm.a mm.H sm.a Nn.H wn.H mm.a mm.a Ha.a H¢.H mo.H oN.H nm.a ss.a zoano so.~ I- oa.H mm.H H¢.H mm.a ma.~ ao.~ wo.~ HH.~ on.~ 0N.~ m~.~ HH.~ mm.H mo.~ mo.~ s~.~ HHa< .0 manmfi subsp. {Effig- Also provided from all the other noteness in the ta: specialized and th A, 1962). Relationships Iii-OTU X 115-char; these taxa. Alfl Siterately from a Cluster together make “P two largc mtomplex a‘ the other cluste Among the 1 (South Almeriea) lated to 5- am Mud“ in the Rafi (Guatemala IIIE ' d1ate") 5t 0 with this taXo wit - h It. M related to thj above. In ger 8th as A. S We line a 122 subsp. rugosa. Also provided in Table 6 is the average distance of each taxon from all the other taxa. This value serves as an indication of re- moteness in the taxa, the more remote taxa tending to be the most specialized and the least remote the most generalized (Ellison EE El" 1962). Relationships among the taxa, based on cluster analysis of the 70-OTU X 115-character data set are similar to traditional systems for these taxa. Alnus maritima is the most distant species, clustering separately from all the other taxa. The subspecies of A, viridis cluster together as the second most remote group. The remaining taxa make up two large groups, the first including the A, incana - A, E237 rulata complex and the A, oblongifolia — A, rhombifolia group, and the other cluster the Latin American species and A, rubra. Among the Latin American taxa, Alnus acuminata var. acuminata (South America) and A, acuminata var. glabrata are very closely re- lated to A. acuminata var. acuminata (Mexico). Several variants included in the analysis as separate taxa ("A, acuminata var. acumi- nata (Guatemala)”, "A. acuminata var. acuminata - var. glabrata inter- mediate", A, ovalifolia, and A, pringlei) are so closely clustered with this taxon that they should undoubtedly be considered conspecific With it. AAEEA ferruginea and its variants are also very closely related to this species, though not quite so closely as the taxa listed above. In general, based on the pattern seen in well-marked species Such as A, serrulata, A, rubra, A, rhombifolia, and A, viridis, the Phenom line for species should lie somewhere between the 1.0 and 1.2 distance coeffici Within spec: vided informatior of the taxa invo migration, isola present species cussed in relati Principal < sulted in the e: (ranging from 2 E1Eenvalues for 10'0) 7-7, and and 5'9 Pel‘cent correlations (a listed in Table Smith and Koehr associated Wit] Ponent, but in Some of the Ch adaptations to horsemen of eital number a listed fOr pr: correlati°ns ‘ terminEd from The Corr 123 distance coefficient lines on the phenograms in Figures 6 and 7. Within species, the distances between pairs of OTU's pro- vided information that may be useful in determining the histories of the taxa involved, showing patterns which can be explained by migration, isolation, divergence, and introgression leading to the present species structures. Several species complexes will be dis— cussed in relation to these data in a later section. Principal components analysis of the 115-character data set re- sulted in the extraction of 19 roots having eigenvalues exceeding 1.0 (ranging from 21.8 to 1.1) and representing 91.6 percent of the trace. Eigenvalues for the first five principal components were 21.8, 12.3, 10.0, 7.7, and 6.1, respectively, representing 24.4, 12.1, 9.8, 7.5, and 5.9 percent of the trace, respectively. Characters having high correlations (at least i 0.600) with each principal component are listed in Table 7. Several authors (Jackson and Crovello, 1971; Smith and Koehn, 1971) have attempted to recognize ”adaptive complexes" associated with the characters correlated with each principal com- ponent, but in the case of A1235, few such complexes are recognizable. Some of the characters of principal component I appear to represent adaptations to a cold environment (number of bud scaleS, leaf size, a protection of the catkins during the winter), but at least an equal number show no obvious adaptive pattern, as do the characters listed for principal components II through IX. Taxa having strong correlations with each of the first 19 principal components, as de- termined from principal axis factor scores, are listed in Table 8. The correlation of each OTU with principal components I and II, Table 7. Morphological c the first nine Principal Component j 1 Shape of hi Length of 1 Ratio of b1 Number of ' Equality 0 Ratio of 3 Degree of Number of Depth of 1 Length of Length of Color of ] Color of ] Density 01 DeBree of Degree of 'Number of Length of Diameter . Ratio of Shape of Length of wldth of Number of Time of f 11 Length 0, Degree 01 SlZe of 5 Length 01 Length 0: Grouping LEngth 0. Time of . H1 DenSity Size of TeXture w Regulari atio of PrOminer Length c IX \~\--~.‘£¥EEE::E 124 Table 7. Morphological characters in Alnus highly correlated with the first nine principal components. Principal Component Charactgr Correlatigfl I Shape of bud tip +.7l4 Length of bud body —.649 Ratio of bud body to stalk length -.780 Number of bud scales -.899 Equality of bud scale size —.903 Ratio of stipule length to width +.606 Degree of compression of leaf base +.612 Number of leaf serrations per cm -.893 Depth of leaf serrations -.729 Length of leaf +.641 Length of leaf from base to widest point +.669 Color of leaf above +.600 Color of leaf below +.629 Density of glands on bud stalk +.79l Degree of development of cross veins +.791 Degree of shoot differentiation -.794 ’Number of nodes on one year's growth +.651 Length of mature cone peduncle -.718 Diameter of mature cone peduncle +.812 Ratio of cone scale length to width -.642 Shape of cone scale tip -.623 Length of fruit wing from base to widest point -.680 Width of fruit wing -.755 Number of male catkins per cluster +.692 Time of flowering +.672 11 Length of pubescence on veins below +.695 Degree of compression of leaf apex +.600 Size of glands on leaf surface below +.733 Length of pubescence on petiole +.663 Length of cone scale -.663 Grouping of pistillate catkins +.677 Length of staminate catkin +.651 Time of pistillate catkin production -,677 III Density of pubescence on petiole -.656 Size of glands on tips of leaf teeth -_633 Texture of fruit wing -.672 IV Regularity of leaf teeth or lobes +.627 Ratio of blade to petiole length +.623 Prominence of veins on leaf below +.677 Length of persistent style on fruit +.606 IX Shape of apex of perianth part +.629 Table 8. Taxa of m with each of morphologies Principal Clem A I a. _r' 11 A TEE V A fl VI A. BE. VII A. £12 a- m VIII A. cas IX A E X A E XVI A E! XVII A .11.] XVIII A° 12‘ XIX A in 125 Table 8. Taxa of Alnus having correlations greater than a 0.500 with each of the first 19 principal components using morphological characters. Principal Correlation Component Coefficient I ‘A. viridis subsp. crispa -.522 II A, maritima -.713 V A. oblongifolia -.510 VI A. 511—133 -.525 VII A, rhombifolia +.548 A. PEP—IE. +.538 VIII A, castaneifolia +.562 IX A, acuminata var. acuminata -.601 X A, incana subsp. incana -.552 XVI A, ferruginea +.511 XVII A, viridis subsp. sinuata +.59l XVIII A, acuminata var. acuminata —.649 XIX A, incana subsp. rugosa +.54l land III, and I and through 10. Nearest sis of the 115-Chan scatter diagram of ( ure 8). The neares‘ rection of the arro‘ mutual nearest neig‘ each of several dif ships, include m 5. m subspp. w, gm, ant and g. acuminata Va 5. castaneifolia. exPectcd from a sul however, including 01$ in the A. E C"mllonent IV, A. i to diVerge from A. related With that WP as a Whole 1 Within SpQCiI Mwl’blon - . um Points are rather ahigh range of v A c inc % SubSp. 126 I and III, and I and IV are plotted as scatter diagrams in Figures 8 through 10. Nearest neighbors, on the basis of distance (from analy— sis of the 115-character data set) are represented as arrows on the scatter diagram of OTU's against principal components I and II (Fig- ure 8). The nearest neighbor of each taxon is indicated by the di- rection of the arrow. Where there is an arrow on both ends of a line, mutual nearest neighbors is indicated. Taxa clustering together in each of several different plots, indicating the most natural relation- ships, include élEEi jorullensis vars. igrullensis and firmifolia; A. viridis subspp. viridis, crispa, and sinuata; A, incana subspp. incana, rugosa, and tenuifolia; A, oblongifolia and A, rhombifolia; and A, acuminata vars. acuminata and glabrata, A, ferruginea, and A. castaneifolia. In most cases these complexes are what would be expected from a subjective evaluation of the genus. Exceptions exist, however, including the presence of Alpg§_rhombifolia and A, oblongi- fglig in the A, incana group. In the diagrams involving principal component IV, A, incana subsp. tenuifolia and A, serrulata are seen to diverge from A, incana subSp. incana, A, incana subsp. rugosa, A, oblongifolia, and A, rhombifolia because they are highly cor- related with that principal component while the others are not. The group as a whole retains its identity in the long run, however. Within species, the OTU‘s appear in a variety of patterns. In ALEgg oblongifolia and A, viridis subsp. crispa, for instance, the points are rather widely-spread in the scatter diagrams, indicating a high range of variability. Taxa having tight clusters include A. incana subsp. tenuifolia, A, maritima, and A, jorullensis var. 127 Figure 8. Projection of taxa of Alnus according to factor scores for principal components I and II using morphological data. .w wusmwm «9. i, :3 was 128 129 Figure 9. Projection of taxa of Alnus according to factor scores for principal components I and III using morphological data. (0 3 CASTA . ACUGL i.) \- \‘o GLABR RUBRA / .o opswflm \i) EB? .\ (.3 ~\ \ ..-/ as Ox :5 m 3 0 3 1 95: \£ \ 3 SE 8:: ..\. \ .9 ”a \39 EE. 9:2: \9 = Esmgsg 422k ma: Q4 5530 n.1,...) , ./ . is“ .aa.‘ KW . , e. <55 MUS. Gk «n5» _: Figure 10. 131 Projection of taxa of Algg§_according to factor scores for principal components I and IV using morphological data. A l/JORUL .‘1 cram: , I 7 132 .oH magmas MS mi SE : The phenet means of a clue data set. The some cases, the closely with t. “new to the A. m relationship 1 In one C] t0 8 single in not correspom illdicate the . genus. Throu ma“? Possibly Cluster 13., t1". the distance dicating thai etic factors quite plausi‘ maticauy C0 it is also l another Teas CluSte] when Separa relation Co firmifolia. The phenetic relationships among the characters were studied by means of a cluster analysis of the 150 characters in the complete data set. The phenogram from this analysis appears in Figure 11. In some cases, the clusters which resulted from this analysis correspond closely with taxa or species complexes. Cluster A corresponds to the A1235 acuminata complex, cluster F_to A, maritima, clusters A_and P to the A, jorullensis complex, £23. In several instances, no clear relationship is seen in the characters of a particular cluster. In one cluster (2) the entire group of characters pertains to a single trait (pubescence) on different organs. This cluster does not correspond to any particular taxon or species complex, but may indicate the unity of pubescence as a taxonomic character in the genus. Through study of the phenogram, it is possible to recognize many possibly-redundant (non-”unit”) characters. For example, in cluster A, the separate characters of leaf length, leaf width, and the distance between lateral veins are very highly correlated, in- dicating that they may be under simultaneous control by the same gen- etic factors. Here such a relationship among the characters seems quite plausible. In other clusters, the characters shown to be mathe- matically correlated are not so logically correlated. In either case, it is also possible that the observed correlation may be explainable by aHOther reason, such as the existence of an adaptive complex. Clusters in the phenogram in Figure 11 appear to be most natural when separated at a phenon line drawn at about the 0.30 or 0.35 cor- relation coefficient level. This is true for all clusters in the tree. Figure 11. 134 Phenogram showing correlations among morphological characters of Alnus based on cluster ana1y31s of 70 OTU's and 150 characters. C HHHHHHII 55$§§§§§§§§§E§§E§a§w§§ sEESz'EEiSas IIIIIIIIIIIUIIIIHHH 555853 I777 77 f7? 7777? 5.335555595183535 ‘ ill/ll/l/l/ll/ #— 135 *IJX) - ’0-90 - 't 0.80 “ + 0.70 - +0.60 ‘- ‘050 - +0.00 - 'OJO - _fi ___i_1———I _____L j—j_____l— — _}——-———-—1 l +0.20 - 90.10 - 0.03 Figure 11. NBSmMNmnwmnwmmNuammhunumflnw {{{rll‘ )- x \- N E___._________.____=m_.m_mE_::E=: s ERR mmmflmvmmnsummmmgnm fl EE§Esa=u 136 ’lfll _- *0.90 — +0.80 - t 0.70 — +0.60 ‘- +0.50 — *OJIO - 1‘030 — ‘02!) - *0.10 - 0.00 Figure 11 (Continued). Smaller clusters to discrete taxa For further ard deviations a character agains cates that a cha taxonomic utilii GM of 0.50 or have reasonably and 19 others w cases and highe found to have I of variation we Using the selected and to states (Table ized and 1 if memt index Wa: hsecond adw Variable Char and Farris (1 sets of Value In terms mo . st primitii 137 Smaller clusters at levels higher than this do not seem to correspond to discrete taxa. For further analysis of the characters used in this study, stand- ard deviations and coefficients of variation were calculated for each character against the OTU's. A high coefficient of variation indi- cates that a character is variable among the OTU's, and therefore of taxonomic utility. Twenty-nine of the 150 characters analyzed had CV's of 0.50 or greater (Table 9). Of these, nine were estimated to have reasonably low within-OTU variation patterns for all the OTU's, and 19 others were found to have low within-OTU variation in some cases and higher variation in others. In most cases, OTU's were found to have relatively few non-overlapping characters when the range of variation was considered. Phyletic Studies Using the method of Wagner (1961, 1962, 1969), 35 characters were selected and assigned generalized (primitive) and specialized (advanced) States (Table 10). Each of twenty OTU's was then scored Q_if general- ized and l_if specialized for each of the characters, and an advance- ment index was obtained by adding the scores for each OTU (Table 11). A second advancement index was obtained by weighting nine of the more variable characteristics by a factor of 0.5 in the manner of Kluge and Farris (1969). Cladograms of the OTU's were prepared using both sets of Values. In terms of advancement on this scale, A, oblongifolia is the mst primitive species studied, and A, maritima is the most advanced. Table 9. Charac at lea Character 12 19 23 26 38 39 46 51 53 57 58 69 88 89 98 113 114 116 129 148 161 171 176 183 189 191 195 198 215 \ 138 Table 9. Characters in Alnus having coefficients of variability of at least 0.5 among taxa. Coefficient Variability Character of Variability Within Taxi 12 111.7 Low 19 82.8 Variable 23 83.6 Variable 26 56.9 Variable 38 93.5 Variable 39 58.1 Low 46 56.6 Variable 51 133.0 Variable 53 58.7 Low 57 84.0 Variable 58 64.0 Low 69 55.3 Variable 88 75.9 Variable 89 93.8 Variable 98 51.9 Variable 113 195.8 High 114 81.4 Variable 116 86.3 Variable 129 52.4 Variable 148 76.4 Variable 161 72.5 Variable 171 56.1 Variable 176 62.5 Low 183 77.0 Variable 189 57.9 Low 191 56.9 Low 195 63.6 Low 198 57.9 Low 215 65.7 Variable I?!" '1 UCOwOHm COaUHUCOU UE¢V£< nahfim mthrH WQNflHmfluwam COHUHUCQC UGNMHQHQCQU Fltvx-mwuu sweet Mgr-HAV n90 mflmkfiHflwF—Q UHUVIHxnkfih 180:“ ~0ij hmuUQNWCU WIHVU nv-Nchvnth ION MQQEJZ OHQNH ponoHo>ovimahoom womoHo>oleHo3 unoanHo>ow mafia ufisum mm uoxoamom uoam mwu oamum ocoo RN ucowoum uCown< Cowmsm nucmwuomludofimaflm mm is; wees fimefl anaemia mm oBu on snow snow on xwm uonEDC uuoa sucmwnom «N unom o>wm nonficc uoauomyn oumcflfioum mm HHmEm swung noQEDC awxumo opmcHEmum NN anesm wcog Suwcwa oHoCDUoQ ocoo HN ”Tao:m wcoq nuwcoa ocou ohsumz ON mumuwfiom omoEoomm wCHasouw awxumu oumaawuwwm aH HHmEm . . owned yonfinc cflxumo oumaawumwm wH 30A swam oHumn fiumcoa waowuoalovoafl wood NH ucomnm so couscom poocsocoum mafiaocoan cflw> hawpcooom 0H msoEOH©owommMHowEom wsoEoqupmammuo coHum¢o> mood ma 0% msoooocmupEoz wsowomoaoo mwocxoflau woman Moog ca U souumz mow: snows wmog ma uuonm wcoa nuwCoH wood NH oum>ono oum>o oamfim mood HH 3958 on $350 35... on? xoaw was; 3 unwound ucomn< COMuDHo>ou Campos wood a HHmEm owpmg swam Luoou wood w non: oHuuHA cowmwogmfioo Noam wood 5 ouso¢ omsuno Ho macadom onzu moon wood 9 HHmEm owuma oNHm oasaflum m osu annu ouoz 039 ”HonEsc wamom tam s ucomn< anomoum xfimum ozm m ucomoum ucown< :oHumecoaowwww uooam N Diem 8h “Ea: H dMMMMMHHMWfiMmeofioon mmwuwpcou wouflaouocoo 1IIIJHHHMMMMMWIIIIIMMMEMM mxmu mDGH< Mo wwwhamcm owuwahnm How tom: muwuumhmso .34 manna Fr t wumfihmmHCD 8030“ H0 HUGH COHUHUCOU “UVNHHQHQMVAHW QUQflhflmeF52 whoa MO 0>flh “05530 080% CH0HN CUHHCQ HMWWIMFNU (c .HQQEHHZ COHUHMVCOHV “UN.“ HQHQCQQ 1111111111 oAUQDCfiuCQUV 0H UNQQH 140 woamoocoo c85u5< wHoo kun HHmEm ouofluwmanb nosow so each 11 coauneaoo sownaanoaam vmmmmwm HouCMS ca chamomxo cwxpmo oumuwuom wdwuoson we oEHH MM umaoz uouwnwfi mo oumEHHo kuocoo mm owwmq pouwnmz CM oHSuwon HHom Nm ouMwHomHuHDZ oNHm Hammo> @003 am ouoe Ho.o>am vamp hon @003 . Honfisc whom dwmhw :wHHom MM cofluflvcoo woNflHmuocoo Houomumco nonfisz .Illll Avoscflucouv 34 wanna A. A1 0‘ A A. INC VHQ—Wrfi UUQUW sea A. VlF A. JORULLEi F~UQMV MHO A. VlRll A. VlRl omhnnsuknv .HOHH .HQU \A 0— VPmU .CH «LOXQU F~090 UflUan F— miw mkflH Much... IvafinVMvnum m,ana.H< Ho lysis of Alnus species. J.C 8118 State of each character for each taxon in the phylet A-uuLC LL. A A Character Weights . VIRIDIS SUBSP. VIRIDIS . VlRlDlS SUBSP. SINUATA A. VIRIDIS SUBSP. CRISPA A. SERRULATA A. RUBRA A. RHOMBIFOLIA A. OBLONGIFOLIA A. MARITIMA A. JORULLENSIS VAR. JORULLENSIS A. JORULLENSIS VAR. FIRMIFOLIA A. A. A. INCANA SUBSP. TENUIFOLIA A. INCANA SUBSP. RUGOSA A. INCANA SUBSP. INCANA ACUMINATA VAR. GLABRATA ACUMINATA VAR. ACUMINATA Characters 141 ooooooocnmmv‘moommoooomoooooooonoooo ................................... HHHHHHHHOOOOOHHooHHHHoHHHHr-IHHHOHHHH H-«reH H H H H14 H.H—.H H H.4—«H H H H H.H H.H».H H HIdFiH H H H H.H H.4H.H H H HvHH H.H H H HvaH H H.H H H H H H H H H H H HH H HH H HHHHH HH HHH HH F.H H r1H H H H H H H H H H H H H H H H P‘H H F.H H H H H H H H H H —.H H ~.H H H —.H H H H.H H H H H H H H H 10 Unweighred Divergence Index heighted Divergence Index sings rubra a general, rath so. The A. j while A. v_ir_i In the . acter 33, th Numerous cas °Pment of th 16, 17, 18, asters). s; 91 Tucker ( In a f ized Charac Closely rel 1“ other c; or PhYloget he coanr g The c 1ationshi, it was in] (1'3 31111813 Rod. Occc inf(”man mold 1161‘ Wei 142 AAAE§_£AEEA_and the members of the A, acuminata complex are, in general, rather unspecialized, while A, rhombifolia is somewhat more so. The A, incana and A, jorullensis groups are moderately advanced, While A, viridis and A, serrulata are highly specialized. In the characters used, one evolutionary reversal is noted (char- acter 33, the advanced state apparently being lost in A, serrulata). Numerous cases of convergence are seen, including the parallel devel- opment of the derived condition in characters 1, 6, 7, 8, 10, ll, 13, 16, 17, 18, 22, 23, 28, 29, 31, 32, 33, and 35 (18 of the 35 char- acters). Similar patterns of convergence have recently been noted by Tucker (1974) in the genus Quercus. In a few cases, it is possible to imagine a loss of the special- ized character (as with character 9, leaf margin revolution) in taxa closely related to those having it (in A, oblongifolia, for example). ‘In other cases, specialized characters typical of supposedly natural :or phylogenetic groups (character 20, small cone size) may actually be convergent within the group. The Cladogram (Figure 12) shows the presumed phylogenetic re- lationships among the taxa studied. Using the 35 characters selected, it was impossible to split AAAEA viridis subsp. crispa from A, 3131? $13 subsp. viridis, and A, jorullensis var. jorullensis from A, iQEElr lensis var. firmifolia. The basal node of the tree, as well as the node occupied by A. oblongifolia are too compressed to provide much information. To correct both of these problems, additional characters would have to be added. Weighting variable characters less than invariant ones to Figure 12. 143 Presumed phyletic relationships among Alnus taxa based on the weighted divergence index. —l .7 Title... 1 .NH magmas uuzuwuu>~ .— anHNHHHoHammmmumuHchmsmwsmeHHHNHmHaH . _ . _ . A 9. H n _ 1 3: 55353 .¢<> Six—Id: .< :3 5425233 .< I a - a 3 £5.53 Java—LS .< a 3: LB: .< 2... .3 st. . . .«mn.h“n...a..:..=..=..=. .2 s: 5;. .< . ‘ 2: 5.25 a 25¢; name.“ 25:; < 2 2 2 2 u u .3 O ‘ S... .3 .sn .3 .2 .2 .v .n .u .C ‘ \ \ . . . 98:: . 2“ z 2 t <39: \\ \ :3 <53; £3.» 25:: .< an J» o: -~ .uCo \\ _ .< 144 V A: <13:— iu 532: .< \ \ . . . o‘ :8 Soon—imp {mm—E 2n 3 2 = .2 .2 .. .u .2 (39:5: u 23.5.55... 65> ”55.1.23... i , \ \ H: 3092 fans» . —.< Lu .3 .2 .2 é .« .8 5.52.5» .< increase the what, but not consistency", sistent with avalue as 1' (1.000 indie doubtedly at and reversal One of the conveni tEChnical 1 number of e the Compute To try the and Charac OTU's used matrix Was Present 1;] Ceding sy: The Preparing the Morse are New Pro] 145 increase the parsimony of the tree changed the relationships some- what, but not greatly. Using Kluge and Farris' (1969) "index of consistency", the weighted tree (c = 0.654) is found to be more con- sistent with the data than the unweighted diagram (c = 0.538). Still, a value as low as 0.654 must be regarded as rather unsatisfactory (1.000 indicating a perfect "fit“). The lowness of the index is un— doubtedly at least partially due to the large amount of convergence hnd reversal of the characters used in the analysis. ‘ Taxonomic 22£3_Matrix Studies One of the most important problems facing taxonomists today is :he convenient storage and retrieval of an enormous quantity of Lechnical information about organisms (cf. Morse, 1974a). One of a umber of experimental systems designed to deal with this problem is he computerized data matrix of Morse (1974b), mentioned briefly above. 3 try the Morse system with the present data, a taxonomic data matrix 1d character list of 23 taxa and 100 characters (including many of the 7U's used in the phenetic analyses) were prepared. Eventually the trix was shortened to include only the 16 taxa recognized in the esent treatment (Printouts 4 and 5, appendix). The format and iing system used are those of Morse §£_alf (1971). The original matrix was used with taxon-comparing, description- paring, key-construction, specimen-identification, and other of Morse routines. Samples of the results of some of these tests provided in Printouts 1, 2, and 3, below. Probably the most useful program for present purposes was the Printout 1. 146 011011 COMPAJ HIGH TAXON’: 16 ALNUS Al ' I -2 I '2 1 '3 I -3 A '4 I '4 I .S t ‘5 A '6 I '6 I ‘6 l '6 ‘ .7 I .7 l .7 I ’ matrix .8 I ' sing taxonomic data .9 ‘ 1 taxon comparisp’IEZR: .9 ‘ e C . . PrintOUt 1. :iESSM with program .13 f '11 1 '16 l ‘16 1 '17 1 '19 ‘ ‘20 ~22 NEXT: I‘RE ““CH COMPA WI CH TWO 1 MONO"50an n; 3 18; \ ””0151 : 50: 55: , 69: 6. 2‘ N50: “R 147 Printout 1. mICH COMPARE ROUTINE?2 HIGH TAXON716 16 ALNUS ARGUTA (MEXICO) -l ALNUS PRINGLEI (I 27) -2 ALNUS ACUMINATA (I 15) -2 ALNUS ARGUTA (CA) (I 17) -3 ALNUS FERRUGINEA (I 19) '3 ALNUS ACUMINATA (N) (I 24) -4 ALNUS OBLONGIFOLIA (I 4) -4 ALNUS JORULLENSIS (I 23) '5 ALNUS CASTANEIFOLIA (I 18) '5 ALNUS FIRMIFOLIA (I 21) ‘6 ALNUS RUBRA (I 6) -6 ALNUS GLABRATA (I 22) ‘6 ALNUS OVALIFOLIA (I 26) '6 ALNUS INCANA (I 29) '7 ALNUS RUBRA (PINNA) (I 7) ‘7 ALNUS TENUIFOLIA (N) (I 13) -7 ALNUS FERRUGINEA (O) (I 20) -8 ALNUS TENUIFOLIA (S) (I 14) -9 ALNUS RHOMBIFOLIA (I 5) -9 ALNUS RUGOSA <5 8) ~10 ALNUS SERRULATA u 10) -10 ALNUS FIRMIFOLIA (5) u 25) ~11 ALNUS RUGOSA (AM) u 9) ~16 ALNUS SINUATA (5) (a 12) -16 ALNUS VIRIDIS (I 28) -17 ALNUS SINUATA (N) u 11) -19 ALNUS CRISPA (a 1) -20 ALNUS CRI SPA (mom (I 2) ~22 ALNus MARITIMA (I 3) ‘73 1=RECYCLE 2=NEW PROGRAM OR MATRIX 3=STOP --?l ICH COMPARE ROUTINE?3 CH TWO TAXA?!6;20 OVERLAPPING DIFFERENCES OF THESE TAXA: CHOTOMOUS: B 1 3 ’ 8 1 I 3 1 .TISTATE: I 4 7 8 2 1 UTITATIVE: 6'20 000 l=RECYCLE 2=NEW PROGRAM OR MATRIX 3=STOP -'?3 148 WWDCNJ NE EH3 UZHmWBOd hmo ZOWQNW ”WE QWECL— “Zuzfit hull! uselel 0‘ All WFWJrogm QNH§(QNW 2° mzmom mzHVflhcu ”H.B(hCHEFW QZQ NHQ‘HJHHW Hm ommw WMJ‘UW IEDGHNB mmoz mo V E Qmmmaoo “gm mMH‘ZHh 0“ NFGZHEUA‘ mo ”H.500“ want Adam 0‘ “H.006 .Fmoa ngZle3 20 “Nu/Eon \flqébwa mHKVQMcH .mW Sample descriptions of Alnus taxa using taxonomic data matrix ALNUSM with program DSCRBE. Printout 2. qnfim Hmo “.321! {wrist-Iain IN. “Joucuucqhfl 149 NH¢ZHZSU€ Mo NEDD¢ XMQQ mqfimuhm 5204 mmwhmzomouz Mm Oh O— WHZMECAHK mQHS m¢ $204 m€ mmzuh N\H m CH m\~ n mm>¢m4 mans .22 @ OH m mmJDmHHm meNmHZG ht MQHS mfiMHwEONUHE N" OH on mZHXH¢U mh¢ddnhmum no mMAUZDme MAm€Hm¢> mmaomhmm hzmommmomm¢ humtm 6204 meHMEomouz mm 08 mg mHHDmh xHGZWA J¢HOH ho hzmommm Oh 08 cm MMQG H¢ QMDuDHQ mmmmHZ€ mmmoq X>A¢U ZCEH meZOA m2fi£¢8m m¢JDQZ44¢DWDIWND¢M4 fists—37* .bo .Qo om .~w ow cu oh ocm 150 W Hm 2M— l—I—DWO.) MD 24¢ .mm -0... W30m0¢mm00 WMQaWJm .und‘mq e‘V “QM; W¢ n—MMHAH WGV WWEHH‘ N OH. N\H “bummed .KGVMNJ em. Mann—CHEM“ ”WE .nno :hufifiu Hun—HI “WEN“ o.- ZAWE mmo: “Nata .uNg‘MlN [IND Eon/HQ IN mkumlurnvz¢m4 .mm nz¢v Gmowam m024¢ .m o.... 30Amm mDOUD¢Jw mw>¢m4 .ma wan: md mama md ¢\— Z4mm¢xw 92mm m4¢0m mzoo m0 xmm¢ .h MQH3 m¢ 0204 md mWZHH N\H N z¢mh wmoz #02 mw>¢m4 .0 ¢md o0 MQHS m¢ mmwn m¢ muZHH m Z¢IH mmoz H02 mmmod h¢w4 om a¢ZZumv cmmDm m024¢ .h .o... Nan? md ammn m¢ mmZHH N 2¢$H mmoz mumOJ h¢w4 om NJONkmm me k0 rhwzmd mmh WMZHH 0~ Z¢IH mmoz 802 mmn¢dm h¢m4 m0 IHEZMJ .m 0 m024¢ .ON a.... mbomodwmoo H02 mmn<4m k¢m4 .Q mumZMAJDm00 mDZA¢ .mm .o... m30m0¢mm00 mmm¢qm hdmd .v MQHB m¢ Emma m¢ WMEHH N CH N\~ mmmoq kqm1~ ieVN EéHmM—nn Mame—H. EE INN 152 ¢2Hhum¢z m324< .m .o... . 35%; so 3ch m5 7: ES. Sow mzCEB msaflflmmfimms mhm4202¢mm MWOJU 20 wzmom hoz mmzoo mh¢2~S<3¢ rdlm¢mm Hzmm mZHxh¢0 MH¢2~E€HW 02Hm¢mm Hw4202¢mm .wd Dmnzwhxm H02 w4¢om m200 k0 Hmoq A¢2HZEMH .bn a¢0v ¢wam< m524¢ or" ..... ¢wzH00mmWh m324¢ .0” ..... QMQZWme m0 mhdwzoqw w4¢0m m200 mo mmOJ 4<2Hzmmh ob" hmoxm 802 mwQ¢Am h¢m4 m0 modhmam mMBOA 20 wozmommmbm .o~ mama mG mmwn w¢ €\~ 2¢zh mmmg mwmoq m¢m4 om“ m251 OH Iom¢2 Bomb #02 02Hmm304h .0 dudomuuzoamo m024< .v ..... mama m¢ 0204 m¢ mmEHH m\u a zdmfi mmmd H02 mm>¢m4 .fl— ¢m4 o¢~ 2h2¢ummm MIR 2¢mh mwhmoxm 802 mzm2¢hm .—~ AI—OZV Qnmm HWC VI? 5.. C Wag—44¢ on . . o o . m W200 k0 NNOJ JQZHZENH DIN éZHEmh tsVN ImuN «mum HED QNQZNHXN m0 HHCwZOJN Nab WDZJ¢ .QN .oo.o W M200 k0 MN 4 WMNOJ had mHQHmH> QMHQEIFVAMH .H‘oz many ”NON? W¢ flammq Mn“ {\N ZGE WW.” mmMNI-IWD‘HU MWOEUE 2N HEB WZHVFHqU MNIFGVJJHIFWHnN IAMVWSCHUEOUV M Uloulfikk. 153 any ¢h¢DZHm mDZaa .ma o...- nZv uld be expected from other lines of evidence). Both parts of :32: viridis subsp. viridis in Europe and Asia have subsp. crispa ‘om the eastern part of North America as their nearest neighbors, illowed by other segments of subsp. crispa and then subsp. sinuata. .bspecies crispa in the eastern United States and Canada is slightly oser to western (European) subsp. viridis than to eastern (ASian) mbers of that taxon. Subspecies sinuata, likewise, is closer to stern members of A, viridis subsp. viridis than it is to the Asian pulations. From these, and similar observations, one can predict that this circumpolar grox anumber of mu western Europe a recently into t large-leaved fo migrating to we hybridization l siderable natui The data A: M subs as nearest nei closer to subs they are close the range of . 5' MILE! sub t°° Close to the Atlantic section of t} range (the 01 Could indica' cations, the Ozarks on th 0f the Speci Intern“ 5‘% $8084 them 157 ircumpolar group was once a single species which has diverged into number of races. The small-leaved form apparently developed in estern Europe and eastern North America, being divided relatively The ecently into the populations called subspp. crispa and viridis. arge-leaved forms appear to have diverged in northeastern Asia, later igrating to western North America. In northwestern North America, ybridization between the two American subspecies has produced con- iderable natural variation in that region. The data show that both southeastern and western members of . incana subsp. rugosa have northeastern members of the subspecies b nearest neighbors. Further, all OTU's of subsp. rugosa are loser to subsp. incana than are those of subsp. tenuifolia, and iey are closer to those of the southern than the northern parts of 1e range of that taxon. Alnus serrulata is slightly closer to . incana subsp. incana than it is to subsp. tenuifolia, but it is not to close to either of these taxa. The OTU's of A. serrulata along ,e Atlantic Coastal Plain show a pattern leading to the southeastern ction of the species, but A. serrulata from the western part of the ‘ge (the Ozark Plateau and Vicinity) is much less like it. This 1d indicate that A, serrulata survived the Pleistocene in two 10— ions, the southeastern United States on the one hand, and in the rks on the other. After the glacial retreat, the southeastern part the species may have migrated north along the coastal plain. Intermediate forms between Alnus incana subsp. rugosa and serrulata were invariably found to be more similar to subsp. osa than to A, serrulata. In the region where hybridization is supposedly occur from _A_. m One might i verging into A. cene. A. M elevations in t North from the subSp. {W1 northern part . the unglaciate Nova Seotia al in contact, tl' intermediate 1 Character 1hbllcéil "adap that thESe Co Diexes of ch a to adaptive ( Those COmple; specialized Exafilina the features may explain to those not has PIObabL EVOIutiOH i 158 supposedly occurring, introgression therefore appears to be proceeding from A. serrulata into A, incana subsp. rugosa. One might predict a circumpolar predecessor of Alnus incana di- verging into A, serrulata in eastern North America before the Pleisto- cene. A, incana subsp. tenuifolia apparently developed at the higher elevations in the West and subsp. rugosa in the lowlands of the North from the Rocky Mountains eastward. After the glacial retreat, subsp. rugosa became re-established on glaciated soil across the northern part of the continent, and A, serrulata spread throughout the unglaciated parts of eastern North America and as far north as Nova Scotia along the Atlantic coast. Where these two species came in contact, they apparently often hybridized, giving rise to the intermediate forms which are found in these regions today. Character analyses showed that the characters chosen do not form logical "adaptive" complexes in the American taxa of A1235, or at least that these complexes are too intricate to be readily apparent. Com-_ plexes of characters corresponding to species (and therefore indirectly to adaptive complexes) are seen in clustering the 150 characters. Those complexes which are apparent in this analysis are all highly specialized types. Examination of the characters themselves showed nearly all of the features to be quite variable and overlapping between taxa. This may explain the fact that the alders are often difficult to identify to those not familiar with them. The confusing variation of the genus as probably arisen from introgression in some instances and convergent volution in others. In const the generaiiz of Wagner (19 in some cases peered in mo: sented in on generalized derived cond as well. Sc 0“ the basis in the Size sociation w “Ch Choice htthods of about the a (if. Whiffi hhy1°genet of error w larger hum Die), The new, as a1 many Chan} Probably in other The Closely v 159 In constructing a cladogram, choices had to be made concerning the generalized and specialized states of the characters. The ideas f Wagner (1969) were generally applied in accomplishing this task. u some cases, characters were considered primitive when they ap- eared in most of the OTU's, with the derived condition being repre- ented in only a few cases (an example being 23252 leaves considered eneralized and obovate leaves specialized). In most cases, the erived condition in such characters seemed the most logical choice 3 well. Sometimes characters were considered primitive or advanced n the basis of observable trends in the genus (such as reduction n the size of the wings on the fruits), and in other cases as- ociation with other primitive characters was used. Admittedly, uch choices are subjective, and the chance of error always exists. athods of phylogenetic analysis avoiding the making of assumptions )out the advancement of individual characters have been proposed :f. Whiffin and Bierner, 1972), but even with these techniques lylogenetic decisions have to be made. It would seem that the chance : error would be less following the Wagner method simply because a -rger number of choices is made (providing a larger statistical sam- e). The techniques of phyletic numerical taxonomy are relatively w, as are those of phenetic analysis, and doubtless will undergo ny changes in the years ahead. For the present, these methods are Dbably best used as a means of deriving hypotheses to be tested other ways. The clusters produced in the cladogram (Figure 12) agree very )ser with the groupings seen using traditional methods, phenetic analyses, and ch tree three major the subgenus A_i one group, whil 160 analyses, and chemosystematic techniques (discussed below). In this tree three major divisions, corresponding to the three subgenera, Alnus, Alnobetula, and Clethropsis, are readily apparent. Within the subgenus Alnus, the Latin American taxa cluster together into one group, while the northern taxa segregate into another. Recently use of biochen or to compleme themselves 31'! note similari hi Fish and W Simpler methc “htographic I determining . on a Presenc as with any Single diffs Shihs- The Recentli’, a nomic hechn Turner, 19, brothel-s, 1 cism (cf, i The c' of attenti \ l SEQ CHEMOSYSTEMATIC STUDIES Recently an increasing number of taxonomists have been making use of biochemical data, either as the sole basis for conclusions or to complement morphological evidence. In some cases, the compounds themselves are identified and their origins determined in order to note similarities in biosynthetic pathways, a technique used recently i )y Fish and Waterman (1973). Many systematists, however, use the simpler method introduced by Alston and Turner (1959) in which chro— latographic patterns, or ”biochemical profiles” are compared without ietermining the actual chemicals involved. The compounds are treated, n a presence or absence basis, as ordinary taxonomic characters, and, s with any taxonomic characters, multiple correlations, rather than ingle differences, are considered most useful in determining relation— hips. The data may be evaluated qualitatively or quantitatively. ecently, a number of workers have successfully applied numerical taxo- Dmic techniques in chemosystematic studies (Adams, 1972; Adams and Jrner, 1970; Flake E£.3£°! 1969; Koshy E£.El°’ 1972; Payne and Fair- ?others, 1973), although these methods have also drawn some criti- ism (cf. Weimarck, 1972)l. The classes of compounds which have received the greatest amount 5 attention by plant taxonomists include phenolics, terpenoids, \ lSee rebuttals by Adams (1974) and Crawford and Born (1974). 161 and protein: phenolic cl. chromatogra employed at the phenoli to be at or Giann; data in ta morphology POSSibie p to this lj 2) to stur the usefu Chemistry other sys Sithatior Nume in hllhhi Qhemical of Saito Species Proteiné tihghis] the Spe Studies 162 and proteins. Of these, one of the most widely-used has been the phenolic class, the compounds being separated by means of paper chromatography. Each of the above classes of substances has been employed at all levels of the taxonomic hierarchy, but, at least for the phenolics and terpenoids, their greatest usefulness has been said to be at or below the level of genus (Turner, 1969). Giannasi and Rogers (1970) list three reasons for using chemical data in taxonomic studies: 1) to test classifications based on morphology, 2) to detect species-specific patterns, and 3) to detect possible phyletic relationships. One might add two additional items to this list: 1) to detect and analyze intraspecific variation, and 2) to study natural hybridization. Alston and Turner (1963) summarize the usefulness of chemosystematics, stating: "what comparative bio- chemistry has to offer is supplementary evidence which, when added to other systematic knowledge, may clarify or help to clarify a given ituation." Numerous chemical compounds of various types have been isolated n AAEAE (Hegnauer, 1964), but few attempts have been made to use hemicals as taxonomic data in the genus. An exception is the work f Saito (1970), who studied relationships among several Japanese pecies by means of acrylamide gel electrophoresis of soluble pollen roteins. In this work it was found that the alders could be dis- inguished on the basis of their pollen protein composition and that e species were relatively constant chemically. Several other tudies, using phenolic data, have proved useful in Betula (Koshy EL -, 1972; Payne and Fairbrothers, 1973). In the present study, phenolic ch: lationships Sample out the ge< were COlIEI from one-yr were dried C. Four t each popui S‘thles f( 0f Saulple: in at Dies of L material fmh 1 tc SPEcimem State Un‘ Phe in Better Milkinsc material chhhinet were ex of boil 163 phenolic chemical characters were used to help clarify taxonomic re- lationships which could not be perceived from morphology alone. Methods 221 Materials Samples of leaves of 13 taxa of élflfli were obtained from through- out the geographical range of each taxon (Table 22, appendix). These were collected in mid-summer (after the leaves had reached full size) from one-year-old branchlets on the lower parts of mature plants and were dried for approximately ten hours in plant presses at about 400 C. Four to eight populations of each taxon were sampled. Within each population, Specimens were taken from five individuals. The samples for each population were later combined to reduce the number of samples for analysis. | In addition to the material collected in the field, eight sam- ples of Latin American taxa were taken from herbarium specimens. This material was selected on the basis of appearance and age, which ranged from 1 to 30 years. A total of 81 samples were analyzed. Voucher specimens are deposited in the Beal-Darlington Herbarium at Michigan State University. Phenolic compounds were extracted from the dried leaves following in general the method of Hanover and Hoff (1966) and Hanover and Wilkinson (1970). After first removing the petioles, 0.4 gm of leaf aterial from each of the five collections at a particular site were combined into a single sample weighing 2.0 gm. Phenolic substances are extracted by homogenizing the samples in a blender with 100 ml f boiling water for two minutes, filtering, washing with 50 ml of boiling wat was found b extract suc The we ethyl ether by five 50 was evapor alcohol. in their v Phenols. substance chromatog Comp Chromatog nicropip. 3hh chro left~han “Peer ed her, whi a hixtur were th anOwed SheEts Rated j (3:17) reaChe 164 boiling water, and then repeating the entire process once. This method was found by Hanover and Hoff (1966) to effectively eliminate from the extract such interfering substances as tannins and chlorophyll. The water extract was washed five times with 50 m1 portions of ethyl ether in a separatory funnel to remove waxes and fats, followed by five 50 ml portions of normal butyl alcohol. The butanol fraction was evaporated to near dryness and brought up to 1 ml with n—butyl alcohol. The ether fraction was found by Hanover and Wilkinson (1970), in their work with spruce foliage, to contain an abundance of simple phenols. For AAAEA, however, only very minute quantities of phenolic substances were obtained when this fraction was concentrated and chromatographed, and it was consequently discarded. Compounds were separated by two-dimensional descending paper chromatography. Fifty microliters of extract were applied with a micropipet to a point on the upper surface of 46 X 57 cm Whatman i BMM chromatographic filter paper 8 cm from each edge in the upper i left—hand corner. The papers were folded on a line 5.5 cm from the pper edge and placed, in groups of eight, in a chromatographic cham- er, which was equilibrated for two hours with the lower portion of a mixture of n-butanol, acetic acid, and water (4:125). The sheets were then irrigated with the upper portion of the solvent mixture and allowed to develop until the solvent front neared the bottom of the sheets (about 16 hours). After drying, the chromatograms were irre- ;ated in the second direction with a mixture of acetic acid and water 3:17) and allowed to develop until the solvent again had nearly eached the bottom of the sheets (about 5 hours). They were then thoroughly drie Completed ment in daylig' illumination, droxide under lightly with c‘ Spots were out actions to ear hi the same c tallied, but 0f any Spot. Ninety- 13 tth anal and in Table Feared on e\ in every Sp, on individu, From 23 to unique Comp 1“ all with at le; the “he: ' and 5. 3C1] \ 165 thoroughly dried at room temperature. Completed chromatograms were examined without chemical treat- ment in daylight and under both long- and short-wave ultraviolet illumination, after a 30-minute exposure to fumes of ammonium hy- droxide under the same conditions, and again after being sprayed lightly with diazotized sulfanilic reagent (described by Smith, 1960). Spots were outlined, their Rf values calculated, and their color re- actions to each treatment noted (Table 23, appendix). Spots produced by the same compound on various sheets were identified, numbered, and tallied, but no attempt was made to determine the chemical identity of any spot. Results Ninety-one compounds were found in the butanol fractions of the 13 taxa analyzed, as shown in the composite chromatogram (Figure 13) and in Table 23 (appendix). Of these, two (numbers 10 and 15) ap- peared on every sheet and four (numbers 6, 13, 26, and 28) were found in every species but not on every sheet. The number of spots present n individual sheets ranged from 16 to 48, with an average of 29. rom 23 to 52 compounds were found in each taxon, and at least one nique compound appeared in all but four (Table 12). In all but three instances, the taxa studied were distinguishable, ith at least one compound always present in one and always absent in he other of any pair. Exceptions to this were Alnus incana subsp. u osa and A, serrulata, A, serrulata and A, acuminata var. glabrata, nd A, acuminata vars. acuminata and glabrata. In these cases, Figure 13. Composite foliage. 166 chromatogram of phenolic c ompounds fr om Alnus A C'D’WATEH CETIC k, BUTANOL_A E 9 c O ‘ however, certa absent in the pair of taxa ( ranged from 0 occurrence of in the analysi Table 12. Un Taxon A. acuminata A. acuminata l- use su' i- Late su‘ A. jorullens A. jorullene A. maritima A. oblongifc A. rhombifoi is his: A. serrulat; h; viridis l A. Viridis x Chemic coefhclem him, i hum, 19 168 however, certain compounds were sometimes present in one and always absent in the other. The number of spots useful in separating each pair of taxa (always present in one and always absent in the other) ranged from O to 11 and is listed in Table 13. The frequency of occurrence of each compound in each taxon was calculated for use in the analyses. Table 12. Unique phenolic compounds found in taxa of Alnus. Number of Unique Taxon Compounds Compounds A. agpminata var. acuminata 30 72 A, acuminata var. glabrata 26 A, incana subsp. rugosa 44 62’ 74 A, incana subsp. tenuifolia 39 A. jgrullensis var. firmifolia 29 86, 91 A, jprullensis var. jorullensis 26 88 A. maritima ' 23 77s 80 A. oblongifolia 36 A, ghombifolia 48 A. rubra 40 22 A. serrulata 45 4 A, viridis subsp. crispa 52 39, 40, 41 A, viridis subsp. sinuata 43 76, 79 l Chemical similarities among the taxa were calculated using the coefficient of association (Sokal and Michner, 1958), the paired affinity index (Ellison.££_AA,, 1962), and the distance coefficient (Sokal, 1961). The coefficient of association and the paired affinity index are both "matching" coefficients, the main difference being that both presence and absence of a character in a pair of taxa are counted as indicators of similarity in the coefficient of association, while only presence of the character in the pair of taxa is used in the rhombii serrula viridis . viridis . rubra A. A. A. A— A. USU Cd UCUDn—fl. misuse—“d NEE A. j orull n o .I. b o o A— COKHU 0:0 Fwd. A. mariti‘ i ll nu mm nu u a a m m r c m n w 0 .m .l. a a a u am. Am. lm— in. in— ledath‘ M0 ”URN”! M0 IHHH MAN! :0 mm howl: NV>INUUIINHHIIWHhUV “to, UCNWNNHM Mknmxsrnn NV “Fur-JOHHEOU ’l H l 01 0 ln ‘LOQ'MKO mqu-m H m m m m N o HQMCDOmww v—i MNHO‘qu—iu—lmo OHNNHMNd'Ooow O «1 r4 N) o: H -¢ :\ .0 H ,0 q 169 ta ta orullensis acumina firmifolia . CIlSEa inu a bsp. s bSp bSp. tenuifolia ta var. glabrata lS SL1 d acumina orullensis var. orullensis var. . v1r1 serrulata incana su A, incana subsp. rugosa A. oblon ifolia A, acuminata var. A. rhombifolia A. viridis su A. maritima A. A. A. rubra A. A. A. A. mumncflm .answ wwwwnfl> mmmfluo .amnsw meHHH> mumHSMHom wands mEHuHHmE .um> mflmcoaasnofl .Hm> wflmcofiasuofl lull maaowHDCou .mmnnw mucosa mmowsu .amnsw mcmocw I'll mumnnmaw .nm> mumcHEDUm lull MHQCHEDUM st> MHMCHEDUM .|<. hm .« um paired affinity For the pi?! a compound was and absent if i by distance, th taxon was calcr are provided ll Table 15, and ( Phenogram 1h, 15, and 16 those for the seen from thee different resr general simil; Affinity affinity indi to find Patte the range of Some apparem was “Cited in in its Widelj is considera sampled than the p°Phlatj taxon ranger showed affi' 170 paired affinity index. For the preparation of data for use with the matching coefficients, a compound was Considered present if it appeared on at least one sheet and absent if it appeared on none of the chromatograms. For analysis by distance, the frequency of occurrence of each compound in each taxon was calculated. Paired affinity indices for each pair of taxa are provided in Table 14. Coefficients of association are listed in Table 15, and distances are given in Table 16. Phenograms from cluster analysis of the data appear in Figures 14, 15, and 16. The methods used in these analyses were the same as those for the morphological data in the previous section. As can be seen from these phenograms, different similarity coefficients give different results in cluster analysis, though the patterns are in general similar. - Affinities within each taxon were examined by calculating paired affinity indices for each pair of collections of each taxon. An attempt to find patterns of decreasing similarity with increasing distance over he range of each taxon was generally unsuccessful, although there were ome apparent trends. Seemingly significant within—taxon variation as noted in several instances, however. Alnus viridis subsp. crispa, n its widely-disjunct population on Roan Mountain, North Carolina, 8 considerably less similar chemically to the other populations ampled than are any of them among themselves. Affinities between he population on Roan Mountain and the other populations of this axon ranged from 50.0 to 57.8 percent, while the other populations howed affinities among themselves of between 54.9 and 73.6 percent‘ affinity index- the paired using of Alnus Similarities between each pair of taxa Chemica l 1.4- Table I» I} I? I» I? I? 19> I? 1:» I? 1:» I» viridis subS} viridis subs serrulata rhombifolia M w 31% sub % sub W acumin \ata 171 I n.00 «.mm 0.00 0.Nm 0.¢m m.mm w.mm inuata . crispa dis subsp. 3 dis subsp viri viri A- A. n 0.mm 5.00 n ¢.0o a .1 m l a m l .1 u a b r r m a w m S r r m. A. m_ N.o¢ m.q¢ m.om 0.0q o.N¢ . oblon ifolia 0.0m m.¢N ¢.¢q m.o¢ w.wm m.n~ 0.0N m.mN w.wm m.od m.Hm m.m¢ w.m¢ orullensis . orullensis var. N.0m 0.mm m.wm 0.0% n.mm H.0m m.mm 0.H¢ firmifolia orullensis var. o.qm m.om w.w0 0.Hw m.wm N.H¢ m.mm N.q¢ N.H¢ é, incana subsp. tenuifolia o.mm N.N0 0.Nn H.0m ¢.wm m.mm w.w¢ ©.Nq 0.0m w.aw bsp. rugosa incana SU «.0N ¢.¢m n.aq N.m¢ m.om m.w¢ w.m¢ m.m¢ o.wq N.q¢ o.N¢ labrata acuminata var. 0.0N w.Nm 0.5m w.m¢ H.mm m.wm o.wd 0.Hq n.wq n.dq H.mq H.nm ta . ta var. acumina acumina muMDCMm .mmnsm mawfinfl> mmmwuo .amnsm wwwflnfi> mumHSHHmm suns“ waaomapaonu anaowawcoano «Ewuwpma mamfimaasuofl .Hm> wwwcoaasuofl IIIIIIIIII mHHOMHEHHw .Hm> wamfioaasuofl Ill-I'll" mHHowfincou .mmnsm seesaw mmowsh .amnsm masocfl mumhnmaw MHMCHEDUQ 0Hm> MHNCHESUN 0Hm> MUNCHEDUM .« .< <| .« ¢ .« ¢ .4 .fl .« .« S S E t 0 f E .I S S a f .1 m e Q J .1 .1. ..|. .1. 1. 1L 1 a a n n d d u a b n t 1|. 1|. n n .1 :1. .l .l r r m... 0 .l U U a a TH. m T. T. T. b 0 .l. T. T. r C C U U .1 .1. e U h b a 0 0 D. n C C V V S T. T. 0 m . .1. .1. a a a o . a o . n n o . . n . A— lflu— A— A— A— A— A— A— A— A— A— A~ Au‘ \I‘ “0 DCUHUAHHUOU USU. WCfimfi. WDCHAV HO Qvnflu. HO Mnnfil £UQQ CUURUQD mUHUHMQHHEHm HNUHEUEU IWH UHQNN ICOfiU‘WdUOUW-U 172 is subsp. sinuata d Vll‘l A. New man 92 are is can «Rm 0...... do. 0.3 Nam or} was: amps... 335.5 2.. - too. 93 .13 0.8 Q? «.3 Nam m... a... mam Wm... Mmmlmm $3.5 2.3.? .< - 9% m.: W? was ..$ 98 mi Wow is 98 333.8. A. - .2. 93 Q: do. is. .33 an. w.$ o.$ - 93 ..$ 9% mg... Q: 3: 9mm 93 3.8359: .4 - 9% #2 :3 ..$ mam .2. was $3 a... - .23 SK 0.: m... 9m. WE .2322. .< - 0.2 0.9. we. Ti 9? a £33 .4 - w.$ are is 0.? 3.3.253 43% a - it. o.$ SK 3.0325... .38.... 23.: hm sense .4 \l I m.mo p.00 wwowsu .awnsm masocfl nfl n w.Nw mumsnmam .ns> mumcflauom ufl l MHmGHEUOm. .Hm.> QUMCHEUU$ c464 S .1 a s i n 1.. e O a a .l. f .1. a t 1.. .1. _.l. t a a u m o a n r r f a r .1 S O .1 .1 S b m .1 . f u 0 a u r n 1.. C c . . e u a r r .L r o a a o o p V V o u r r s a P P a a b a .1. S S S S V V u .1 1|. .1. .1. b b s a 1 o s s u u a a t o f a n n S s t t S a f .1 m e e a a .1. .I. .1 .1. _.l. 1.. a a n... n d u a b n t l l n n i i .1. r r m 0 .1 U u a a m m .... a a .m .1... a a a m m u w . C V S r r o m . . .1 .1 a a O O O i I 0 O O O O O I A. A. A. A. A. A. A. A. A. A. A. A. A. viridis sx ____— N 9 . virid A. serrulata A IUCUAHUHWQM‘OU f i orullet A. maritima Am. A: A. rubra A. rhombifo' oblong GUCmumflV 020 ME4WJ mic orulle- Am. H< n40 Hmvnflda .1 c u A. An. NO hwfil 50mm &. A. I N CUUBUUQ WNVHUfivaHu—urtflm HUUHEQ-‘Nnv WH GHQ rH , I. 173 o crispa bsp. sinuata bsp A. viridis su serrulata A. viridis su A. A. rubra H.0m o.N¢ 0.mm 0.0m A. rhombifolia o.mq ¢.nq m.0¢ m.N¢ H.m¢ ia . oblongifol A. oucmHUHHHGOU GUCMHmHU vSJ w24nj ¢.H¢ m.Nm n.5m q.¢q o.¢¢ ¢.m¢ m.m¢ n.¢m m.wm ¢.H¢ m.o¢ H.mm m.mm orullensis 'orullensis var. ' A- 031714“ A.) 31:...) labrata orullensis var. firmifolia acuminata var. A. incana subsp. tenuifolia A, incana subsp. rugosa A. A. 0.0q 0.0m H.mm H.m¢ 0.mq 0.5m m.mN 0.Nm N.¢N m.Nm m.mq 0.NN ta acumina acuminata var. A- muwDCHw .mmnsm wwwwpfi> mmwfluo .awnsw wflwwuw> MHmHfihhwm mussy maaoeananu maaowamcoapo mEHuflHmE wwwcwaaspofl .Hm> mamCoHHDHMfi mwaowHEuflw .Mm> mamdoaassofl mwaowwscou .awnsm wcmocH mmowdh .amnsw wcwodw muwuanw .Hm> mumcfifisum mHmCflEDUm 0HM> m 4.1 m c a E 5 o m <3] <1 .31 <3| <51 «5| <51 <| «4| <2| rphology. These groups may have entered North America in very .cient times, while present-day members of the subgenera A1235 and ‘nobetula in northern North America may be more recent arrivals om Europe and Asia and represent a different evolutionary line. It is possible, of course, that some of the chemical relation- ips seen here are merely artifacts, and it would be wrong to put uestioning faith in them when constructing a classification for the us. But neither should one ignore them altogether. As noted by e-Smith (1963), Heywood (1966), and others, the results of chemo- tematic work must be used with other kinds of data, not in isola- D.- “2131-” .,-' From are not p and m 50, and 5 ized com. on morph alders c groups. men, a: and at ‘ shrub h 0f othe to rem. T' w and 5' The La Piltte' rOund for e PHYLOGENETIC TRENDS Divergence and Convergence From anatomical evidence (Tippo, 1938; Hall, 1952) the Betulaceae e not primitive but are low in the Amentiferae. Tippo regards Betula d AAAAA as the most primitive genera, Corylus and Ostryopis less , and Carpinus and Ostrya least primitive. Generalized and special- ed conditions in.AlEE§ have already been discussed in the sections morphology and numerical taxonomy. At the infrageneric level, the ders of the New World have diverged into three very distinctive~ oups. Subgenus Alnobetula is specially adapted to a cold environ— It, as shown by its present distribution in the boreal latitudes i at high elevations, and by morphological features including the rub habit and reduction in the size of many structures. A number Other characters, such as the presence of long peduncles, appear remain primitive in the subgenus. The basic pattern seen in the "primitive" members of subgenus 23 (illustrated best in America by élEEi oblongifolia, A, £2233! A, acuminata) suggests adaptation to a mesic, temperate existence. Latin American members of this subgenus have diverged from this tern to a greater or lesser extent. All have adapted to a year— ld warm climate with rainy and dry seasons, and none become dormant an extended period, as do their northern relatives. Some species 190 ‘1;ng 1" of this g1 drier hab leaves, t a more xe southern many of ' adapted be seen leaves, catkins subSp inter °¢Cur is bl ment View 191 nis group, in addition, have developed the ability to live in : habitats (as A, jorullensis). These taxa have thicker, smaller As, thicker bark, and other characters suggesting adaptation to e xeric habitat. Alnus serrulata and A, rhombifolia of the ern United States have also diverged in this direction and have of the same specializations. A1333 incana, like A, viridis, has ed to a cool environment, but not to the same extent. Here can en convergence in such characters as the shrub habit and smaller 3, while the subgeneric characteristics (such as the pistillate ‘s exposed during the winter before anthesis) are retained. Although some have concluded that subgenus Clethropsis is the primitive segment of the genus (Takhtajan, 1960; Murai, 1964), maritima seems to be considerably advanced in a host of charac- including the loss of fruit wings, the semi-shrub habit, 1 flowering, and relatively small obovate leaves. Its peculiar rphical distribution in America suggests great antiquity, but :8 not seem more primitive, at least in readily-observable char- , than such mesophytic tree species as A, oblongifolia, A, £2952, acuminata. wo species of Alnus in America are represented regionally as cies. In each of these, the subspecies have diverged, but they rade where their ranges overlap. The strongest divergence in A1235 viridis subspp. crispa and sinuata. Subspecies crispa :er adapted to lower elevations, while subsp. sinuata occurs in : conditions, although both are somewhat flexible in this respect. the species as a whole, the montane habitat and the larger leaves of the two. tween the elevatior derived treelike adopted Cause t1 cene. A_l occurre true f. are re size), lar Er othen from trib make Pail cal SYn 192 ves of subsp. sinuata suggest that this is the most primitive of two. In A, incana there is less morphological differentiation be- en the subspecies, although subsp. tenuifolia is adapted to higher vations while subsp. rugosa is shrubbier in habit (apparently a ived state since the other subspecies of A, incana are all more elike). Alnus acuminata has diverged to a lesser extent than Niridis.and A, incana in North and South America, and it has not pted a radically different habitat in either area, probably be- se the two populations have been separated only since the Pleisto- e. Alnus is an old genus, and, as discussed above, convergence has irred repeatedly in many characters. This seems to be especially a for some of the easily-seen characteristics by which the species recognized in the field (habit, leaf shape, pubescence, and cone 2), and in most cases, the specializations have developed in simi- environmental conditions where they have appeared in two or more :rwise distinct species. Similar phenomena have been described A l A the genus Quercus by Tucker (1974). Mt The amount of convergence in AlEEE’ together with its wide dis- tion and apparent repeated migrations over a long period of time the construction of a phylogeny difficult. However, certain rns are repeated in the results of the morphological, biochemi— and phytogeographic studies reported here. A phylogenetic tree, esized from all of this data, is provided in Figure 19. The base Figure 19. 193 Phylogenetic tree of the New World taxa of Alnus and their Eurasian vicariants. 'V VWILIUVW o - cuzfi-Inunr‘ ' V 194 >. <~m~u~m mcmmu. m~zc>4> >. >. <~muc~m mcmmv. <—m—U~m >. mmmxcr>4> >. uzn>z> mammv. 4m2c~mor~> >. _zn>z> mcwmv_ mcmow> >. Hzn>z> wcwmp. ~zn>z> >. mcmm> >. owrozmumor~> >. onzw~10r~> >. >nc3Hz>H> <>m. >nc3_z>4> >. >nc2~z>4> <>m. mr>wm>4> ,>. concrrmzmum <>z. m~wz_morH> >. Loxcrrmzwum <>x. Lomcrrmzmam >. 3>x~4_3> SUBGEN. ALNOBETULA SUBGEN. CLETHROPSIS SUBGEN. ALNUS PRIMITIVE ALNUS Figure 19. __/ “ran-.7 of this (1 the tips species, several 5 times in Neverthe ways lea the pres 195 is dendrogram represents the supposed origin of the genus, and ips of the branches represent present American species, sub- as, and varieties, together with the Eurasian vicaniants of 11 species. Because extinction has certainly occurred many in the genus, there are bound to be gaps and missing branches. :heless, this scheme gives an approximate picture of the path— leading to the present taxa of North and South America based on resent study. AA tanning gent m. for sh water, moistL from : Culpe den i 83th< than ”the Acen 00h Pro a d Ufa th ECONOMIC IMPORTANCE .lnus has long been used by man, especially for lumber, in .g and dyeing, for the production of charcoal, and as an astrin- 1edication. It was important to the ancient Greeks and Romans ,ipbuilding, and medieval builders, seeing that alders grew near concluded that the wood of this tree must be resistant to ,re and constructed much of Venice and Amsterdam on piles made ts trunks. mong the more obscure uses for alders is one reported by Nicholas er in 1653, who writes that "leaves gathered while the morning on them, and brought into a chamber troubled with fleas, will them thereonto, which being suddenly cast out, will rid the E of those troublesome bedfellows." Coon (1963) states that igs, bark, and catkins are a source of a black dye that has own and used for centuries. Used alone, it dyes wool a reddish nd used with copperas, produces a good black. Young shoots colors from yellow to cinnamon, while the leaves will produce or leather." Other uses of élflfli in EurOpe, including the man- e of wooden shoes and clog soles, are reviewed by Eldin (1964). though seldom a major timber source in America, élflE§.iS never— important for a variety of wood products. In the Pacific st the wood of Alnus rubra is used in cabinetry as well as for 196 'Iz=fl?==*‘fi‘ the manufz boxes, an: also an i into its years (cl uses thr( (1939) s ture mak availabl Calupf, it is u: eXporte Co North A Small 0f siz and ”t hitheI class labor sPeci dlffr 197 anufacture of toys, trays, brush handles, spools, shoe soles, , and other small items (cf. Worthington SE 31., 1962). It is an important source of pulp for paper in this region, and research its productivity, management, 232° has been increasing in recent (cf. Smith, 1968; Williamson, 1968). A, acuminata finds similar :hroughout its range from Mexico to Argentina. Acosta-Solis A states that in Ecuador alder wood ”is used in carpentry, furni- iaking, and cabinet work, and before Eucalypt lumber became 1ble, was employed in general construction in the same way as 5, Sauce, Algarrobo, and Arrayan. In the province of Tungurahua used for making the boxes and packing cases in which fruit is zed." iommenting on the shrubby stature of the alders of eastern America, Michaux (1859) states that ”the common alderl is too to be applicable to any use in the arts: from its inferiority 1e, it will probably one day give place to the European Alder" +e dwarfish stature of all the species of Alder that have to been discovered in North America excludes them from that f vegetation to the description of which I have restricted my ; but I could not forbear mentioning the two most remarkable . 2 . . . , of which one merits attention on account of its abundant 3 . . . . . on, and the other on account of a striking peculiarity in serrulata. serrulata. incana subsp. rugosa. ‘5:qu ‘ —‘ the c even Among regic sild. l9A3 bent den pot tur ter 198 :olor of its leaves." But in spite of their small size, the shrubby alders of the New World have important uses. ; these is their use as a source of firewood for the natives of )ns where other fuels are scarce, including parts of Alaska (Por— , 1939) and the high Andes of South America (Record and Hess, A. The ability of alders to fix atmospheric nitrogen may someday fit the forestry industry. According to Tarrant (1968), "evi- : from many studies indicates that alder (A3233 spp.) has a ltial relation to forestry similar to that of legumes to Agricul- ," Presently, however, members of the genus have not found ex- .ve use in this way in North America. The bark and foliage of AAAAA are very astringent and are used 'imitive American societies for medicinal purposes. Smith (1923, 1932, 1933) reports a number of such uses among American Indians, ,ding the making of poultices to reduce swellings and infusions Aas to treat sores, the passing of blood in stools, cases of piles, he flux. He (1923) states that "the white man has held valuable stringent properties in the treatment of diarrhea and haema- . The liquid has been used as a mouth wash or gargle in the ent of stomatitis and pharyngitis. When injected into the vagina, said to cure leucorrhaea." Standley (1920) mentions that in an infusion of bark is used as a lotion for cutaneous diseases at a decoction of the bark is sometimes taken internally for ls and venereal diseases. Martinez (1959) reports that the f A, 'orullensis var. firmifolia is sold in Mexican markets for such pur to cicat flamatic bleedin; (1960), of “int practit in a p. 0 dyeing bride, Steyei lives 0i so lollc of 11 Gem. mine 199 , purposes. In Peru, the leaves of alders are crushed with butter .icatrize wounds and also used without fat to protect against in— .ation. Applied to recent wounds, the leaves are used to stop ding (Macbride, 1936). In the United States, according to Vines 0), the bark of élflfli serrulata was formerly used in the treatment intermittent fever." Smith (1932) states that "the eclectic titioner in the United States and Canada employed it [the bark] powdered condition for dusting upon chafed body surfaces.“ Other native American uses of the alders include the tanning and ng of leather and textiles with preparations of the bark (Mac- e, 1936; Smith, 1923, 1932, 1933; Standley, 1920; Standley and ermark, 1952). In some regions alders are of limited value for stock forage. All of the species are important in the reduction oil erosion since AlEE§,iS among the first of the colonizers to 3w fires, lumbering, volcanic eruptions, and other disturbances ‘e natural vegetation. In Great Britain, the Netherlands, and ny, alders have been used for some time in the rehabilitation of spoils (Tarrant, 1968). “amp-€91“ _' In viation Stafleu tutions 0f Albe Museum Of Agr herbar Polyte (F);A (1ND) Nacio Méhit Gard. Notr Okla ium. TAXONOMIC TREATMENT In the citation of specimens, herbaria have been given the abbre- tions used in the fifth edition of IE}; Herbariorum (Lanjouw and fleu, 1956). Collections were examined from the following insti- ions: Arnold Arboretum of Harvard University (A); The University Alberta (ALTA); The University of Arizona (ARIZ); The National sum of Canada (CAN); Plant Research Institute, Canada Department Agriculture (DAO); Dudley Herbarium of Stanford University (DS); )ario de la Escuela Nacional de Ciencias Bioldgicas, Instituto 'tecnico Nacional, Mexico (ENCB); Field Museum of Natural History Gray Herbarium of Harvard University (GH); Indiana University A); Jepson Herbarium, University of California (JEPS); Herbario onal del Instituto de Biologfa de la Universidad Nacional de co (MEXU); The University of Michigan (MICH); Missouri Botanical en (MO); Beal-Darlington Herbarium of Michigan State University ); University of Notre Dame (ND); Greene Herbarium, University of e Dame (ND-G); New York Botanical Garden (NY); University of home (OKLA); Ohio State University (OSU); Rocky Mountain Herbar- University of Wyomong (RM); University of California, Berkeley ; National Museum of Natural History, Smithsonian Institution (US); arsity of Wisconsin (WIS); University of Washington (WTU). Approxi- .y 8,500 specimens were examined in this study. 200 Wan-1:" ‘ r The that spe cal taxc Pfl. Hi ovate gentl i conSp age s Delta brin Tedd ifer fixi stir min ovo C06 201 The symbol "*" before an herbarium abbreviation indicates . specimen was used in the collection of data for the numeri- taxonomic studies. ALNUS P. Miller Alnus Tournefort, Inst. Rei Herb. 1: 587. 1700; Ehrhart, Oekon. Hist. 2: 211. 1753; Miller, Gard. Dict., Abbr. ed. 4. 1754. Betula L., Sp. Pl. 2: 983. 1753, in part as to species 5. Betula-Alnus Marshall, Arbust. Am., p. 19. 1785. Deciduous monoecious trees and shrubs with narrowly to broadly : to obovate leaves, watery sap, smooth to (in age) scaly astrin— bark, and terete branchlets with triangular pith and (usually) .icuous circular to elliptic lenticels, the twigs, buds, and foli— .parsely to densely covered with simple straight hairs and minute ,te glands; sapwood white, soft, close, straight-grained, and le, becoming red when exposed to the air before dry; heartwood sh-brown, usually forming only a small core; roots often stolon— us, the rootlets fibrous, bearing nodules containing nitrogen- g endophytes. Leaf buds with 2 or 3 equal mostly valvate lar scales or 5 or more imbricate scales, raised on well-defined e stems (stipitate), elongate, often slightly three-angled, to oblong and acuminate, acute, or rounded at the apex, resin- 1; apical bud pseudoterminal. Leaves alternate, simple, pin- r—veined, singly- or doubly-serrate to nearly entire, petiolate, rous, shed while still green; in the bud enclosed in their ‘.-.-_—-.:>‘-* -.‘ “' stipu along l‘Ollm 202 lles, becoming conduplicate in expansion, plicately-folded ; the lateral veins. Leaf scars often elevated, obtuse to ,ed below and somewhat notched above, with three large approxi- y equidistant circular to crescent-shaped bundle scars, the t itself often obviously composed of three smaller scars. 1es ovate, elliptic, or obovate, acute to rounded at the apex, ular, glabrous to densely pubescent, ciliate-margined, deciduous derately persistent. Flowers unisexual, sessile, in modified as reduced and arranged into pedunculate bracteate aments, the :les from the axils of leaves or minute leafy bracts, opening 3 early spring before or with the unfolding of the leaves, or in summer or autumn, often partially developed during the previous 1g season and either exposed or enclosed in the bud during icy. Pistillate inflorescences ovoid to oblong or cylindric, pedunculate, formed in the axils of the leaves of a branchlet Iping in the axil of one of the upper leaves on the main axis the staminate inflorescences, appearing as a racemose cluster itary on the main stem; bracts subtended by and adnate with 4 oles, imbricate, and somewhat fleshy. Pistillate flowers 2 per without perianth. Pistil 1, compressed, 2- (or rarely 3-) late; ovary inferior, 2- (rarely 3—) locular below and 1-locular ovules 2, one per locule, axially attached near the summit of :ule, pendulous, anatropous; styles 2, linear, free, each tic near the apex. Pistillate inflorescences enlarging, the becoming thick and woody after anthesis; mature scales ob- 3- to 5-1obed or truncate at the apex, forming a persistent ‘_=.a'='-=‘. “ ‘- subgl Stami clusi shor tend 203 iglobose, ovoid, or cylindric strobilus—like infructescence. minate catkins elongate, pendulous, in one or more racemose sters or solitary in the axils of leaves or leafy bracts; bracts rt-stalked, peltate, adnate at the base to 3 or 5 bracteoles, sub— ding 3 (to rarely 6) minute flowers. Perianth of one series, (or infrequently l- to 6-) parted, the segments ovate, elliptic, obovate and connate at the base, glandular-margined; stamens 2 or occasionally 5 or 6), short (never much exserted from the catkin), arted at the base of the perianth parts and often basally adnate :hem; filaments short to long, undivided; anthers erect, dorsifixed, rorse, 4-sporangiate and 2-locular, the thecae parallel, partially irate, contiguous, dehiscing longitudinally. Fruit a small, :tened, ovate, elliptic, orbicular, or obovate nutlet or samara, Lted and crowned at the apex by the remnants of the styles, wing— . (with a narrow wing—like chartaceous border) or broadly membra- ous winged on 2 margins; pericarp of 2 coats, the outer thin and ranaceous, the inner thick and crustaceous. Seed solitary by tion, filling the cavity of the fruit, the hilum apical, testa ranaceous, without endosperm. Embryo large, straight, the cal superior and shorter than the flat, fleshy cotyledons. Type species: A3333 glutinosa (L.) Gaertn. (Betula 33333 a ElEEET L., Sp. Pl. 2: 983. 1753, lectotype). Al333 is the classical Latin name for the alder tree and for :ts made from its wood, including ships and boats. In this action it was used by Virgil, Pliny, and others. It is presumably red from the Latin verb alo (to nourish), referring to its usual clos the Phi 912 non the bh 204 )se association with water.1 The genus was generally considered as separate from Betula until : two were combined as Betula by Linnaeus in Species Plantarum (1753). .lip Miller was the first to resurrect A3333 in his Gardner's .tionary, Abridged, ed. 4 (1754), but since he did not use binomial .enclature one of his species cannot be designated as the type. In 8th edition of the Gardner's Dictionary (1768), where he first used omials, Miller ommitted the genus altogether, apparently partly by take, referring the reader first under "Alder-Tree" to A3333, and n under '33333fl to Betula, but not including any alder species re. The first treatment employing binomial nomenclature to list 33 species, excluding Hill's 333 British Herbal (1756), which is consistent in its use of binomials, as noted in the International 1‘33 Botanical Nomenclature (Stafleu et al., 1972), is Gaertner's I'ructibus 33 Seminibus Plantarum (1791). The only species of Alnus :d in this work is A, glutinosa (L.) Gaertn., based on Linnaeus' 33_alnus a glutinosa (Sp. Pl. 2: 983. 1753). .rtificial Key £3_the Subgenera, S ecies, and Infraspecific Taxa Winter buds stakled, covered (sometimes incompletely) by 2 or 3 equal stipular scales; leaf—bearing stems usually not forming both long and short shoots; staminate and pis- tillate inflorescences produced mid to late in the growing season, not with new growth in the spring ................. 2 1Wood (1845) states that it is derived from the Celtic 33 T) and 1an (riverbank), also referring to its usual habitat. 205 Lateral veins of the leaves terminating in teeth at the margin; pistillate inflorescences (and later infructes- cences) borne on short branchlets in racemose clusters; flowering occurring at the beginning of the growing season (spring). Subgenus Alnus ....................... 3 3. Leaves mostly ovate (rarely elliptic), finely serrate or serrulate to rather coarsely double-serrate ...... 4 4. Leaf margins finely and evenly serrate or serru- late, sometimes slightly lobed; trees of the west- é. rhombifolia ern United States .................. 3. 4. Leaf margins serrate or double—serrate to irregu- larly-toothed .................................... 5. Margin of leaf blade strongly revolute; large trees of northwestern coastal North America ... l. A. rubra oon.o.00.00....con-gnoooooooooo- 5. Margin of leaf blade flat or only slightly to moderately revolute; trees and shrubs ......... 6 6. Leaves lanceolate to narrowly ovate or ovate ...................................... 7 7. Major teeth of the leaves sharp and acuminate, usually standing out well 8 above the secondary teeth ............... 8. Internodes, petioles, and lower leaf surfaces and veins glabrous; stami~ nate flowers with 4 stamens and peri— anth parts; trees of central and 7. 206 southern Mexico ...................... ........... 4b. A, acuminata var. glabrata 8. Internodes, petioles, and lower leaf surfaces and veins usually at least sparsely pubescent, often villous to velutinous; staminate flowers usually with 2 large and 2 smaller perianth parts and stamens, 2 of the stamens often absent; trees of the south— western United States and adjacent Mexico ................ 2. A, oblongifolia Major teeth of the leaves acute to obtuse, short to long ............ 2. A, oblongifolia 6. Leaves moderately to broadly ovate, the major teeth obtuse to rounded .................... 9 9. Leaves usually large, the blade 5-19 cm long, relatively finely double-serrate or serrate, the apex usually acuminate, densely glandular below; infructescences 11-45 mm long; bark usually with trans- verse constrictions or ridges; trees of Mexico, Central America, and South Ameri- Ca .......... 4a. A. acuminata var. 353333353 Leaves usually smaller, the blade 4-10 cm long, more coarsely double-serrate, the apex usually acute to obtuse, sparsely to 207 only moderately glandular below; in- fructescences 10—17 mm long; bark with- out transverse constrictions; large shrubs and small trees of Canada and the northern and western United States ...... 10 10. Leaf blade moderately thick, the major teeth acute; large shrubs of eastern Canada and the northeastern United States ....................... ............. 6a. A, incana subsp. rugosa 10. Leaf blade thin and papery, the major teeth usually rounded; large shrubs or small trees of the western United States and Canada ................... ......... 6b. A, incana subsp. tenuifolia 3. Leaves mostly elliptic, oblong-elliptic, or obovate, occasionally tending toward ovate ................... 11 11. Leaf apices rounded to retuse; moderately large trees naturalized in the northeastern United States and adjacent Canada ................. 8. A:.&l2£i§2§i 11. Leaf apices acute to obtuse or sometimes slightly rounded; trees and shrubs of the United States, Canada, Mexico, and Central America ............. 12 12. Leaf margins finely and evenly serrulate; leaf texture papery to slightly leathery ....... 13 13. Leaf blades broadly elliptic to obovate; 13. 208 staminate flowers with 4 stamens; shrubs of eastern North America ......... 7. A, serrulata Leaf blades narrowly elliptic or rhombic; stamens 2 (or 4 with 2 reduced in size); trees of the western United States ........ ............................. 3. A, rhombifolia 12. Leaf margins rather coarsely and unevenly toothed or wavy, especially near the apex; leaf texture very leathery; trees of Mexico and Central America .......................... 14 14. 14. Lower leaf surface densely covered with bright yellow glands ...................... ........... 5a. A, jorullensis var. jorullensis Lower leaf surface with relatively few pale whitish, yellowish, or brownish, or dark brown glands ..... 5b. A, jgrullensis var. firmifolia Lateral veins of the leaves usually terminating by anastomosing with other veins near the margin (infrequently ending in the teeth); pistillate inflorescences (and later infructescences) solitary in leaf axils along main stems; flowering occurring near the end of the growing season (late summer or autumn). thropsis ................................... 10. Subgenus Cle- A. maritima Winter buds sub-sessile (stalks not over 1 mm long), covered by 5 or more unequal imbricate scales; leaf— 5 bearing stems usually with both long shoots and short 209 spurs, the latter bearing the leaves; staminate inflorescences produced late in the previous growing season, pistillate inflorescences pro- duced along with the first new growth of the season. Subgenus Alnobetula .............................. 15 15. Leaves finely serrate or serrulate (rarely double— serrate), leathery, dark green, and sometimes sparsely to densely pubescent below; shrubs of the eastern United States and and adjacent Canada, . and northern Canada to Alaska ...................... ........................... 9a. A, viridis subsp. crispa 15. Leaves moderately to coarsely double-serrate (rarely singly or finely serrate), thin to membranaceous, light or yellowish green, glabrous; shrubs of the mountainous western United States and Canada ....... .......................... 9b. A, viridis subsp. sinuata lnus subgenus Alnus AABEE sect. II. Clethra W.D.J. Koch, Syn. F1. Germ. Helvet., . 663. 1837. Lectotype: A, glutinosa (L.) Gaertn. Algg§_sectio II. Gymnothyrsus Spach, Ann. Sci. Nat. ser. 2, 15: 04. 1841; Alggs subgenus IV. Gymnothyrsus Regel, Bull. Soc. Nat. osc. 38(3): 425. 1865. Lectotype: A, glutinosa (L.) Gaertn. Al23§_sectio I. Phyllothyrsus Spach, Ann. Sci. Nat. ser. 2, 15: 4. 1841; Alnus sect. Gymnothyrsus subsect. Phyllothyrsus erepanov, Notul. Syst. Herb. Inst. Bot. Rom. Acad. Sci. U.R.S.S. 17: 210 98. 1955. Lectotype: A, acuminata H.B.K. Alnus subgen. Euclethrus Petermann, Deutschl. F1. 516. 1849. Lectotype: A, glutinosa (L.) Gaertn. A1325 sectio II. Betulaster Regel, Mem. Soc. Nat. Mosc. 13(2): 144. 1861. Type: A, lindeni Regel (= A, rhombifolia Nutt.). A1223 sectio IV. Eualnus Regel, Mem. Soc. Nat. Mosc. 13(2): 152. 1861. Lectotype: A, glutinosa (L.) Gaertn. Alnus sectio III. Pseudalnus Regel, Mem. Soc. Nat. Mosc. 13(2): 145. 1861. Lectotype: A, acuminata H.B.K. Alnus sect. Pycnantha Muller, Madrofio 5: 152. 1940. Type: A. densiflora Muller (= A, incana subsp. tenuifolia (Nutt.) Breitung). Alnus sect. Gymnothyrsus subsect. Hedroiostachys Czerepanov, Notul. Syst. Herb. Inst. Bot. Kom. Acad. Sci. U.R.S.S. 17: 101. 1955. Type: . glabrata Fern. (= A, acuminata var. glabrata (Fern.) Furlow). Algg§_sect. Gymnothyrsus subsect. Phyllothyrsus ser. Acutissimae zerepanov, Notul. Syst. Herb. Inst. Bot. Kom. Acad. Sci. U.R.S.S. 7: 99. 1955. Type: A, acutissima (Winkl.) Call. (= A, acuminata H.B.K. ar. acuminata). Alggg sect. Gymnothyrsus subsect. Phyllothyrsus ser. Ferrugineae zerepanov, Notul. Syst. Herb. Inst. Bot. Kom. Acad. Sci. U.R.S.S. 7: 99. 1955. Type; A, ferruginea H.B.K. (= A, acuminata H.B.K. var. cuminata). Alggg sect. Gymnothyrsus subsect. Podostachys Czerepanov, Notul. yst. Herb. Inst. Bot. Kom. Acad. Sci. U.R.S.S. 17: 99. 1955. pet A? serrulatoides Call. Alnus sect. Gymnothyrsus subsect. Podostachys ser. Glutinosae 211 Czrepanov, Notul. Syst. Herb. Inst. Bot. Kom. Acad. Sci. U.R.S.S. 17: 100. 1955. Type: A, glutinosa (L.) Gaertn. AAEEE sect. Gymnothyrsus subsect. Podostachys ser. Jorpllenses ‘Czerepanov, Notul. Syst. Herb. Inst. Bot. Kom. Acad. Sci. U.R.S.S. 17: 101. 1955. Type: .5“ jorullensis H.B.K. éiflfli sect. Gymnothyrsus subsect. Podostachys ser. Rhombifoliae Czerepanov, Notul. Syst. Herb. Inst. Bot. Kom. Acad. Sci. U.R.S.S. 17: 100. 1955. Type: A, rhombifolia Nutt. AAEEA sect. Gymnothyrsus subsect. Podostachys ser. Rubrae Czere- panov, Notul. Syst. Herb. Inst. Bot. Kom. Acad. Sci. U.R.S.S. 17: 100. 1955. Type: A, EEEEE Bong. Alggg sect. Proskeimostemon Czerepanov, Notul. Syst. Herb. Inst. Bot. Kom. Acad. Sci. U.R.S.S. 17: 102. 1955. Type: A, hirsuta Turcz. (= A, incana (L.) Moench). AAEEA sect. Proskeimostemon ser. Incanae Czerepanov, Notul. Syst. Herb. Inst. Bot. Kom. Acad. Sci. U.R.S.S. 17: 104. 1955. Type: A. incana (L.) Moench. Aiggg sect. Proskeimostemon ser. Hirsutae Czerepanov, Notul. Syst. Herb. Inst. Bot. Kom. Acad. Sci. U.R.S.S. 17: 103. 1955. Type: A, hirsuta Turcz. (= A, incana (L.) Moench). Alggg sect. Glutinosae Murai, Bull. Gov. For. Expt. Sta. Jap. :54: 63. 1963, ESE: 329'? élEEE subgen. Gymnothyrsus sect. Glutinosae lurai, Bull. Gov. For. Expt. Sta. Jap. 154: 67. 1963, 232, §y2,; A33§_subgen. Gymnothyrsus sect. Glutinosae Murai, Bull. Gov. For. xpt. Sta. Jap. 171: 50. 1964, BEE: All_g, (Latin description or iagnosis not provided). 212 Large shrubs and small to large trees; twigs and young branches not differentiated into long and short shoots; bud stalks well developed; buds covered by 2 (or sometimes 3) equal stipular val- vate scales. Leaves coarsely to very finely double—serrate or ser— rulate; lateral veins ending in major teeth at the margin. Pistillate inflorescences borne on short, stout peduncles, usually without sub- tending leaves, in racemose clusters on short branchlets along a major branch, the latter bearing the staminate catkins at the upper nodes, usually without subtending leaves, in one or more racemose clusters. Pistillate and staminate inflorescences both formed during the previous growing season and exposed during the dormant period (where present); anthesis occurring in the spring before new growth commences; fruits maturing at the end of the current growing season; staminate flowers with 4 (or occasionally 2) stamens; perianth parts 4, 2 sometimes reduced in size. Fruits lacking Wings or merely narrowly wing-margined on two sides. 1. Alnus rubra Bongard AAEE§_£EB£A Bongard, Mem. Acad. Sci. St. Petersb. ser. 4, 2: 44. 1833; AAAEE incana n £3233 Regel, Mem. Soc. Nat. Mosc. 13(2): 157. 1861. Type: ”the island of Sitka" (LE?, not seen). £1222 castaneaefolia Douglas ex Hooker, Fl. Bor. Amer. 1: 158. 1838, 223 Mirbel, Mem. Mus. Hist. Nat. 14: 463. 1827, p£2_§yg, Algggloregona Nuttall, North Amer. Sylva l: 44. 1842. Type: luttall ELEL: "on the borders of the rivers Boisée and Brulée, which ass into the Shoshonée not far from Walla-Walla, and at intervals it 213 continues more or less common to Point Chinhook, near the shores of the Pacific (BM7, not seen). Al333 maritima hort. ex Wetzel, Bot. Archiv 25: 264. 1929, 323 Spach, Ann. Sci. Nat. ser. 2, 15: 206. 1841, £52,313. A1333 washingtonia hort. ex Wetzel, Bot. Archiv 25: 264. 1929, REE.§X2: A1333'333£A_var. pinnatisecta Starker, Jour. For. 37: 415. 1939; Al223_333£3 f. pinnatisecta Rehder, Bibliog. Cult. Trees and Shrubs, p. 104. 1949. Type locality: ”T. Norman Nelson farm, 16 miles northwest of Portland, Ore." (original material not seen). Narrow, somewhat pyramidal—crowned trees, up to 28 m in height; trunks usually several, erect, up to 1.5 m in diameter, bark thin, gray, whitish, or irregularly mottled, smooth to slightly rough with small bumps and irregularities, with inconspicuous lenticels when young, broken into shallow plates on older individuals; young stems medium to dark red-brown, dull to moderately lustrous, not glaucous to slightly glaucous, usually without a heavy resinous coating, not differentiated into noticeable long and short shoots, often with prominent longitudinal ridges originating at the nodes; internodes usually glabrous, moderately glandular, glands small to medium in size, brown or dark brown; nodes and stems bearing inflorescences densely glandular; lenticels oval to elongate, 0.5-1.2 mm long, 0.3—1.1 mm wide, whitish to yellowish, moderately prominent; leaf scars 1.5—2 mm high, 2-4 mm wide, the bundle scars moderately prominent. Buds ellipsoid, acute to slightly rounded at the apex, heavily resin-coated; stalk 2-8 mm long, 1.2-2.5 mm in diameter, 214 glabrous, densely glandular; body 6-10 mm long, 3-5 mm in diameter; scales 2 (-3), stipular, equal, more or less valvate, glabrous to moderately pubescent, glandular; pubescence and glands usually ob- scured by the heavy coat of resin. Leaves ovate to elliptic; apex acute to obtuse; base broadly cuneate to rounded; blade (3.5-) 6-11 (-15) cm long, (2.5-) 3.5-7 (—9.5) cm wide, medium to dark green and dull to moderately lustrous above, light to medium green or green-brown and dull below, coriaceous; margin strongly revolute, double-serrate or crenate; major teeth (3-) 5-15 (~25) mm apart, 2-6 mm deep (up to 35 mm deep in forma pinnatisecta (Starker) Rehder), regular; secondary teeth 3-6 per cm, 0.2-1 mm deep, regular; abaxial surface glabrous to sparsely pubescent, sparsely to moderately glandular; adaxial surface and veinlets glabrous to sparsely pubescent, densely to very densely glandular, slightly to moderately resin-coated; major ‘veins and vein axils near the base usually moderately pubescent to tomentose or wooly-pubescent (rarely only sparsely pubescent); pubes- cence whitish to yellowish; glands small to medium in size, whitish to yellowish (rarely brownish). Lateral veins 9-14, 4—11 mm apart at id-leaf, straight or slightly ascending, usually not branching again, terminating in major teeth at the margin; cross veins between lateral eins poorly to well developed. Petioles (7-) 8-18 (-22) mm long, .7-2 mm in diameter, sparsely pubescent to moderately villous, oderately to densely glandular. Stipules ovate, elliptic, or ob- vate, the apex acuminate, acute, or obtuse, 6-8 mm long, 1.7-2.5 mm ide, green to light brown, glabrous to sparsely pubescent, moderately landular; hairs, when present, yellowish; glands yellowish. 215 Pistillate inflorescences borne in racemose groups of (3-) 4-5 (-8) on short branches diverging moderately to strongly from the main axis, produced during the previous growing season, erect, ovate to elliptic, at anthesis (3.5-) 5-7 mm long, 1.7-2.1 mm in diameter, on peduncles 1-2 mm long, 0.5—1.5 mm in diameter; staminate catkins borne in one or more racemose clusters of (2-) 3-6 at the end of the main branch above the pistillate inflorescences, produced during the previous growing season, pendent during dormancy and anthesis, at anthesis 3.5-l4 cm long, 6-10 mm in diameter; floral bracts 1-2 (-3) mm high, (1.5-) 2-3 mm wide. Staminate flowers 3 per bract; perianth of 4 parts, these elliptic to obovate, obtuse to rounded at the apex, 1.6-2.1 mm long, 1.2-1.6 mm wide, lined with minute glands at the margin; stamens 4, opposite and separate from or only slightly basally adnate to the perianth parts, usually appearing much longer than the peri- anth, filaments 0.8-1.6 mm long, anthers 1.3—1.8 mm long and 1.3-1.8 mm in diameter, thecae separate for 30-60% of their length. Mature infructescences ovoid (sometimes ellipsoid) or subglobose, (lO-) 14-25 (-34) mm long, (6-) 8-14 (—16) mm in diameter, on peduncles 0.5-7 (-10) mm long, 1-1.8 mm in diameter; scales 4-7 mm long, 4.5—6 nm wide at the apex, 1.7-2 mm wide at the base; apex moderately thick ‘ to very thick, flat, the terminal lobe truncate, not extended. Truits narrowly wing-margined, brown; body ovate or elliptic, 2-2.5 mn long, l-1.5 mm in diameter; wings 2.5-4 mm long, 0.5-1 mm wide, :hartaceous; persistent styles 0.5-0.7 mm long. Plates 2B, 8C, 10D, .5B, and 20. 216 Distribution and_Ecology: Along the Pacific coast from the southwestern corner of Alaska to central California; east along the Columbia River and its tributaries to eastern Washington and western Idaho. 0n rocky, gravelly, sandy, or humus stream banks and moist floodplains, lake shores, coasts, and open slopes, mostly below elevations of 300 meters (occasionally up to 1000 meters). Often with Acer, Seguoia, Pseudotsuga, Thuja, or Larix. Figure 20. Common Names: Red alder, Oregon alder, alder. Representative §pecimens: CANADA. British Columbia. North arm ‘ of Cowichan Lake, Allen s.n., Aug. 5, 1938 (cut-leafed) (UC); Lake Cowichan, Vancouver Island, Allen s.n., Aug. 6, 1938 (cut-leafed) (UC); 5 mi NW of Merville on road between Campbell River and Courte- nay, Calder é MacKay 28519 (*DAO); Bull Harbour, Hope Island of N end of Vancouver Island, Calder é MacKay 21211 (DAO, UC); 21 mi by road W of Terrace along road to Prince Rupert, Calder g£_313 14228 (DAO, NY, UC); E end of Mosquito Lake near Moresby Logging Camp, Moresby Island, Calder £5.2l3 21912_(DAO, NY, UC); about 2.5 mi E of Masset, Graham Island, Calder 3£_§l: 21222 (*DAO); just S of Lawnhill on road from Tiell to Skidegate, Graham Island, Calder 3E 31. 21131 (*DAO); Henslung Bay, Langara Island, Calder EE.§£: 22523 (DAO); small lake on road from Gray Bay to Sheldens Bay, Moresby Island, Calder 23 El: 2344: (DAO, NY, UC); S side of narrows near lead of Crescent Inlet, Moresby Island, Calder 23.3l! 24218 (DAO); Yale Mountain, Fletcher 3:2;J Aug. 5, 1883 (DAO); Bowen Island, ca. 5 li W of HorSeshoe Bay, Huber 1025 (*UC); Cheam Woods, Aqassiz, Ledingham Plate 20. 217 Representative specimen of Alnus rubra Bong. L__ 218 —1 Hum: of Nonbrrn (a... I... cl OIL-ul- Inmnourr calm Ctuucnu .. Jon”- v 1.... .. .(ui‘t 'd «7.4 Anna-.1“ In 1141/ .1119 0. V. ._ ‘ Plate 20. .._. . _r'. Figure 20. 219 Distribution of Alnus rubra Bong. Figure 20. 221 49—495 (DAO); Chilliwach Valley, between Lat. 49° and 49°10I and Long. 121025' and 122°, Macoun fl, June 13, 1901 (NY); Elk Lake near Victoria, Macoun _s_._n_._,‘Mar. 9, 1914 (*CAN); within 5 mi of Lillooet, Macoun _s_._n__._, July 2, 1916 (CAN); Hazelton, Skeena River, Macoun fl, June 24, 1917 (CAN); Sidney, Mflk’ Aug. 19, 1912 (CAN); British Columbia Forest Experiment Station, Lake Cowichan, Vancouver Island, Matthews s.n., June 26, 1940 (cut-leafed) (UC); Bella Coola, McCabe E (UC); Pulteny Point, Malcolm Island, McCabe fl (UC); Clam Cove, Nigei Island, McCabe 106—5 (UC); Stamp River, Alberni, Rhoades 14 (DAO); Dist. of Renfrew, Vancouver Island, Rosendahl éfl‘lfli (MSG, NY, RM, UC); Cowichan Station near Duncan, Vancouver Island, Wessel 59 (*DAO); Prospect Lake near Victoria, Young 11 (DAO). UNITED STATES. Alaska. Juneau, Anderson 62% (CAN, DAO, F, NY, *RM); 13 mi NW of Juneau on Glacier Highway, Afllfié Chunys 6_1_0_1_ (*CAN); Washington Bay, Kuiu Island, Eyerdam i345 (WTU); on the banks of Indian River in Sitka V National Monument, just E of Sitka, Heller l_49__3_2 (UC, WTU); Sitka, Trelease é Saunders _3_5_1i (M0). California. Alameda Co.: Berkeley, Bioletti s.n., Apr., 1894 (*UC); Berkeley, Blasdale s.n., Mar. 12, 1896 (*RM). Del Norte Co.: Crescent City, Applegate E (*F, UC); 2 mi S of Smith River, Furlow 321 (MSC); Gasquet, Mfl (UC); Humboldt Co.: 9 mi NE of Blue Lake, Furlow fl (MSC); 3 mi NW of Neafus Peak, Nelson 15$ (*UC); Trinidad, @294, July 26, 1941 (*DAO); near Carlotta, on Van Duzen River, Etc—sclfl (*UC); Trinidad, mesi (UC). Marin Co.: Inverness, Howell,w (NY, UC); W side of Mount Tamalpais, Lawfer EL (WTU); near Tomales Bay, .IEEi—a 2409 (RM, UC); Muir Woods, Rose 50032 (DAO). Mendocino Co.: 4 mi 222 NE of Point Arena, Cif_f_12 (UC); 13 mi N of Fort Bragg, @102 (UC); 4 mi E of Mendocino City, Nelson _10_3 (UC). San Mateo Co.: Kings Mountain, stream banks, coastal region, my; (RM); Tunitas, M26134 (*US); Tunitas, 303355—189 (*RM). Santa Cruz Co.: near mouth of cafion of Liddell Creek, Bacigalupi £._n_:_, March 29, 1922 (RM); Afitos, Dudley 3:11;, Oct. 13, 1902 (WTU). Idaho. Bonner Co.: Whiskey Rock Bay, Pend Oreille Lake, Johnson fl8_1_ (RM, UC). Latah Co.: Mos— cow, Becraft é Jahn 8363 (NY, UC); Marion, Henderson s.n., Apr. 21, 1905 (RM). Nez Perces Co.: Forest, Heller _g Heller i489 (DAO, MO). Oregon. Clatso Co.: Seaside, Demaree 13$]; (*NY). Columbia Co.: Drain, Applegate _Zfl (US); W of Tenmile, Steward 1482 (DAO, WTU). Coos Co.: Crawford Pt., Smi_th__3_§7_6 (UC). Curry Co.: streams about Brookings, Henderson 5692 (RM); Port Oxford, flw (NY); Chetco River Valley, M12883 (UC). Hood River Co.: without location, Henderson 212 (*MO). Lane Co.: Lowell Junction, on Noisy Creek, 1 mi above Dexter, Beattie _1_17_6§ (NY); Alderwood Wayside State Park, Furlow _29_6 (MSC); 1 mi W of Greenleaf, Furlow £8 (MSG); 8 mi SW of Cottage Grove, Furlow £9 (MSC). Lincoln Co.: without lo- cation, @1394! Aug., 1909 (NY). Linn Co.: near Fish Lake, Coville _&_ Applegate 215" (US). Multomah Co.: sandy river at Troutdale, 52213.8‘. Clausen fl (UC); without location, Matthews mm Dec., 1947 (cut-leafed) (DAO); Palmer, MM (NY). Tillamook Co.: Trask River, Merrill 1% (WTU). Washington Co.: Forest Grove, MEL: Nov. 15, 1893 (NY); Portland, Lunnell &, June 20, 1903 (*RM); T. Norman Nelson farm, Dixie Mountain, Matthews s.n., Oct. 3, 1940 (cut-leafed) (UC); farm of Mr. Nelson near Hillsboro, Nelson s.n., Illlllllllllllll |\|I| llll\l 1.“. _Illll V 223 Nov. 1, 1938 (cut-leafed) (NY, UC). Washington. Clallam Co.: Olympic Mtns., ElEE£.gZ§2 (NY). Clark Co.: Vancouver Lake, Sheldon 11689 (*UC). Grays Harbor Co.: Quinault, Conrad 14; (NY); 5 mi SE of Humptulips, Furlow 222_(MSC); 12 mi S of Aberdeen, Furlow 224 (MSC); Olympic Peninsula, along US rt. 101, near Quinault, Yuncker §_ngsh l§§§2 (UC). Island Co.: West Point, Deception Pass State Park, Whidbey Island, Sgéth'329_(WTU). Jefferson Co.: near Ruby Beach, Porter g Porter leg (DAO, RM); Olympic Peninsula, Hoh River Valley, Thomas E Thomas 2211 (DS). King Co.: Seattle, Eyerdam 1222 (F, *MO); North Seattle, Eyerdam s.n., Aug. 29, 1936 (*F, UC); Evans Creek Bog, 5 mi W of Redmond, J222§_§§§§ (WTU); near Seattle, Thompson 5128 (*WTU). Klickitat Co.: along creeks near Blugen, Suksdorf 224 (NY); by streams and springs, W Klickitat Co., Suksdorf 2184 (F, NY, UC). Lewis Co.: near Centralia, Palmer 21229 (NY). Mason Co.: Waketichie Creek, Jones 8501 (WTU); Shafer State Park, 15 mi N of Elma, Maas 641 (MSG). Pacific Co.: near Elk Horn Creek, 5.4 mi N of Raymond, Bartlett é Grayson 629_(NY); 2 mi N of Naselle, Furlow 225 (MSC); sapling beside tidal slough, T10N, R10W, 320, MacCleery §5_(MSC). Pierce Co.: along the White River, 6 mi NW of Cayuse Pass, Mount Rainier National Park, Furlow 229 (*MSC); 7 mi SE of Enumclaw, Furlow 222_(MSC); Puyallup, Porter 4291 (RM). San Juan Co.: Stuart Island, Lawrence 1§§_(WTU); Friday Harbor, San Juan Island, Zeller g Zeller 2§§_(NY). Skeminana Co.: E bank of Clearwater Creek, Columbia National Forest, Matthews s.n., Sept. 9, 1940 (cut-leafed) (UC); Columbia Forest, T8N, R6E, $14 at Clearwater Shelter, Wyas s.n., June 25, 1936 (cut-leafed) (US). Snohomish Co.: Catheart 224 Heights, Benson 1118 (WTU). Thurston Co.: along Perry Creek, 5 mi W of Olympia, Rogers 824 (UC, WTU). Whatcom Co.: scattered over the W slope of Lummi Island, Gaines 506 (NY, WTU). Whitman Co.: brook- side, Wawawai Canyon, St, John 5885 (NY). £122§.£2932 may be the most generalized species of the genus in America. It is more or less restricted to the mesic coastal fog belt along with Seguoia and other formerly more widespread taxa. Morphologically, this species seems little specialized in habit, leaf shape and size, cone size, and many other features, although it appears to be advanced in a few other characters (including the reduced fruit wings). Nuttall (1842), and later Kuntze (1891), recognized and commented on the fact that élEE§.£BP£2 bears strong affinities to the alders of Latin America. Such a relationship is indicated by the results of the experimental work reported above. In the western United States and Canada, this alder was formerly very widely known as A1322 oregona Nutt. The name Alnu§_£2b£a Tuckerman, Amer. Jour. Sci. 45: 32 (1843), based on Betulafélgu§_£ub£a Marshall, Arbust. Am. p. 20 (1785), is a later homonym applying to A, serrulata (Ait.) Willd. é1g2§_£ub£a is readily recognized by its size, gray mottled bark, and revolute-margined, coarsely-toothed, ovate leaves. The largest reported individual occurs in Polk 00., Washington, with a trunk circumference of 4.1 m, a height of 27.6 m, and a spread of 16.2 m (Pomeroy and Dixon, 1966). In two or three widely-separated locations a deeply cut-leafed form (A, rubra f. pinnatisecta (Starker) Rehder) has been seen growing along with individuals of the 225 ordinary kind. The depth of the leaf lobes in such plants is not the same in specimens from various trees or locations. Algu§_£ub£g_is found in varying habitats, ranging from exposed coastal bluffs to river floodplains and pond shores, sometimes form— ing great expanses of forest in low-lying areas of Washington and Oregon. It follows the Columbia River system eastward away from the Pacific Ocean and the coastal fog belt, reaching western Idaho in several disjunct populations (Johnson, 1968a, 1968b). The limiting factors responsible for the restricted distribution of this species are not well understood. Although it occurs in a rather moderate climate naturally, plants transplanted to an experimental garden were found to be hardy in the relatively severe winters in central Michi- gan. 2. Alnus oblongifolia Torrey Alnus oblongifolius Torrey in Emory, Rept. U.S. Mex. Bound. Surv. 2: 204. 1859; Alnus serrulata Y oblongifolia Regel, Bull. Soc. Nat. Mosc. 38(3): 432. 1865. Type: Wright 1864, "banks of the Mimbres and near Santa Barbara, New Mexico” (NY!, holotype; USI, isotype). Open, round-crowned trees up to 15 (-30) m in height; trunks usually several, erect, up to 1.5 m in diameter; bark light gray to brown and smooth when young, dark brown and broken into plates on old individuals, the lenticels usually inconspicuous on smooth trunks and branches; young stems medium red-brown, slightly to moderately lustrous, Slightly to moderately resin-coated, not differentiated into long and 226 short shoots, sometimes with slightly to moderately conspicuous longi— tudinal ridges originating at the nodes; internodes sparsely pubescent to velutinous, sparsely to moderately glandular; nodes and stems bearing inflorescences very densely glandular; hairs yellowish to brownish; glands small to medium in size, yellowish to brownish; lenticels of twigs circular to elliptic, 0.2~0.7 mm long, 0.1-0.5 mm wide, whitish, inconspicuous. Buds ovoid, slightly rounded at the apex, moderately to heavily resin-coated; stalk 1.5-4 mm long, l-l.5 mm in diameter, glabrous to sparsely pubescent, densely glandular; body 4-8 mm long, 1.5-4 mm in diameter; scales 2, stipular, equal, valvate or often incompletely covering the underlying unexpanded leaves, glabrous to moderately villous, glandular; pubescence and glands usually obscured by the resinous coat. Leaves narrowly ovate or lanceolate to elliptic (or occasionally rhombic); apex long to short acuminate or acute (rarely obtuse or rounded); base narrowly to broadly cuneate or rounded; blade (3—) 5—9 (all) cm long, (24) 3:6 (-7) cm wide, medium to dark green and dull (sometimes lustrous when young) above, medium green and dull to moderately lustrous below, chartaceous to moder- ately coriaceous; margin flat, slightly thickened, sharply to coarsely doublenserrate to serrulate; major teeth usually acuminate, standing well above the secondary teeth, 5-13 (e16) mm apart at mid-leaf, up to 5 mm deep, regular to irregular; secondary teeth 3-8 per cm, 0.5- 2 mm deep, slightly uneven to irregular; abaxial surface sparsely pubescent to rarely moderately villous (sometimes glabrous), moderately to densely glandular; adaxial surface and veinlets sparsely to moderately villous, moderately to densely glandular, slightly to 227 moderately resin-coated; major veins and vein axils near the base densely tomentose to wooly-pubescent; pubescence whitish to yellowish; glands small to medium in size, whitish to yellowish (rarely brownish). Lateral veins 9—13 (-15), 3-8 mm apart at mid-leaf, straight or slightly ascending, often branching once again, especially near the base, terminating in major teeth at the margin; cross veins between lateral veins poorly developed and usually not meeting. Petioles (3-) 7-18 (-22) mm long, (0.7-) 1-1.5 mm in diameter, moderately villous to velutinous, moderately to densely glandular. Stipules ovate to elliptic or obovate, acute to obtuse at the apex, 5-7 mm long, l-l.5 mm wide, green to light brown, glabrous to velutinous, the hairs yellowish, moderately glandular, the glands yellow or pale brown. Pistillate inflorescences borne in racemose groups of (2-) 4-5 (~7) on short branchlets not diverging strongly from the main axis, these sometimes subtended by leaves, several such groups also often Clustered together, produced during the previous growing season, erect, ovate to elliptic, at anthesis 4-5 mm long, 2-2.8 mm in diameter, on peduncles 1.5-5 mm long, l-l.5 mm in diameter; staminate Catkins borne in one or more racemose clusters of 3—6 at the summit Of the main branch above the pistillate inflorescences, produced during the previous growing season, pendent during dormancy and anthesis, at anthesis 3.2-8.5 (_10) mm long, 5—8 mm in diameter, on peduncles 3—13 mm long, 1-1.5 mm in diameter; floral bracts 1—2 (—3) mm high, (1.5-) 2-3 (—4) mm wide. Staminate flowers usually 3 per bract; perianth of 4 parts, these elliptical to obovate, rounded at the aPEX, 1.3-1.8 mm long, 0.8-1.2 mm wide, 2 frequently reduced in 228 size, the margins lined with minute glands or glands absent; sta- mens 2 or 4, if 4 then 2 frequently reduced in size, opposite and basally adnate to the perianth parts, usually appearing much longer than the perianth, the filaments l-l.3 mm long, the anthers 0.9-1.6 mm long, l-l.4 mm in diameter, the thecae separate for 50-70% of their length. Infructescences ovoid, ellipsoid, or cylindric, (9-) 15-24 mm long, 0.5-1.5 mm in diameter; scales 3-4 mm long, 2.5-4 mm wide at the apex, 0.5-1.3 mm wide at the base, the apex thin to moderately thick- ened, flat, the terminal lobe-tip acute, somewhat to very extended. Fruits narrowly winged, brown; body broadly elliptic to obovate, 1.8-3 mm long, 1.2-2.3 mm in diameter; wings 2.2-3.5 mm long, 0.5-1 mm wide, chartaceous; persistent styles 0.6-1 mm long. Plates 3A, 4A, 8A, 12C 9 18B, and 21. Distribution and Ecology: Central Arizona and west-central New Mexico south to south-central Chihuahua and northeastern Sonora. Occurring on sandy or rocky streambanks and adjacent moist slopes, often in mountain canyons, from elevations of about 1500 to 2300 meters (occasionally as low as 1000 meters). Often associated with ‘ ' ' 'n fairl Pinus, Quercus, Juniperus, or Pseudotsuga, sometimes formi g y dense groves. Figure 21. Common Names‘ Arizona alder, New Mexican alder, Mex1can alder, §—' aliso. Specimens Examined: MEXICO. Chihuahua. Sierra Charuco, o I {f0 Fuerte, Gentry 1530 (ARIZ, MEXU, MO, UC); Sierra Canelo, R10 Mayo, 3entry 2875 (ARIZ, F, UC); Balleza [Belleza] W of Parral, Engblggk Plate 21. 229 Right: holotype of Alnus oblongifolia Torrey. specimen of A, acuminata H.B.K. Left: T!" ' Lil/fungal: (Bu. 1:.- In]. 230 Plate 21. Figure 21. 231 Distribution of Alnus oblongifolia Torr. IO I20 IIO I00 Figure 21. ‘-'~.7..' a . . ‘7 233 768 (ENCB, MSC); Cuiteco, S of Creel near Rio Cuiteco, Knoblock 948 (*MSC); Batopilitas River, LeSueur 1304 (ARIZ, *F); La Pulvosa, l 1 28°11 N Lat., 108038 w Long., Martin 56017 (MICH); sw Chihuahua, Palmer s.n., in 1885 (NY, US). Sonora. Huchuerachi, Hartman 322 (*F, NY, UC, US); 6 mi N of Huachinera, Hastings 2 Turner 65:53 (*ARIZ); Santa Rosa Cafion, EEEES.3§2 (ARIZ, MICH); Cahon de la Escalera, mgfllfi (ARIZ, MICH); Ca’fion de Bavispe, Wh_it3_3_1_1_§ (MEXU, MICH); Cafion Internacional, Wh1£g_34§1 (F, *MICH); Canon Palpito, Mun. de Agua Prieta, Muller 3128 (*MICH). UNITED STATES. Arizona. Apache Co.: Bog Creek, 10 mi E of McNary, Goddard 681 (*UC). Coconino Co.: near Cathedral Rocks, 2+ mi below Sedona, Rossbach 2321 (UC). Gila Co.: Russell Canyon, Pinal Mountains, Ferris 1926; (UC); bank of East Verde River, at bridge, N of Payson, Foster é Arnold 198 (*US); Sierra Ancha, Jackson 13 (*US); without definite location, Johnson 8581 (*NY, UC, US); Workman Creek Falls, Sierra Ancha, Johnson £121, Sept. 3, 1957 (ARIZ); sandy banks of the East Verde River, near the bridge, Nelson_§ Nelson 1288 (RM); Pinal Mtns. near Miami, Peebles 33.213 2242 (ARIZ). Graham Co.: Wet Canyon, Graham Mts., Anderson g£_§13 192 (ARIZ); Wet Canyon, along stream, Pinaleno Mountains, EEEE 60-239 (ARIZ); Swift Trail at Wet Canyon, Pinaleno Mountains, 2335 68—569 (ARIZ); Mt. Graham, near Wet Canyon Recreation Area, Furlow 221 (*MSC); Mt. Graham below Wet Canyon Recreation Area, Furlow 251 (MSC); Graham Mountains, Kellogg s.n., May 15, 1902 (ARIZ); Graham Mountains, Swift Trail Road, Maguire 12122 (DAO, NY, WTU); Frye Canyon, above dam, Graham Mts., Moeller 403 (ARIZ); Mt. Graham, Peebles E£_21. 4398 (ARIZ); Pinaleno Mountains, Wet Canyon Campground, 234 Pinkava_g£.§1.l18211 (NY); Graham Mountains, Thornber 8 Shreve Z842 (*ARIZ); Graham Mountains, Thornber 8 Shreve 8811 (*ARIZ). Greenlee Co.: Clifton, A.T., 83381 888 (F, NY, *UC). Pima Co.: near Summer House, Santa Catalina Mts., Graham §;2;) Aug., 1908 (NY, UC); near Mud Springs, Mt. Lemmon Trail, Santa Catalina Mts., Harris Cl634l (NY); near Soldiers Camp, Santa Catalina Mts., Harris Cl6424 (NY); Mt. Lemmon, St. Catalina Mts., Lemmon 8 Lemmon 312;, July, 1915 (ARIZ, UC); Mt. Lemmon, Loomis g£_313 2188 (*ARIZ); Tucson, Neally 148 (NY); by streams of the Santa Catalina Mts., Pringle 342;, June 16, 1881 (F, NY); Rose Creek, Santa Catalina Mountains, Shreve 2818 (ARIZ, UC); Santa Catalina Mts., Sabino Cahon, Thornber 881 (ARIZ, UC); Santa Catalina Mts., at mouth of Sabino Cahon, Thornber 888 (ARIZ, UC); Santa Catalina Mts., Sabino Canyon, Thornber 8188 (ARIZ); Santa Cata- lina Mountains, Webber's Camp, Thornber 8 Kellogg 213;, Apr. 17, 1902 (ARIZ); Santa Catalina Mts., Toumey 312;, June, 1894 (ARIZ, NY, RM, UC). Beaver Creek, Fernow §;E;! Aug., 1896 (*US). Oak Creek, Pearson 888 (*US). New Mexico. Catron Co.: foothills of Mogollon Mountains, between Mogollon and Glenwood, KEPE£.§ Salamun 12844 (DAO, NY). Grant Co.: Cameron Canyon, Fort Bayard Watershed, Blumer 88 (NY); Gallings Creek at Devil's Backbone, W slope, Black Range, Gila Forest, Eggleston 18884 (MICH); Pinos Altos Mts., Greene E121, Oct. 11 (*F); near Silver City, Greene 2121, March and Oct., 1880 (M0); Pinos Altos Mts., 8222§_88888 (UC); Gila River bottom near Cliff, Metcalfe 778 (ARIZ, MO, NY, RM); Pinos Altos Mts., Mumford 879 (NY). Luna Co.: banks of the Mimbres and near Santa Barbara, Bigelow s.n., Without date (NY); without location, Wright 1864 (*NY, US). Sierra Co.: 235 Kingston, Metcalfe_1188 (ARIZ). Socorro Co.: in the Mogollon Moun- tains in or near the W fork of the Gila River, Metcalfe 888 (ARIZ, MO, NY, RM, *UC); Santa Magdalena Mts., MEL, June, 1881 (NY); near Halt's Ranch, in the Mogollon Mountains, Wooton 312;, July 20, 1900 (ARIZ, *RM). Pecos Cafion, Eastwood 18888 (*NY). Blue Creek, Wooton s.n., Aug. 13, 1902 (UC). Though it may have once been more widespread, Alnus oblongi- folia is today rather restricted in distribution relative to other species of the genus. In Sonora this species intergrades into ‘4. acuminata var. acuminata, and a case might be made to combine it with that taxon. Sargent, in his Silva 21 North America (1896), tentatively included it in Alnus acuminata pending the accumulation of additional specimens, noting that it appeared to be identical with certain of the Mexican material. In many ways, however, A. oblongifolia shows closer affinities with 4. rhombifolia, with which it is sometimes confused, including the presence of only two stamens in some cases, an occasional tendency to take on the serrulate and rhombic leaf form of the latter species, and the smooth bark (lacking the transverse constrictions of 4. acuminata). Its leaf shape varies, often on the same branch, from a distinctive sharp-toothed, lanceolate, acuminate-tipped form to elliptical or rhombic with very small teeth, reminiscent of 41283 rhombifolia, to a double-serrate ovate type similar to that of A, incana subsp. tenuifolia (which occurs in the same geographical region, though at a higher elevation). Identification is usually difficult, however, only with herbarium specimens having few leaves. 236 Alnus oblongifolia is found at relatively high elevations in the mountains of the southwestern United States and adjacent northern Mexico. The habitat of this species in Arizona is des- cribed in detail by Whittaker and Niering (1965) in their study of the vegetation of the Santa Catalina Mountains. This species, like Alnus rubra, is rather generalized in morphology and also occupies a mesophytic habitat. Both taxa are probably rem- nants of wider ranging species which existed during cooler periods. 3. Alnus rhombifolia Nuttall Alnus rhombifolia Nuttall, North Amer. Sylva l: 49. 1842; Alnus rhombifolia var. ypica Callier, Fedde Rep. Sp. Nov. 10: 229. 1911. Type: Nuttall s.n., "in the vicinity of Monterey in Upper California" (BM?, not seen). Alnus g1utinosa 6 serrulata lusus d. californica Regel, Mem. Soc. Nat. Mosc. 13(2): 166. 1861. Type locality: "aus Californien" (original material not seen). Alnus rhombifolia var. ovalis Winkler, Pflanzenreich 19(4.61): 115. 1904. Type locality: "Californien" (original material not seen). Alnus californica hort. ex Winkler, Pflanzenreich 19(4.61): 115. 1904, pro syn. Alnus rhombifolia var. bernardina Munz & Johnston, Bull. Torr. Bot. Cl. 52: 222. 1925. Type: Munz 8_Johnston 8468, junction of South Fork and Santa Ana River, San Bdno. Mts., San Bernardino Co., California (POM, holotype; Fl, NYl, isotypes). 237 Wide, round, open-crowned trees up to 25 (-35) m in height; trunks usually several, erect, up to 1 m in diameter; bark light gray, whitish, or irregularly mottled, smooth to slightly rough, with inconspicuous lenticels when young, brown and broken into scales on old individuals; young stems light green to red-brown, slightly to moderately lustrous, slightly to moderately glaucous, lightly to mod- erately resin-coated, not differentiated into long and short shoots, usually without longitudinal ridges; internodes glabrous to moderately villous, sparsely to moderately glandular; nodes and stems bearing inflorescences very densely glandular; hairs whitish to yellowish; glands small, yellowish to brownish; lenticels of twigs circular to elliptic, 0.1-0.7 mm long, 0.1-0.3 mm wide, whitish, inconspicuous; leaf scars 1.7—2 mm high, 1.5-2.5 mm wide, the bundle scars moderately prominent. Buds ellipsoid to obovoid, slightly rounded at the apex, moderately to heavily resin-coated; stalk 3-5 mm long, 1.2-1.5 mm in diameter, glabrous to sparsely pubescent, moderately to densely glandular; body 3-9 mm long, 2-3 mm in diameter; scales 2, stipular, equal, mostly valvate, often incompletely covering the underlying organs or even apparently absent from the buds nearest the apex of the stem, glabrous to moderately villous, glandular; pubescence and glands often obscured by the heavy resin coat. Leaves ovate, elliptic, r rhombic; apex acute, obtuse, or rounded (rarely acuminate); base roadly cuneate to rounded; blade (3-) 4.5-8.5 (-l3) cm long, (1.5-) .5-4.5 (-7.5) cm wide, medium to dark green and dull (sometimes ustrous when young) above, light to medium green and dull to moder- tely lustrous below, chartaceous to coriaceous; margin flat, not 238 thickened, finely serrate, serrulate, or (rarely) double-serrate (mainly on very vigorous shoots), sometimes slightly lobed; major teeth 5-8 (-10) mm apart at mid-leaf, up to 3 mm deep, regular to irregular; secondary teeth 4—9 per cm at mid-leaf, 0.7-1.5 mm deep, slightly uneven to irregular; abaxial surface glabrous, sparsely pubescent, or moderately villous, moderately to densely glandular; adaxial surface and veinlets sparsely pubescent to velutinous, moderately to densely glandular, slightly to moderately resin—coated; major veins and vein axils near the base densely tomentose to wooly- p L t; r L " whitish to yellowish; glands small to medium in size, yellowish to brownish. Lateral veins 9-12 (-15), (2-) 4-7 (—10) mm apart at mid-leaf, straight, often branching once again, especially near the base, terminating in major teeth at the margin; cross veins between lateral veins usually poorly developed. Petioles (3-) 7—15 (-23) mm long, 0.7—1.2 (-l.8) mm in diameter, moderately villous to velutinous, moderately to densely glandular. Stipules mostly elliptic, the apex acute, 6-11 mm long, 2-2.5 mm wide, green to light brown, glabrous to moderately villous, the hairs yellowish, oderately glandular, the glands yellowish. Pistillate inflorescences orne in racemose groups of 3-6 on short non-strongly-divergent ranchlets, produced during the previous growing season, erect, ovate o elliptic, at anthesis 3—6 mm long, 1.5-2 mm in diameter, on pe- uncles 0.5-1.5 mm long, 0.6-1 mm in diameter; staminate catkins orne in one or more racemose clusters of 3-7 at the end of the main ranch above the pistillate inflorescences, produced during the revious growing season, pendent during dormancy and anthesis, at 239 anthesis 3-10 cm long, 4-7 mm in diameter, on peduncles 2-12 mm long, 1-1.5 mm in diameter; floral bracts 1-2 (-3) mm high, (1.5-) 2-3 (-3.5) mm wide. Staminate flowers 3 per bract; perianth of 4 parts, these elliptic or obovate, the apex obtuse to rounded, 0.9—1.7 mm long, 0.4- 1.1 mm wide, 2 frequently reduced, the margin lined with minute glands or glands absent; stamens usually 2 (occasionally 4, but if so 2 reduced), opposite and basally adnate to the perianth parts, 1 usually appearing much longer than the perianth; filaments 0.8-1.7 mm long, anthers 1.1—1.9 mm long and 1.2-1.7 mm in diameter, the thecae , separate for 35-45% of their length. Infructescences ovoid, ellipsoid, \ or cylindrical, 10-17 (-22) mm long, (6-) 7-9 (-10) mm in diameter, l on peduncles 0.2-7 (-10) mm long, 0.7-1.2 mm in diameter; scales 3-3.5 mm long, 3.5-4.2 mm wide at the apex, l-l.2 mm wide at the base, the apex moderately thickened, flat, the terminal lobe-tip acute to rounded ‘nd somewhat extended. Fruits narrowly wing—margined, brown; body roadly elliptic, 2-2.5 mm long, 1.5-2 mm in diameter; wing margins .5-3 mm long, 0.1—0.3 mm wide; persistent styles l-l.5 mm long. lates 3C, 70, 9C, 15A, 22, and 23. Distribution and Ecology: Southern Washington and adjacent estern Idaho southwest to northern California and south to the Mexi- an border. On rocky streambanks and adjacent slopes from near sea evel at the coast to elevations of over 2300 meters inland. Often ssociated with Pinus, Quercus, and Abies. Figure 22. Common Names: Alder, white alder, California alder, mountain lder, western alder. Plate 22. 240 Specimen of Alnus rhombifolia Nutt. Isotype of rhombifolia var. bernardina Munz & Johnston. Alnus 241 flmwr: ‘I'F- ' uncnco mum I l l I l msmm IUSEUI .. , ' NEGATIVE no, cenllmeler! 051535 11 I21 1’! [“1 l5 8 Pomona College Herbarium PLANTS OF SOUTHERN CA use... 4",, L'l'c' New“ m.a.~h4. ' grist... q M 4...; all 3m a... «2..., Lia rum ML. .3» Bernardino. Date. 1.3 S)» , «1,4 (- 00 R. Coll. Philip A. Munz.~ J)" was. No. 8455- Plate 22. Plate 23. 242 Representative specimen of Alnus rhombifolia Nutt. / Art as" 1‘" " Wm, / ., » a V2 ... S fiflMNuruu‘) Plate 23. CALIFORNIA \‘icinixy of]; Grunge. Snnisllus County 1‘ rhombifolia Yutt. folia, Pcrry .\H:n flowers Augufi. 17, 1951 anunry 25. 196? Figure 22. 244 Distribution of Alnus rhombifolia Nutt. Figure 22. 246 Representative Specimens: UNITED STATES. California. Alameda Co.: along Redwood Creek, NE of Redwood Peak, Constance_888 (UC); Bay of San Francisco, without collector or date (herbarium of the U.S. Exploring Expedition under the command of Capt. Wilkes) (NY). Amador Co.: (Pyramid Pk., Gifford 848 (UC); 1% mi W of Plymouth, Nordstrom 188 (UC). Butte Co.: cahon of Big Chico Creek E of Chico, Heller 11181 (NY); Cherokee Mine, Table Mt. NE of Oroville, EE§22.21£§ (UC). Calaveras Co.: near Gardner, Hall 8 Chandler 4779 (UC); wet ravine near Murphy, Redfield 848 (NY, UC). Colusa Co.: Big Stoney Creek near Stoneyford, Sharsmith 4818 (UC). Contra Costa Co.: Alamo Canyon, Mount Diablo, Bowerman 1888 (UC); Marsh Creek, 6 mi E of Clayton, EEES.§§§QZ (UC, WTU). Del Norte Co.: Kelley's Flat, Dar- lingtonia, Smith River, EEEEE.§.£§£E§.2&Q§£ (RM). El Dorado Co.: mouth of Rock Canyon, Jensen 200 (UC); Whitehall, Johannsen 8 Gifford 57 (UC). Fresno Co.: under br. at Bald Mill Crk. on RR Grade Rd., a. 2700 ft, on WNW facing slope above San Joaquin River, Quibell 8881 NY). Glenn Co.: Alder Springs in the Coast Range, Heller 18188 NY, UC, WTU). Humboldt Co.: along the Trinity River 7 mi SE of illow Creek, Furlow 884 (MSC); Van Duzen River, 5 mi above Carlotta, rac 5306 (UC). Kern Co.: Stanley Smith Ranch, S fork of Kern iver, Onyx, Voegelin 88 (UC). Lake Co.: Snow Mtn., Blankenship s.n., ug. 3, 1929 (M0). Los Angeles Co.: San Antonio CaHon near Claremont, aker 8881 (F, MO, NY, RM, US); Little Dalton Canyon, San Gabriel Mts., am bell 81_(*DAO). Madera Co.: 4 mi N of Madera, Benson 8888 (WTU); adow Valley, Cockrell 8414 (NY, UC). Marin Co.: Lagunitas Canyon, well s.n., May 12, 1940 (NY, UC). Mariposa Co.: along the Merced ‘m' :1- =._.‘~w V 247 River, 5 mi W of Yosemite Village, Yosemite National Park, Furlow 307 (MSC); Mariposa, Jussell 88 (UC). Mendocino Co.: by streams, Ukiah, Pringle s.n., Aug. 14, 1902 (F); 12 mi N of Leggett along US 101, along S fork of Eel River, Stevens 1818 (MSC); Willits, 1:881 8188 (UC). Merced Co.: 2 mi SW of Snelling, Nordstrom 888 (UC). Monterey Co.: Carmel River, 12 mi above Carmel, Applegate 1818 (DAO, RM, UC); along Arroyo Seco, Santa Lucia Memorial Camp, Santa Lucia Mts., Ewan 9028 (UC, WTU). Napa Co.: White Sulphur Springs, SW of St. Helena, Jepson_§;p;, Nov. 26-27, 1895 (JEPS). Nevada Co.: W of Greenhorn Creek, T16N, R9E, S24, 88X88_8814 (F). Orange Co.: Tucker Wildlife Sanctuary, sapling growing in a wash in Harding Can~ yon, £222§.§;2:J Jan. 12, 1972 (MSC). Placer Co.: without location, Carpenter E;E;: Aug.-Oct., 1892 (UC). Plumas Co.: near mouth of Little Grizzly Creek below Genessee, Heller 8 Kennedy 8841 (NY). Riverside Co.: along stream, Snow Creek, N base of San Jacinto Mts., 322E 2&9; (*WTU); Whitewater, BBEE.§§QQ§.(NY’ *WTU). San Benite Co.: % mi S f Cement Works, T133, R4E, Axelrod 640 (UC). San Bernardino Co.: 0 junction of South Fork and Santa Ana Rivers, San Bdno. Mts., Munz 8 ohnston 8468 (F, *NY); mouth of Mill Creek, Munz_8 Johnston 8679 (*NY); an Bernardino Valley, Parish 8 Parish 542 (F, *JEPS, UC). San Diego 0.: near mouth of San Diego River, Mearns 4038 (UC); on bank of San elipe Creek above ranch house of San Felipe Ranch, Wiggins 2031 (*WTU). an Mateo Co.: creeks near Searsville Lake, Demaree 7297 (M0); Corte adera Creek, Lewis s.n., May 4, 1908 (WTU). Santa Barbara Co.: San ogue Canyon, Santa Barbara, Pollard s.n., Jan. 5, 1958 (DAO). Santa lara Co.: Stanford University, Abrams 2257 (M0, NY, UC); Los Trancos 248 lreek, back of Stanford University, Keck 1357 (F, NY, RM, WTU). Ehasta Co.: 0.8 mi N of Silver Bridge Road along Little Cow Creek, Stevens 1069 (MSC); on banks of N fork of Swift Creek and of Ewift Creek, Wayton s.n., July 1, 1913 (NY). Sierra Co.: Downieville, Eastwood_8181 (UC). Siskiyou Co.: Shasta River, Butler 18 (*UC); along :reeks leading to the Box Canyon of the Sacramento near Mt. Shasta City, 35EEE.122£§ (NY, UC). Stanislaus Co.: vicinity of LaGrange, 81188 181;, Aug. 17, 1961 and Jan. 25, 1962 (JEPS). Tehama Co.: Cold Fork )f the Cottonwood Creek, near foot of Tom's Head, Yollo Bolly Mtns., Iepson s.n., Apr. 28, 1899 (JEPS). Trinity Co.: along the Trinity Liver 5 mi NW of Del Loma, Furlow 305 (MSG). Tulare Co.: below fohnsondale, 20 mi N of Kern County line, Munz 12162 (RM, WTU). Tuolumne Co.: near Woods Creek, along state hwy. 49, 5 mi S of lonora, Barclay 8£_81. 1534 (MSC); Bear Creek at Rawhide, Ferris 1487 *NY). Ventura Co.: 1% mi NW of Casitas, Sowder 145 (*UC); Horn agon, Ojai, Thacher s.n., Apr. 24, 1918 (*JEPS). Yuba Co.: Green- ille, 81y18_81 (UC). Idaho. Canyon Co.: Big Willow, Macbride 118 ). Clearwater Co.: Orofino, Christ 1811 (NY). Idaho Co.: Rocky nyon Creek near Cottonwood, Christ 18848 (NY); on the banks of eep Creek, at the Pete Wilson Ranch on Snake River, 103 mi S Lewiston, Christ 18111 (NY). Latah Co.: near Kendrick, Christ 151 (NY). Nez Perces Co.: Willow Creek, E of Fayette, Christ 8444 Y); about Lewiston, Heller 8 Heller 8111 (DAO, *MO, NY, UC). shington Co.: 6 mi up Mann Creek from highway Idaho 95, Christ 034 (NY). Oregon. Curry Co.: bank of Cheteu R. 7 mi above rbor, Peck 8905 (NY). Grant Co.: Dayville, Mason 3595 (UC). 249 Hood River Co.: E line of county, Henderson 818 (MO). Jackson Co.: Neil Creek, 6 mi S of Artland, Applegate 888 (*UC); Gold Hill, Walpole 188_(*US). Jefferson Co.: 7 mi W of Culver near hydro- electric power station, Gillis 8811 (MSC). Josephine Co.: Grant's Pass, Brandegee 2:2;9 Sept. 17, 1885 (UC). Malheur Co.: upper end of Sucker Creek Canyon, 25 mi S of Adrian, Glarkson 884 (*DAO). Wallowa Co.: from 0.5 to 2 mi SE of Imnaha, Bartlett 8 Grayson 884 (NY); 10 mi S of Grant's Pass, on the bank of the Applegate River, Furlow 888 (MSG). Wasco Co.: The Dalles, along Ghenowith Creek, 83823.3lég (NY); Kah-Nee-Ta Hot Springs, floodplain of the Warm Springs River, Gillis 4818 (MSC); along the Deschutes River, £233§.§§§§ (*WTU). Jashington. Asotia Co.: along Asotia Creek, Asotia, Hitchcock §.§Efl‘ $311533 (1m). Klickitat Co.: 7 mi NE of Klickitat beside Klickitat ireek, Gillett 8 Taylor 11066 (DAO); along creeks near Bingen, Suksdorf 42;, Feb. 13, Mar. 18, 1893 (*UG); streams, W Klickitat Co., Suksdorf :2;2 Mar. 2, May, 1881 (*F, UC). Walla Walla Co.: grassy stream ank 2 mi 8 of Walla Walla, Hitchcock p Muhlick £2.52 (NY, WTU). litman Co.: along stream on N side of Snake River, 14 mi W of swiston, Hitchcock 8 Muhlick 88888 (RM, UC, WTU); Wawawai, 81, 8888 {2_3_ (NY, UC). Alnus rhombifolia, like many other species in the genus, netimes appears with densely pubescent leaves. Such a variant : named var. bernardina by Munz and Johnston (1925). Study of 5 Species from all parts of its range in western North America wed this form to merely represent one extreme of the overall tern of variation, and it does not seem worthy of formal recognition. 250 Occasionally, especially on vigorous sprouts and young plants, the leaves assume a more lanceolate form with the tip becoming acuminate and the margin double-serrate, suggesting somewhat the shape of 4. oblongifolia, to which it is undoubtedly closely allied. The largest iknown individual of this species is reported by Dixon (1961) to grow (in the Angeles National Forest in California and measure 3.4 meters *in circumference and 28.4 meters in height with a spread of 13.7 meters. This alder normally occurs in habitats with moderately warm lwinters, and in the southern part of its range it does not lose its ileaves and become dormant. But it also survives in some more rigorous situations (in the Sierra Nevada Mountains). Plants from Shasta Co., California (originally growing about 20 meters above sea level), when transplanted to an experimental field at Lansing, Michigan, were Killed back to ground level during each of three consecutive winters, )ut always resumed growth the following spring. No specimens were seen from Mexico, but since Alnus rhombifolia (I: found as far south as San Diego, California, it might be expected 5 adjacent Baja California as well. Alnus acuminata H.B.K. Alnus acuminata H.B.K., Nov. Gen. Sp. Pl. 2: 20. 1817. Narrow-crowned trees up to 30 m in height; trunks one or several, act to spreading, up to l m in diameter; bark gray to gray-brown, >oth to slightly rough, scaly on old individuals, often broken by 251 transverse ridges or constrictions encircling the stem; young stems medium green-brown, brown, or dark red-brown, dull to moderately lustrous, sometimes slightly to heavily glaucous, without conspic- uous longitudinal ridges originating at the nodes; internodes glabrous to moderately villous or velutinous, moderately to densely glandular; nodes and stems bearing inflorescences densely glandular; hairs yellowish to brown (occasionally dark brown); glands small to medium in size, yellowish, brownish, or dark brown; lenticels of twigs cir— cular to elliptic or elongate, 0.3-1.5 mm long, 0.2-1 mm wide, whitish or yellowish, inconspicuous to moderately prominent; leaf scars 1-3 mm high, 1.5—4 mm wide, with inconspicuous bundle scars. Buds ovoid, ellipsoid, or obovoid, acuminate or acute (to slightly rounded) at the apex, lightly to heavily resin-coated; stalk 1-7 mm long, 1-2 mm in diameter, glabrous to moderately villous or velutinous, £dense1y glandular; body 3-10 mm long, 2-4.5 mm in diameter; scales 2, hore or less equal, stipular, valvate, often incompletely covering the Lnderlying organs, glabrous to sparsely pubescent (to rarely densely 1 pillous), glandular; pubescence and glands usually obscured by the esinous coating. Leaves lanceolate, narrowly to broadly ovate, ob- Ong-ovate, elliptic, or (infrequently) obovate; apex long-acuminate, cuminate, acute, obtuse, or rounded; base acute, cuneate, obtuse, r rounded, sometimes oblique; blade (3.5—) 5.5-14 (-19) cm long, _2-) 3-9 (-10.5) cm wide, medium to very dark green and dull, mod— rately lustrous, or very lustrous above, light to medium green or rown and dull below, chartaceous to coriaceous; margin slightly to oderately revolute or sometimes nearly flat, unthickened or slightly 252 thicker than the blade itself, coarsely, moderately, or finely double— serrate to serrulate; major teeth 8-14 (~20) mm apart, up to 5 mm deep, regular, slightly uneven, or irregular; secondary teeth (2-) 4~8 (~10) per cm, (0.1-) 0.3-1.5 mm deep, regular, slightly uneven, or irregular; abaxial surface glabrous to sparsely pubescent (rarely moderately vil- lous), sparsely to densely glandular; adaxial surface and veinlets glabrous to moderately villous (sometimes tomentose), densely glandular, f moderately resin-coated; major veins and vein axils near the base tomen- I tose to wooly ,"h c “t; r'“ """"" whitish, yellowish, or brown; glands small to medium in size, whitish, yellowish, or brownish (rarely dark brown). Lateral veins (7—) 10-15 (~18), (3-) 5-8 (~15) mm apart at mid-leaf, straight to slightly ascending, usually branching once again, especially near the base, terminating in major teeth at the margin; cross veins between lateral veins poorly to well-developed. Petioles (4-) 7-23 (~35) mm long, (0.8-) 1—2 (-2.5) mm in diameter, glabrous to velutinous, moderately to densely glandular. Stipules ovate to elliptic, the apex acuminate to acute, 4-8 mm long, l—l.5 mm wide, hreen to light brown, sparsely pubescent to velutinous, the hairs [ellowish to brownish, moderately glandular, the glands yellowish. [istillate inflorescences borne in racemose groups of (2-) 3-6 on ihort non-strongly-divergent to strongly-divergent branchlets, these enerally subtended by leaves, produced during the previous growing eason, erect, ovate to elliptic, at anthesis 3-6 (-8) mm long, 1.5— ‘.2 (~3) mm in diameter, on peduncles (l-) 2-5 (~6) mm long, 1-l.5 (~2) F in diameter; staminate catkins borne in one or more racemose clusters F 2-6 at the end of the main branch above the pistillate inflorescences, 1 253 the lowermost often subtended by small leaves, produced during the previous season, pendent during and before anthesis, at anthesis (3~) 5-11 (~15) cm long, 5-10 (~11) mm in diameter, on peduncles 2-10 (~22) mm long, l~1.8 (~2) mm in diameter; floral bracts 1-2 (~3) mm high, (l.5~) 2-3 (-3.5) mm wide. Staminate flowers 3 per bract; perianth of 4 parts, these elliptic or obovate, the apex rounded to obtuse, 1.2-1.8 mm long, 0.6-1.2 mm wide, the margins lined with small to large glands; stamens 4, opposite and basally adnate to the perianth parts, appearing shorter than to longer than the perianth, the filaments 1.1-1.8 mm long, the anthers 1.2-2 mm long and 0.9—1.9 mm in diameter, the thecae separate for 35-55% of their length. Infructescences ovoid, ellipsoid, or cylindric, 11—28 (~45) mm long, 8-12 (~15) mm in diameter, on peduncles 0.2-10 mm long, 1~2 nm in diameter; scales 3—5 mm long, 2.5-5 mm wide at the apex, l~1.8 mn wide at the base, the apex moderately thickened and flat, the :erminal lobe-tip acute to rounded and not much extended to very :xtended. Fruits narrowly wing-margined, dark brown; body elliptic, Lo obovate, 1.5-3 (~5) mm long, 1.5-1.8 (-2) mm in diameter; wings F3 (-5.5) mm long, 0.2-1 mm wide, chartaceous to coriaceous; per- istent styles 0.5-1 mm long- The status of éiflfli acuminata and the other species of this genus ascribed by Humboldt, Bonpland, and Kunth,'4. jorullensis and 4. 188- $8888, has long been confused. Virtually all of the Latin American .ders have gone under each of these names in one taxonomic treatment another. In his monograph of the Betulaceae, Regel (1861) used the me.é' acuminata for the South American alders, including 4. jorullensis 254 of Mexico as a variety. The remaining Mexican forms were separated as 4, arguta (Schlecht.) Spach, and the extremely narrow-leafed 4, castaneifolia Mirbel of South America was left as a separate species. Winkler (1904), in his monograph of the family, trans- ferred all of the forms, including 4, castaneifolia, to 4, jorullensis, where they were retained in the most recent treatment of the genus by Murai (1964). These taxa are very closely related, but there are consistent differences, at least between 4, acuminata and 4, jorullen- sis. Photographs of the types of the H.B.K. species at Paris show that 41888 acuminata bears broadly ovate, acuminate-tipped foliage, while 4. jorullensis has more elliptical or obovate and acute or round- tipped leaves. From the material examined, it appears that 4, 18881- lensis does not occur farther south than Guatemala, while 4. acuminata extends deep into South America. Alnus ferruginea has frequently been used as the name of an extremely pubescent alder, whether in South America, where it was originally collected, or in Mexico and Central America. This pubescent- leaved form is much more frequent in extreme southern Mexico (Chiapas and Oaxaca), Central America, and northern South America than in other parts of the range of 4, acuminata, but it is no more than an extreme in the continuous variation pattern of the pubescence character in this species. 41888 acuminata shows a considerable amount of variation through- )ut its range, but probably not much more than is seen in other such dde-ranging species of the genus, includingl4. viridis and 4. incana. \vs—fg-F 255 Especially variable characters in 41888 acuminata include leaf pubescence (which ranges from almost completely glabrous to densely tomentose or wooly), the density and size of the glands covering the lower leaf surface, leaf shape (varying from narrowly lanceolate to nearly orbicular), and the shapes of the leaf apices and bases. The fact that the foliage is so variable sometimes makes this species confusing and difficult to determine, but taken as a whole, it forms a natural unit. The area of best development of 41884 acuminata, and also of the greatest variability, is southern Mexico and northern Central America, pointing to this region as a possible site of origin. This species is relatively unspecialized, as discussed above. It may have been derived from a widespread prototype similar to present day 41888 rubra, having a mesophytic habitat and a large stature. Alnus acumi- nata and A. rubra are similar in many ways, indicating such a close relationship, but these taxa also have numerous differences, suggesting a probable long period of isolation. a. Alnus acuminata H.B.K. var. acuminata Alnus acuminata H.B.K., Nov. Gen. Sp. Pl. 2: 20. 1817; Alnus cuminata a genuina Regel, Mem. Soc. Nat. Mosc. 13(2): 147. 1861; 1891. lnus jorullensis var. acuminata Kuntze, Rev. Gen. Pl. 2: 638. ype: Humboldt 8_Bonpland s.n., ”crescit in Andibus Peruviae inter axamarca et Micuicampa, regione Escalloniae et Valleae stipularis 1t. l700~l800 hex.” (P, holotype; MSCI, microfiche photograph of type). Alnus ferruginea H.B.K., Nov. Gen. Sp. Pl. 2: 21. 1817; Alnus 256 acuminata Y ferruginea Regel, Mem. Soc. Nat. Mosc. 13(2): 148. 1861; Alnus jorullensis var. ferruginea Kuntze, Rev. Gen. Pl. 2: 638. 1891; 41888 ferruginea var. a. typica Callier, Mitt. Deutsch. Dendr. Ges. 27: 161. 1918. Type: Humboldt 8 Bonpland E;EL’ "crescit locis ex- celsis frigidis Andium Novogranatensium prope Santa Fe’de Bogota, alt. 1400-1600 hex.” (P, holotype; MSCI, microfiche photograph of type). 41888 castaneaefolia Mirbel, Mem. Mus. Hist. Nat. 14: 463. 1827; 18888_18ru11ensis B castanifolia Regel, Bull. Soc. Nat. Mosc. 38(3): 425. 1865. Type: Dombey, ”a’Tarma au Perou" (P?, not seen). Betula arguta Schlechtendal, Linnaea 7: 139. 1832; 41888 arguta Spach, Ann. Sci. Nat. ser. 2, 15: 205. 1841; Alnus arguta a genuina Regel, Mem. Soc. Nat. Mosc. 13(2): 151. 1861. Type: Schiede 81, "prope San Miguel del Soldado, Naulingo, Acatlan, et Chiconquiaco” (HAL?, not seen; M01, isotype or isosyntype). 41488 mirbelii Spach, Ann. Sci. Nat. ser. 2, 15: 204. 1841; 41888 acuminata B mirbelii Regel, Mem. Soc. Nat. Mosc. 13(2): 148. 186l3.él23§ jorullensis var. s mirbelii Winkler, Pflanzenreich 19(4.61): 126. 1904. Type locality: "Peruvia" (original material not seen). 41888 arguta Y ovata Regel, Mem. Soc. Nat. Mosc. 13(2): 152. 1861. Type locality: “Naulingo, Acatlan, Chiconquiaco, und Tabina in Peru“ (original material not seen). 41888 arguta 6 punctata Regel, Mem. Soc. Nat. Mosc. 13(2): 152. 861. Type: 8818, “in Peru und Chili" (not seen). 41888 acuminata Y spachii Regel, Bull. Soc. Nat. Mosc. 38(3): 424. 865; 41888 spachii Callier in Schneider, Ill. Handb. Laubh. l: 132. 904, pro syn.; Alnus spachii Callier, Mitt. Deutsch. Dendr. Ges. 257 27: 163. 1918. Original material not seen. Alggg lanceolata Philippi, Anal. Univ. Chile 91: 514. Type: Paulus Ortega 342;, "Januario 1881, prope Lurin haud procul a Lima in regione litorali Peruviae” (not seen). £1233 jorullensis var. ; acutissima Winkler, Pflanzenreich 19(4.61): 127. 1904; A1335 acutissima Callier, Mitt. Deutsch. Dendr. Ges. 27: 163. 1918; Alnus mirbelii var. acutissima Callier, Mitt. Deutsch. Dendr. Ges. 27: 163. 1918, 232 syn. (erroneously attributed to Winkler). Type: Poe i , "Peru: an Bachen des Huanuca-Thals" (B?, syntype, not seen); Weberbauer 182, ”Thal Von Huillapolschi, sfidwfirts von Matucana" (B?, syntype, not seen). Alnus ferruginea var. 31322 Lorenz & Hieronymus ex Winkler, Pflan— zenreich 19(4.61): 126. 1904, p£2_§yfl. Alnus jorullensis var. n acuminata f. media Winkler, Pflanzenreich 19(4.61): 127. 1904, in part. Alnus pringlei Fernald, Proc. Amer. Acad. 43: 62. 1907. Type: Pringle 19125, Michoacan, by streams near Uruapan, alt. about 1525 m., 13 November, 1905 (CH, holotype; DAOI, F1, MEXU1, M01, M301, NY1, UCI, U31, isotypes). Alnus arguta var. cuprea Bartlett, Proc. Amer. Acad. 44: 610. 1909. ype: Pringle 19221, "Oaxaca, wet canon near base of the summit ridge f the Sierra de San Felipe above the city of Oaxaca" (GH, lectotype; A01, ENCBX, F1, MSCI, U01, U81, isolectotypes). Alnus arguta var. subsericea Bartlett, Proc. Amer. Acad. 44: 610. 909. Type: Pringle 10252, "Oaxaca: wet cafion near the base of the ummit ridge of the Sierra de San Felipe, above the city of Oaxaca" 258 CH, holotype; DA01, ENCBl, F1, MSCI, U01, U81, isotypes). Alnus ovalifolia Bartlett, Proc. Amer. Acad. 44: 611. 1909. Type: §mi£h_21223 Guatemala, San Lucas, Dept. Zacatepequez, alt. 5500 ft. (GH, holotype; USI, isotype). Alnus ferruginea var. obtusifolia Callier, Mitt. Deutsch. Dendr. Ges. 27: 162. 1918. Type: Hartwig 1312, "Colombia: Bogota: in Andi- bus" (BRM?, not seen). Alnus guatemalensis Gandoger, Bull. Soc. Bot. Fr. 66: 289. 1919. Type: von Tfirckheim II-1013, Guatemala, Dept. Alta Verapaz, Coban, 1350 met., Februar 1907 (F1, M01, NYl, isotypes). Rather narrow-crowned trees up to 25 m in height, sometimes shrubby, sprawling, or prostrate on exposed sites; trunks one to several, up to l m in diameter; young stems dull to moderately lus— trous, occasionally slightly to heavily glaucous; internodes gla- ‘ brous to moderately villous or velutinous; glands small to medium in size, yellowish to brownish or dark brown. Lenticels of twigs O.3~l.5 1 mm long, 0.2-1 mm wide, whitish or pale yellowish, inconspicuous to moderately prominent. Buds ovoid, ellipsoid, or obovoid, acumi- ‘nate to acute at the apex, lightly to heavily resin—coated; stalk :1—7 mm long, glabrous to moderately villous or velutinous; scales ‘glabrous, sparsely pubescent, or densely villous. Leaf apex long- acuminate to acute, obtuse, or rounded; base cuneate to rounded, sometimes oblique; blade 3.5-10 (-l9) cm long, (2-) 3.5-9 (-11) cm wide; margin slightly to moderately revolute; the major teeth 8-17 mm apart at mid-leaf (often apparently absent), slightly uneven to irregular; secondary teeth (2—) 3-10 per cm, 0.1-1.5 mm 259 deep, slightly uneven to irregular; adaxial surface and veinlets glabrous to moderately villous or velutinous. Lateral veins (7—) 10-16 (—18), (3-) 4-8 (—15) mm apart at mid-leaf, slightly to moderately ascending; cross veins between lateral veins well-devel- oped. Petioles (4-) 7-16 (-35) mm long, 1-1.8 (-2.5) mm in diameter, glabrous to moderately villous or velutinous. Stipules 4-15 mm long, 2.5-4 mm wide, sparsely pubescent to velutinous. Pistillate inflorescences at anthesis 3-8 mm long, 1.5-3.2 mm in diameter, on peduncles 1.5-6 mm long, 1-2 mm in diameter. Staminate catkins at anthesis 3.5-15 cm long, 5-11 mm in diameter, on peduncles 2-22 mm long, 1-2 mm in diameter. Staminate flowers with 4 perianth parts, these obtuse to rounded at the apex, 1.2-2.2 mm long, 0.6-2 mm wide, the margins lined with small to large glands; stamens appearing shorter than, equal to, or longer than the perianth, the filaments 0.8-1.8 mm long, the anthers 1.2—2 mm long and 1.1—1.9 mm in diameter, the thecae separate for 20-50% of their length. Infructescences ‘(ll-) 15-45 mm long, 8—12 (-15) mm in diameter, on peduncles 0.5-8 (~10) mm long, 1.2-2 mm in diameter; scales 3-5 mm long, 2.5-4.5 mm wide at the apex, 1~1.8 mm wide at the base. Fruits narrowly wing- margined; body 2.7-4 mm long, 1.2—2 mm in diameter; wings 2-3 mm long, 0.2—1 mm wide, chartaceous to coriaceous; persistent styles 0.5-1 mm long. Plates 11A, 12A, 13A, 18A, 24, 25, 26, 27, and 28. Distribution and Ecology: Mexico from central Sonora southeast along the Sierra Madre Occidental and the Sierra Madre del Sur to Oaxaca; east and north from Michoacah to southern San Luis Potosf Plate 24. 260 Specimen of Alnus acuminata H.B.K. var. acuminata. Specimen of Betula arguta Schlecht. Type 261 1? V1134“.- 1. * -. .1 §Z§ 'ifiuu- ... ' f/ 6 min 7r; ‘ [mark/IE v' d”, ': / Plate 24. Plate 25. 262 Specimen of Alnus acuminata H.B.K. var. of Alnus pringlei Fern. acuminata. __..—-—— Isotype 263 Plate 25. W 7' 8c, o. rumour, “MR PLANT/1E MEXICAN/E. IOOB. 7" " —mu or mm— 7 ' ' 1 1012.3 Alnus [’rlnglc-l. Fernald n. ~p. I; was" "or llruupan, 5000 [I .\ Imall Irw- I.'l. Km 264 Plate 26. Specimen of Alnus acuminata H.B.K. var. acuminata. Iso- type of Alnus guatemalensis Gandoger. g 265 (TA. M f , r 1 1 qt. 4 111...).«11? out. 6L1 2, Lawfuk/ #5,!) TYPE 241440 FLORA VON GUATEMALA. DEPT, ALTA VEHAPAZ. I M/y ' r’f ' /'/',’(./ {/1 7% 1/ zl/‘I/ 4 ,'/(/1 (4' '(2 (”Kan /u7/ ...;l‘ MA...“ 470/ I ’ a 15ml; '3 H II. IW' TIUIIJIIIIH Plate 26. Plate 27. 266 Representative specimen of Alnus acuminata H.B.K. var. acuminata. p___.._____, 267 PLANTS OF SOUTH AMERICA hy-lum dun. "datum ”Mn.“ \ ’ em WWI“?- AIOW Jun! t-o. I P‘E'RY COLL. IAGIIIDI C mTNlIIYOII 1' M Plate 27. 268 Plate 28. Specimen of Alnus castaneifolia Mirbel (= 5? acuminataHoB-K var. acuminata). L L———..———»_ ___. ..J—_— __—r PX nl'r)’,A HERBARIUM HORYI BOTANICI MATRITENSIS Plantae a -Ruiz e1 Pm't'ufi in vice-rcgnu Pcruviuno cl Chilcnsi lectae. llTTS-ITSSI W ‘ Mia/f4. k“? . Jaw ' /”"4, m aw % 421/1: Plate 28. 270 and northern Veracruz, and central Chiapas; southern Guatemala, central Costa Rica, and southwestern Panama; eastern Venezuela and northern Columbia south along the Andes to northern Argentina. Along streams and adjacent moist slopes at elevations from about 2000 to 3500 meters (though sometimes found as low as 1500 or as high as 3800 meters). Usually with Pinus, Quercus, or Abies. Figures 23 and 24. Common Names: Aile, jaul, aliso. Representative Specimens: ARGENTINA. Valle de Tafi, Prov. Tucu- ‘ man, §EEEE.E;EL’ in 1908 (NY); Prov. de Jujay, Yala, cerios, Burkart é Troncoso 11222 (US); Jujay, Cabrera g Schwabe 181 (US); Dept. de Andal- gala, Prov. de Catamarca, Jorgensen 1426 (M0, UC); Villa Nougues, Prov. Tucumah, Krapovickas § Cristobal 14522 (*MO, UC); Saladillo, Prov. Tucuman, Dept. Chicligasta, Mayer 13.956 (*MO); Cumbre de Taficillo, Prov. Tucuman, Dept. Tafi, M31 5 g; a1._2115 (DAO, *NY, UC); de Yala a Lagunas de Yala, Prov. Jujuy, O'Donell 2§§4 (*NY); camino de Yala a 1as Lagunas, Prov. Jujuy, O'Donell 4222 (*MO); Cabrohoreo, Prov. Tucuman, Dept. Taff, 91ga_125 (NY, UC); Cerro 1a Cuera Sta. Cruz, Prov. Salta, Dept. Orau, Pierotti 222 (*NY, UC); Sierra del Cajén, Prov. Salta, Rodreguez 1221 (*NY); Infernillo de Diablo, Prov. Tucuman, Senn 4134 (DAO); Hogava, Venturi 1047 (*UC); Siacubon, Venturi 1047a (*F); Siacubon, Venturi 3865 (*MO); along Rio Yala, 3 km w of Yala, West 6238 (M0, uc); 1a lagunita, Prov. Tucuman, Dept. Tafi, Xescole s.n., Jan., 1944 (*NY, UC). BOLIVIA. Without location, Bang 1893 (MICH, M0, NY, U3); La Paz, Buchtien 680 271 Figure 23. Distribution of Alnus acuminata H.B.K. var. acuminatain Mexico and Central America. 272 Figure 23. Figure 24. 273 ata in Distribution of Alnus acuminata H.B.K. var. acumin South America. a -L- -. 274 Figure 24. 275 (*MO, NY); location illegible, Buchtien 2146 (*NY, US); La Cruz, Kuntze 313;, in 1892 (NY); without location, Kuntze 312;, Apr., 1892 (NY); Tunari, Kuntze 5:51;, Apr. 5, 1892 (*NY, US); Okara, Ia£e_§;n;, Apr. 26— 29, 1926 (NY); Pulcheri, $133231 (MICH, *NY). COLOMBIA. Bogueron de Bogota: AEQES.Z29 (*NY); Paramo de Guasca, Cundinamarca, 23113.22168 (*UC); Caldas, 16 km from Manizales along road to Bogota, Breteler 3452 (NY); Caldas E of Neira, near Cemento de Caldas, Breteler 3412 (*NY); without location, Celestine M311§_122& (US); Vereda de Roto, 4 km S of Cota, Fassett 25675 (UC); Rio Guabo at mouth of Quebrada de Pueblo Viejo, 4 km SE of Piedrancha, Fosberg_2112§ (NY); S side of valley of Rio Gomeza (Rio Arobispo) 6 km E of Socha, Fosberg 22210 (MICH, NY); Cundinamarca, Cordillera Oriental, Paramo de Guasca, Garcia-Barriga § 1 Schultes 12543 (NY); Dept. Norte de Santander, Garganta 121 (F); 1 Dept. Norte de Santander, E slope of Paramo de Santurban, toward 1 Mutiscua, Killip é_§m1£hi12§1§ (*NY); Dept. Norte de Santander, be— I tween Mutiscua and Pamplona, Killip §_§m1ghl12§21 (*NY); Quebrada Surbar along road between Dultama and Charala: Langenheim 2251 (UC); Paramo de Guantiva near km marker 221 on road between Belen and Susacon, Langenheim 1632 (UC); canoncito in sabrano, Dept. of Cundinamarca SW of Las Cruces, Bogota, Pennell 2121 (*NY); hills of Chapingro, £512g_21§_(M0); hills at W base of Monserrate, E of Bogota: §m1£h_1912 (MO, UC); paramo de San Antonio, entre la Laguna de La Cocha y el Valle de Sibundoy, Prov. Putumayo, Schultes 22215 (*F). COSTA RICA. Roads above San Isidro de Coronado, A11En_§4§ (*F); Las Nubes, Coronado, Echeverria 122 (F, *UC); Prov. Cartago, Cordillera de Talamanca, Lems 5031 (*NY); Volcah Poas, Lems s.n., 276 July 26, 1963 (NY); in pasture below crater of Turrialba Volcano, L_e_n_r;§l§ (F); near Sanatorio de Tierra Blanca, Cartago, slopes of Volca’n Irazu, Rodriguez 9. 1E (MICH, *UC); 1 mi above Las Nubes, E of San Isidro de Coronado, Prov. of San Jose’, _S_to_r_lgfl9 (MICH); in Podocarpus-Quercus forest on slope of Volca/n Poas about 20 km NNW of Alajuela, Taylor 1%! (NY); bords des rivieres au Copez, Tonduz _Llfig (MEXU, MICH, *NY). ECUADOR. Chillo Valley, Rio Pita, Anthony §1_a_t_e_29_(*US); Prov. Loja, Nudo de Cajanuma, 7 km S of Loja, Camp E—122 (NY); Prov. Loja, Cerro Villanaco, ca. 7 km W of the city of Loja, ginpwé (NY); Prov. Loja, Cerro Villanaco, ca. 7 km W of the city of Loja, Emmi-6L6 (*NY); Prov. Canar, uplands called Huai- racaja, 10-40 km NE of Azogues, gimp E-1779 (NY, UC); Prov. Azuay, along the Rio Matadero, W of Cuenca, gml E—l930 (NY, WTU); Prov. Azuay, along the Rio Cumbe, 25—30 km S of Cuenca, Camp E-207O (*F, 1 1 1 NY); Prov. Canar, dry chapparral scrub and pa/ramo, with occasional ~moist valleys, near E1 Tambo, gimp E-2432 (*NY); Prov. Azuay, Quebra- das leading into the Rio Collay, 3-8 km N of Sevilla de Oro, gimpE; _5_12§_ (NY); along banks of Rio Blanco, 16 km from Sigsipamba, m 117-2_7_6_ (*MSC); near Bolivar, mw (MSC, NY); valley of the Rio Matadero, a few km W of Cuenca, 9133311 (*US); Agua de Oro, Heinrichs _8_20_ (NY); Provs. Imbabura and Pichincha, Otavato to Maichiugui, M— wm (*NY); Cayambe, Little _6_9_9_7 (US); Rio Tasquasa near Angel, Prov. Carchi, 143131533 (tuc); Tilulun, Pachano 102 (NY); Mt. Tungurahua, Rimbach 252 (F, NY); W slope of Mount Tungurahua, @— flflg (MICH); vicinity of Cuenca, ME: (*NY); in Andibus Equadorensibus, Spruce 5755 (*NY); Prov. Loja, W slopes of Cordillera 277 de Condor and NW slopes of Nudo de Sabanillas, Steyermark £3166 (F, NY); Puente de San Luis between San Gabriel and Bolivar, Wiggins M (NY); Tarqui Valley, 3 of Cuenca, Wiggins 10_7_3_§ (NY); Pa/ramo Alpacado 82 km S of Cuenca, Wiggins 19214 (NY); N side of valley of e1 Rio Leon, 85 km S of Cuenca, Wiggins 19246 (*MO, NY, UC). E1 SALVADOR. Northern slopes of Santa Ana Volcano, Dept. of Santa Dept. Santa Ana, Volcah de Santa Ana, Carlson 119 (F, UC); cloud forest, Mountain Cerro Verde, Dept. Santa Ana, Molina 2,'§ Montalvo 21229 (NY); N slope of Volcah de Santa Ana, Tucker 1224 (F, MICH, MO, NY, UC). GUATEMALA. On the Tecum Uman Ridge at km 154 on Ruta Nacio- nal No. 1, ca. 20 km E of Totonicapan, Beaman 4292 (*MSC, UC); 23 mi N of San Sabastian (N of Quezaltenango), Beaman 5834 (MSC); 23 mi 1N of San Sabastian (N of Quezaltenango), Beaman 5834 (MSC); 5 mi N of San Juan Ixcoy on road to Soloma, Breedlove §§Z2 (*F); bank of Rio Panajachel near Lake Atitlan, EEEEE.§ Wilson 22g (*F); Antigua, Dept. Sacatepéfiuez, Kellerman 4222 (*US); about 11 mi W of Quezaltenango, £335.2l22 (NY, UC, *US); road to Iximche Ruins, Tecpan, Dept. Chimaltenango, Molina 23 §1_31, 16121 (*F); S. Rafael, Rittier 22 (*US); Dept. of Huehuetenango, Skutch 1110 (*F, NY); San Lucas, Dept. Zacatepéouez, Smith s.n., M. Apr., 1800 (*US); Dept. Guatemala, slopes of Volcah de Pacaya, between San Francisco Sales and the base of the active cone, Standley 80553 (*F); Dept. Huehue— tenango, Aguacatah road, 10 km E of Huehuetenango, Standley 82135 *F); Dept. Huehuetenango, about Laguna de Ocubila, E of Huehuetenango, A tandle 82695 (*F); Dept. San Marcos, mountains along the road between 278 1 San Marcos and Serchil, Standley 22221 (*F); Dept. Quezaltenango, oak—pine forest, above Los Vahos, Cerro Quemado, Standley 66626 (*F); Dept. Quezaltenango, oak-pine forest, above Los Vahos, Cerro Quemado, Standley 66122 (*F); Dept. Quezaltenango, Volcan Zunil, Steyermark 26622 (*F); Dept. E1 Progreso, between Calera and summit of Volcdn Siglo, Steyermark 52622 (*F, NY); Dept. Alta Verapaz, Coban, 4200 ft, 323 Tfirckheim‘221 (*MICH, US); Dept. Alta Verapaz, Coban, 1350 m, X23 Tfirckheim II-1013 (*F, M0, NY); pine forest region in Sierra Madre Mountains where depts. of Huehuetenango, Totonicapah, and Quezaltenango join, Williams E£.El' 22718 (*F). MEXICO. Chiapas. Along road to Zontehuitz, Breedlove 6662 (ENCB, *F, MICH); along the road to Chenalho near the schoolhouse of Yal Ichin, Breedlove 1221 (ENCB, *F); along creek near road 1 km N of Aguacatenango, Breedlove Z222 (ENCB, *MICH); on road from San Cristdbal Las Casas to Tenejapa, Breedlove 2262 (F, *MICH); 1 mi W of Nabenchauk along Mexican highway 190, Breedlove 2222 (*MICH); creek bank at NE bound- ary of Aguacatenango, Breedlove 2662 (MEXU, MSC, *US); SE city limits of Teotisca, Breedlove 11222 (*MEXU); near the center of Amatenango, Breedlove 12122 (*MSC); in the barrio of Tuk, paraje of Matsab, - Breedlove 12457 (*MSC); between Las Casas and Tenejapa, Carlson 2412 (*MEXU, MICH); on the NW side of Muk'ta vits (Cerro Huitepec), Laughlin 1870 (MSC); near Zinacantéh Center, Laughlin 2347 (*MEXU); de S. Cristdbal a Buenavista (Chepetic), Miranda 4987 (*MEXU); Cerro del Boqueron, Purpus 6981 (NY, *UC); paraje de Mahosik', Ton 1134 (*MSC); in the paraje of Yash'anal, Ton 2130 (MEXU, *NY). Chihuahua. Arroyo Hondo, Sierra Charuco, Gentry 8163 (MEXU, MICH); Basigochi, 279 SE of Creel, Knoblock 622 (ENCB, MSC, WIS); near Conchefib, LeSueur 222 (*F); 44 mi S of Creel, steep barranca wall, EEEEEHQ Foreman 1221 (*ENCB, MEXU, MICH); Cerro Mohinora, 10 mi S of Guadalupe y Calvo, §E£§ZH§ Foreman 2216 (ENCB, MEXU, MICH). Distrito Federal. Vallee d. Mexico, Pedrigal, Bourgeau 262 (US); on Mex. hwy. 16, 1.9 mi W of San Bartolo, Denton 1222 (MSC, WTU); Cafiada de Contreras cerca de 20 dinamo, Espinosa 661 (*ENCB); Valle de Mexico, Amecameca, Matuda 12222 (MEXU). Durango. North slopes of Cerro Huehueto, S of Huachicheles, about 75 mi W of C. Durango, Maysilles 1221 (MEXU), MICH); Laguna del Progreso, 34 road mi N of railroad at Coyotes, Maysilles 2222 (MICH); San Ramon, Palmer 221 (NY, UC, *US). Guana- juato. Mountains ESE of San Jose Iturbide and about 5 mi W of Cerro Zamorano, wooded canyons in oak forests near summits called Mesa del Gato, McVaugh 12221 (MEXU, *MICH); San Luis de la Paz, Salazar 2_(MEXU). Guerrero. Puerto de los Velanquez,2§2§_§ Martinez 1662 (ENCB); Pilas, Distr. Mina, Hinton 12122 (*NY, UC); Carriaal-Espadines, Distr. Mina, Hinton 12222 (NY, UC); Paraje Javalin, Distr. Mina, Hinton 12212 (MICH, NY); approximately 6 mi IW of Mazatlah, Rowell 2212 (*MICH); 2 km a1 E de 0mi1temi, Rzedow- . 321.16222 (*ENCB); 5 km al W de Camotla, Mun. de Chichihualco de :Leonardo Bravo, Rzedowski 16622_(ENCB, *MEXU, MSG); Cerro Alcuitran, cerca de Mazatlah, Rzedowski 22622_(ENCB); 11 km al E del Aser- radero Agua Fria, Rzedowski 2 McVaugh 262_(*ENCB, MSG); near Omil— temi, 262£p 441587 (MEXU). Hidalgo. Mun. Chapulhuacan, km 296 on Laredo highway, 2111y_§_Rickett 2_(MICH, MSG); alrededores de Zacualti— pan, Gonzalez Quintero 318 (ENCB, MEXU, *MSC); alrededores de Tenango 1 1 1 1 1 280 de Doria, Gonzélez Quintero 222 (ENCB); Santiago Tepepa, Mun. de Acoxochitlah, Gonzélez Quintero 222 (*ENCB); Rancho Viejo, Mun. de La Mision, Gonzalez Quintero 1222 (*ENCB); 30 km a1 NE de Jacala, Gonzélez Quintero 1221 (*ENGB); Xochicoatlan, Mun. de Molango, Gonza/lez Quintero 126_0 (*ENCB); Santuario, 22 km a1 NNE de Ixmiquil- pan, Gonzalez Quintero 2212 (*ENCB); cerca de Chapulhauacan, Rzedowski 12262 (*ENGB, MSG); alrededores de Zacualtipah, Rzedowski 12221 (ENCB, MSG); Tezoantla, cerca de Real del Monte, Rzedowski 12211 (ENCB, MEXU, MSG); on the mountain between city of Mineral (Real) del Monte and highway 105 by-passing city, §E£Efln§ Gregory 1121 (MICH); Lindavista, sobre el camino a Tenango de Doria, 1613.2 Rzedowski 222 (*ENGB); Cerro Jihuingo, 17 km NW of Apam, 2§§1_1222 (MICH). Jalisco. Etat de Jalisco, Liguet.2121, without date (*NY); mountain in oak-pine forest, near summit of pass 7-8 road mi NW of Los Volcanes along 'road between Ayulta and Mascota, McVaugh 12222 (MICH, *US); Sierra de la Campana, along road to Mascota, 7-8 mi NW of Los Volcanes, McVaugh 12112 (MICH); about 6-10 mi SW of Talpa de Allende, above Aranjues in the valley of Rio Gharco Verde and near its headwaters, McVaugh 12222 (*MICH); 10-12 mi W of Talpa de Allende, in the headwaters of an E branch of Rio de Talpa, 3 mi above Los Sauces, McVaugh 21222 (*MIGH); headwaters of Rio Mascota (about 20 km, air- line, SE of Talpa de Allende, McVaugh 21212 (*MIGH); 16 km S of El Chante (ca. 25 km SE of Autlah), McVaugh 22222 (*ENGB, MICH); Sierra del Halo, near a lumber road leaving the Golima highway 7 mi SSW of Tecalitlan and extending SE toward San Isidro, McVaugh‘g Koelz 1123 (*MICH). Mexico. Rancho Tobias near Villa Guerrero, ,-_ _.—_....__.-——- 281 Alexander 2 Hernandez 23 1212 (*NY); on slope of mountain about 1 km E of Mancho de Agua, 19 km E of ZitaEuaro, Clausen §;2L’ Oct. 5, 1955 (MEXU, NY); 4.8 km E of the Mékico-Michoacéh state boundary along rt. 15, Furlow 222 (MSC); Rancho Santo Tobias near Villa Guerrero, 2111y_122 (*MSC); along small stream just N of Ameca- meca, 6211y_6_2622§_1‘(*MSG, NY); Rancho Santo Tobias near the town of Villa Guerrero, 2111y_61_§1, 122 (*MSC); 5 km a1 SW de Cahuacan, Gonzalez Quintero 212 (*ENCB, MSG); La Labor, Distr. of Temascalte- pec, Hinton 2162-(*US); Rincdh, Distr. Temascaltepec, Hinton 2122 (*NY). Michoacah. South of Morelia at km 263, 2232 63-586 6_211S§ 212 (*UC); Jose Morelos Nat. Pk., 30 km E of Morelia, Furlow 221; 23 km W of Mil Cumbres, Furlow 222 (MSC); 1.5 km S of Opopeo, Furlow 221 (MSC); ZitaEuaro-Carpinteros, Hinton 11212 (DS, NY, *UC); west-facing slopes of Cerro de Carboneras above the Rio Cupa, 212g.2_Soderstrom 2222 (*MICH, NY, UC); 6 mi N of Tancitaro on S mt. slope, Leaven— mfl (F, MICH, MO, NY); 8—10 mi NW and WNW of Ciudad Hidalgo, among mountains W of Cerro San Andre; and 6—7 mi N of village of San Pedro Aguaro, McVaugh 2222_(MEXU, MICH, *MO); San José’Purua, 12331 2222 (*ENCB); by streams near Uruapan, 5000 ft, Pringle 12122 (DAO, F, *MEXU, M0, MSG, UC, US); 14 km a1 E de Zitabuaro, sobre 1 Morelos. Al N de Coajomulco, Palacios s.n., Jan. 9, 1965 (*ENGB). Nayarit. 5 mi N of Compostela, near the bridge over Rio Mira— valles, McVaugh 2_Koelz 627 (*MICH). Oaxaca. San Miguel Suchis- tepec, Alexander 596 (*MEXU); between Ayulta and Santa Maria, Camp ,2751 (MICH, *NY, UC); Cerro San Felipe, Gonzatti 2215 (*F); Hacienda 282 Guadulupe, Conzatti ghioifl (*MICH); Vicinity of Cerro Zempoalte- petl, Hallberg 2‘9. (*ENCB, MICH); Cerro de San Felipe, Mg; (F, M0, *NY, NC, WTU); by brooks, Sierra de San Felipe, Pringle _10_221 (DAO, ENCB, MSC, *UC, US, WIS); by brooks, Sierra de San Felipe, 31:31: 513% (DAO, ENCB, F, MSC, UC, *US, WIS); Puertecillo de Lanchao, Rzedowski w (*ENCB, MEXU, MSG). Puebla. Huanchinango, Aguirre E R_&Lo_ _1_5(l (*NY); entre Teziutla’n y Atempan, Ern 434 (ENCB); 5 km a1 sw de Huauchinango, Esginosa 601 (*ENCB, MEXU); near Beristain above highway between Tulancingo and Huachinango, _G—1_1_13L7_2. (MSC); Los Molinos, near Atlixco, _G_1_l_ly_l§3_(*MSC); Tezuitlah, Orcutt EL: Sept. 6, 1910 (*F, MO); Sierra de Zacapoaxtla, Palacios s_.n_., Jan. 19, 1968 (*ENCB); E1 Para1’so, V. Jua/rez, Sarukhan e_ta_1. _s_._n_., Apr. 12, 1962 (*MEXU). Queretaro. 51 mi NE of Zimapa’n, Waterfall 14_292_ (US). San Luis Potosi. Soledad de Zaragoza, Mun. de Xilitla, Rzedowski 2E (ENCB); El Guajolote, 10 km a1 W de Rosa de Castilla, Rzedowski fl (*ENCB). Sinaloa. Along the Arroyo E1 Surutato, 2 mi N of Surutato, Sierra Surutato, Breedlove _1_6i84 (*MICH); 3 mi N of Los Ornos along road ‘ to Ocurahui, Breedlove g Thorne L841: (*MICH). Sonora. Horconcitos, Rio Huachinera, mare; (*ARIZ, MEXU, MICH). Veracruz. Orizaba Mexique, Bilimek fl (NY); 32 km SW of the city of Orizaba, Furlow fl (MSC); Me/xico, Prov. de Vera Cruz, Galeotti fl (US); San Miguel del Soldado E de Jalapa, Gomez-PomEa fl (MEXU); orillas del Rio Jamapa, cerca de Ixhualtlan del Cafe, Egg (GH, *MEXU); 2/3 way up Cerro de San Cristobal, 1/2 mi S of Orizaba, Manning é Manning 2&6. (GH); 17 mi N of Jalapa, Manning g Manning 53813 (MEXU); La Joya, Rosas B. 543 (GH); Acatlan, Roses 3: 600 (GI-1); 283 Malpais, entre la Joya, Ver. y Rancho Dos Hermanos, Ver., Ye_1a Galves _198 (*ENCB); Champilico, Ventura A. 3 (ENCB, MSC); La Florida, Mun. de Atzalan, Ventura _A. $3 (*ENCB, MSC); valley NW of La Perla (vicinity of Orizaba), Weaver e_til: 1102 (*MICH). PANAMA. Llano del Volca’n, MM (F, *MICH, MO, NY, UC); Boquete Distr., Chi- riqui Prov., Davidson E (F); Volca/n de Chiriquf, Davidson 221 (F); vicinity of Boquete, Finca Collins, E1 Velo, fine—til: fl (MICH); 8 mi NE E1 Volca/n, Tyson _6; Dwyer 828 (*MO); Finca Le/rida to Pefia Blanca, Woodson g Schery £1 (*MO). PERU. Chumbes, Pajonal L to Ayacuche, flaw (F, UC); highway between Huanuce and Chinchao, DIV—91% (M0, UC); Valle del Urubamba, Calca, Herrera 209—2 (*F); Canyon of Rio Huasahuasi below Huasahuasi, Hutchison fl (*NY, UC); Cerros Calla Calla, E side, 7.5 km above Leimebamba on the road to Balsas, Hutchison é Bennett 4633—6 (F, NY, UC, US); Dept. Junin, Ocopa, Killip gmgzgoi (NY, us); Dept. Junin, Carpapata, Killip gm 24480 (NY); near Yungay, at Ranrahirca, Kinzel s.n., July 23, 1964 (US); Matucana, Macbride é Featherstone 561 (F); Tarma, Macbride g Featherstone 1021 (*F); Mito, Macbride §c_ Featherstone 1527 (F); Nuanuco, Macbride é Featherstone 2083 (F); Cerro Puma Urco, SE of Chachapoyas, Pennell M (*US); Abancay, flfi'.’ June 27, 1957 (F); Rio Acopalca, flag, July 3, 1957 (*F); in Peruviae memonibus, _Ifl .1__l6§_2_ (F, photograph); without location, Bu—fiém 21;, in 1788 (F); km 40, Dept. Huanuco, Huanuco to Carpish, Seibert £22 (MO, *US); without location, Soukup E (F); Huacho, 8 km N of Huanuco, along banks of Ruallaga River, M é: Horton fl (UC); Dept. La Libertad, Prov. Santiago de Chuco, Cachicadan, Stork é, Horton 9978 (UC); 284 alrededores de Huancayo, Tovar 2770 (*UC); Cuzco, Vargas E. 8109 (*MO); Dept. Apurimac, Vargas 9, 8144 (*MO); Rio Apurimac, Dept. Cuzco, Weberbauer 5821 (US); vicinity of Panao, Woytkowski 111 (*F). VENEZU- ELA. Between Merida and Mucuchies, Alston §1Q§ (NY); Chama valley, Apartaderos, Breteler 4412 (*NY); Chama valley, near Apartaderos, Breteler 4416 (*NY); Chama valley, near Apartaderos, Breteler 4411 (NY); Chama valley, between Mucuchies and San Rafael, Breteler 4419 (NY, US); Chama valley, between Nueuchies and San Rafael, Breteler 4480 (*NY); Valle Rio Chama, 9 km from Merida along road to Tabay, Breteler 4481 (*NY); Chama valley between Merida and Tabay, Breteler flfléi (NY); State of Tachira, La Granja, 55 km from Tovar along road to La Grita, Breteler 4429 (NY); Quebrada de Saisay, Gehriger 14 (NY); Merida, 2323.122 (*NY); Moconoque, Mer., everywhere, Pittier 12129 (MO, *NY) . Alnus acuminata var. acuminata is the only alder occuring in South America. Several minor variants, including A, mirbelii Spach, é, spachii Callier, and A, acutissima Callier differ only Kin leaf margin and indumentum characteristics, and these taxa are therefore not recognized. Regionally, Alnus acuminata in South America shows generally smaller leaves and infructescences toward the south (Argentina) than Iin Bolivia and Peru. Farther north, in Venezuela, Colombia, and Ecua— ? ‘dor, the leaves appear generally broader and more rounded at the japex, although acuminate-tipped foliage is not uncommon there. In aEcuador and Venezuela, infructescences are again somewhat smaller than those from other parts of the range. 285 The foliage of South American var. acuminata is quite variable, the apices ranging from long—acuminate to very rounded, the bases varying from rounded to long-cuneate, the general shape varying from narrowly lanceolate to broadly ovate, and the indumentum ranging from nearly glabrous to densely pubescent. Among the Mexican and Central American populations, several minor variants have been named as separate species as well. One such taxon is Alnus pringlei Fern., the only tangible distinguishing characteristic of which is the presence of pubescence on the petioles. Alnus arguta var. Subsericea Bartlett and A. ovalifolia Bartlett, simi- larly, vary only in leaf shape and indumentum from the typical variety of 4. acuminata. Throughout the range, infructescences vary from rather small to very large. The leaves are almost always ovate, though ‘they sometimes tend to become elliptic or even obovate in rare instan- ces, with the apices ranging from acuminate to rounded. Pubescence on the leaves varies from absent to very dense, and leaf surface glands may appear either rather large and somewhat dense, or smaller and sparser, though the glands are never entirely absent, as stated by Fernald (1904b), Standley (1920), and Standley and Steyermark (1952). In Costa Rica and Panama this variety sometimes resembles £1335 jorullensis in that its leaves are often much more glandular and some- times elliptic or slightly obovate in shape. Specimens from this re- gion, nevertheless, have the overall aspect of A. acuminata and probab- ly simply represent part of the relatively greater amount of varia- ility seen in Central America than elsewhere, as discussed above. Alnus castaneifolia Mirbel is a form with extremely narrow 286 leaves having sharp teeth (Plate 28). I have seen only one specimen (and one photograph of a specimen) of this taxon, both collected by Ruiz in Peru. It is apparently quite rare in that it has not been more frequently collected. It seems possible that this form may be only a sporatic or local variant of A, acuminata var. acuminata, but without additional material, it would be impossible to unequivocally resolve its status. It is therefore tentatively being included in this variety pending the accumulation of additional material. Alnus acuminata var. acuminata is usually found in Mexico and Central America growing along mountain streams or on moist slopes in pine-oak forests at relatively high elevations. In Central America it shows a preference for moderate moisture conditions by its absence on the Pacific Slope, where rainfall is more abundant (Standley and Steyermark, 1952). In South America it occupies simi- lar habitats, but at higher elevations. Here, too, alders are not found on the moist western side of the Andes in Chile. 4b. Alnus acuminata var. glabrata (Fern.) Furlow, comb. 22 Stat. nov. Alnus glabrata Fernald, Proc. Amer. Acad. 40: 26. 1904. Type: 4. Dugés s.n., April, 1882, Monte San Nicolas, Guanajuato (CH, lecto- type of Standley, 1920). Alnus jorullensis n acuminata f. media Winkler, Pflanzenreich 19(4.61): 127. 1904, in part. Alnus glabrata var. durangensis Bartlett, Proc. Amer. Acad. 44: 611. 1909. Type: Palmer 965, in the vicinity of the City of Durango, .‘. ..:—-- ~__- Amy 287 State of Durango, April to November 1896 (CH, holotype; F1, NYI, UCI, U81, isotypes). Narrow-crowned trees up to 30 m in height; young stems dull to slightly lustrous, not glaucous to moderately glaucous; inter- nodes glabrous, moderately glandular; lenticels 0.3—1 mm long, 0.2-0.7 mm wide, whitish or yellowish, moderately prominent; , leaf scars l-l.5 mm high, 1.5-2.5 mm wide. Buds ovoid to ellip- soid, slightly rounded to rounded at the apex, moderately to heavily resin-coated; stalk 2-6 mm long, 1-2 mm in diameter, gla- brous, densely glandular; body 5-9 mm long, 3-4 mm in diameter. Leaves narrowly ovate to lanceolate (or rarely elliptic), the apex long-acuminate (or sometimes acute), the base acute to obtuse or rounded; blade (3-) 6-13 (—15) cm long, (2-) 3-6.5 (~8) cm wide, medium to dark green above, light to medium green (or brown) below, chartaceous; margin flat, unthickened, usually sharply double-serrate; major teeth 8-16 (-l8) mm apart at mid-leaf, 2-5 mm deep, regular, usually more or less acuminate; abaxial surface glabrous; adaxial surface and veinlets glabrous (rarely very sparsely pubescent), moderately to densely glandular; major veins and vein axils near the base glabrous to sparsely , L “t; r"h “ r , when present, whitish to pale yellowish. Lateral veins 8-13, 4-8 (-11) mm apart at mid-leaf, straight or slightly ascending; cross veins between lateral veins poorly developed. Petioles (8-) 12—21 (-27) long, 0.8—1.5 mm in diameter, glabrous. Stipules 7-9 mm long, -2 mm wide, green to light brown, sparsely to moderately pubes- ent, the hairs yellowish, moderately glandular, the glands yellowish. 288 Pistillate inflorescences at anthesis 4-6 mm long, 2-3 mm in dia- meter, on peduncles 1-4 mm long, l-l.7 mm in diameter; staminate catkins at anthesis 6.5-9.5 cm long, 6-8 mm in diameter, on pedun- cles 2-14 mm long, 1—2 mm in diameter. Staminate flowers with 4 perianth parts, these elliptic to obovate, the apex obtuse to rounded, 1.5-2.1 mm long, 0.8-1.1 mm wide, the margin lined with minute to small glands; stamens usually appearing equal to or longer than the perianth, the filaments l-l.7 mm long, the anthers 1.6-1.8 mm long and 1.5-1.8 mm in diameter, the thecae separate for 35—45% of their length. Infructescences (10-) 15-25 (~30) mm long, 6-12 (-l5) mm in diameter, on peduncles 0.2-3 mm long, 1.2—1.8 mm in diameter; scales 4.5-5 mm long, 3.7-5 mm wide at the apex, 1.2-1.5 mm wide at the base, the apex moderately thickened, the terminal lobe—tip acute to rounded, often somewhat extended. Fruits narrowly wing-margined; body 2.2-4 mm long, 1.2-2 mm in diameter; wings 2.5—3 mm long, 0.3-0.8 mm wide, firm to coriaceous; persistent styles 0.6-1 mm long. Plates 12B, 13C, and 29. Distribution and Ecology: Central Durango southeast to Guana- juato, Mexico, Tlaxcala, and north-central Oaxaca. Usually found along streams and on moist slopes, often with Pinus and Quercus, at elevations from 1500 to 2500 (or rarely to 3000) meters. Figure 25. Common Names: Aile, aliso. Specimens Examined: MEXICO. Distrito Federal. Cafiada de Con- reras, Martin Contreras §_(ENCB); Tlalpah, Espinosa 491 (ENCB); Cahada e Contreras Es inosa 502 (ENCB)- T1e1pefn, Fisher s.n., Aug. 1 1924 v _L__._ a — 3 289 Plate 29. Representative specimen of Alnus acuminat§_var. filflEEEEE (Fern.) Furlow. Plate 29. c. o. PRIFOLB. PLANT/E MEXICAN/E. 1m, mm: mm: ~02“! Mulls ucumlnnln. Illlli. a. .....n. ,. rm.- 7500/. r... .- 291 Figure 25. Distribution of Alnus acuminata var. glabrata (Fern.) Furlow. 292 Figure 25. 293 (F, MO, RM); Tlalpan, Pringle §_(MEXU); by streams, valley of Mexico, Pringle 4291 (F, MO, *MSC, NY, UC); by streams near Tizapah, 7500 ft, Pringle 9911 (F, *MEXU, M0, M80, NY, UC); by streams near Tlalpah, Pringle 11999 (MSC); Tlalpah, Rose et a1. 8483 (NY); Cahada de Con— treras, cerca del primer dinamo, Rzedowski 11199 (ENCB); Cafiada de Contreras, Rzedowski £131, May 15, 1954 (ENCB); Contreras, Salgado 43 199 (ENCB); slopes, Rio de la Magdalena, between Contreras and the 2nd dynamo, Eha£p_449 (MEXU); San Angel, Torres 124 (*MEXU); Tlalpan, Valle de Mexico, Urbina §_(MEXU). Durango. At the city of Durango and vicinity, Palmer 99: (F, NY, UC, *US). Guanajuato. 14.5 mi from Guanajuato on the road to Dolores Hidalgo, Johnston 1441 (MEXU, MICH, *MSC); ca. 15 mi from Guanajuato on road to Dolores Hidalgo, Solbrig é Ornduff 4509 (*UC). Guerrero. Petlas- cala, 23513.2212 (F, MO, *NY, UC). Hidalgo. Barrancas W of El Salto Station, Distr. Tula de Allende, Mail—451 (MEXU, *MICH, UC); Tula, mfl (NY). Me’xieo. 3 km al N de Magu, fl ‘Cisneros 1991 (ENCB); Atlacomulco, Detling 8919 (ENCB); cruce de la autopista a Queretaro con el camino a Atizapan de Zaragoza, Espinsoa 591 (ENCB); 4 km a1 SW de Tepozotlah sobre el camino a Presa de la Concepcidh, Espinosa 291 (ENCB); just N of Amecameca, Furlow 211 (MSC); just N of Amecameca, 9111y'§_2299§'1 (MSC, NY); Barranca de Texalotengo at Rancho Santo Tobias near Villa Guerrero, 91111 9 LSimpson §_(MICH); Villa de Carbdh, Matuda 19929 (MEXU); Calacoaya, _Medellin 199 (ENCB); al E de Tenago del Aire, a los lados del Rio Tenango, Pineda 1: 211 (ENCB); Parque Nacional ”Los Remedies", Rzedowski 19301 (*ENCB, MSC); 2 km a1 SE de San Pablo Ixayoc, 294 Rzedowski 1419 (*ENCB, MSC). Oaxaca. 5 km adelante de Tlaxiaca, Pennington §_Sarukhan 1, 9119 (*NY); Rancho del Cura cerca de Con— cepcidh, Buenavista, Distr. de Coixlahuaca, Rzedowski 12119 (*ENCB); 1 km a1 E de Ihuitlan Plumas, Rzedowski 14989 (*ENCB, MSG). Puebla. Vicinity of Puebla, Arsene 119 (US); alrededores de Atzala, §£fl_11§ (ENCB); Los Molinos near Atlixco, §h§£2_4§49§ (*MEXU). Tlaxcala. Rancho Nuevo, Tlaxco, Aguilar 8-A-42 (ENCB); 4 km a1 W de Apizaco, Ruiz Bedolla s.n., July 9, 1967 (ENCB); San Pablo, Rzedowski 19 (ENCB); orillas del Rio Zahuapan, cerca de Tlaxcala, Weber 169 (ENCB). In describing élEEE glabrata, Fernald (1904b) did not designate one of his cited speimens as the type. In 1920, Standley, in 13335 329 Shrubs 2£_Mexico, listed the type of this species as from "Monte San Nicolas, Guanajuato,” thus establishing one of Fernald's elements 94. Dugés s.n., April, 1882) as the lectotype. This specimen is the first listed by Fernald in his protologue, and its selection may represent an arbitrary choice of his first listed element, as expli- citly prohibited in the International Code 21 Botanical Nomenclature (Stafleu et al., 1972). Without stronger evidence that this is the case, however, the specimen of Dugés should be allowed to stand as the lectotype. This variety occurs at generally lower elevations than var. 322217 nata, but it is sometimes found growing sympatrically with it. It ay eventually prove to be only a minor variant of the typical variety, ut from the material available, it seems distinct enough to be given arietal status of its own. The most useful distinguishing characters re the completely (or nearly so) glabrous lower leaf surface and the 295 lanceolate or narrowly ovate, acuminate, sharp-toothed leaf form. There is abundant evidence of natural hybridization between this and the typical variety throughout the range, as shown by numerous specimens having intermediate leaf shapes and indumentum. Where such putative hybridization occurs, it is often difficult to determine these taxa. In the specimens cited here, such intermediates have been assigned to one variety or to the other, depending on their closest affinity. They have not been taken into consideration in constructing the key. 5. Alnus jorullensis H.B.K. Alnus jorullensis H.B.K., Nov. Gen. Sp. Pl. 2: 20. 1817. Spreading trees up to 20 m in height; trunk usually single, up to 1.8 m in diameter, the branches sometimes massive; bark gray to dark brown, smooth to corky, often broken by deep transverse constrictions which encircle the stem, the lenticels inconspicuous on smooth branches; young stems light to medium brown or dark red-brown, dull to slightly lustrous, not glaucous to heavily glaucous, without conspicuous resin- coating, not differentiated into long and short shoots, usually with- out conspicuous longitudinal ridges originating at the nodes; inter- nodes glabrous, sparsely pubescent, or velutinous, moderately to densely glandular; nodes and branchlets bearing inflorescences very ensely glandular; hairs yellowish to brown; glands medium to large, ellow to brown. Lenticels of twigs circular to elliptic, 0.5-1.5 long, 0.3-0.7 mm wide, whitish to yellowish or brownish, moderately 296 prominent; leaf scars 1-2 mm high, 1.5—2.7 mm wide, with inconspicuous bundle scars. Buds ellipsoid, slightly rounded to rounded at the apex, moderately to heavily resin—coated; stalk 1-3 mm long, 1.2 mm in diameter, sparsely pubescent to moderately villous, densely glandu— lar; body 2-7 mm long, 1.5-3 mm in diameter; scales 2, stipular, equal, valvate, glabrous to moderately pubescent, glandular; pubescence and glands obscured by the resin coating. Leaves narrowly elliptic, ellip- tic, elliptic-oblong, oblong, or obovate (rarely ovate); apex acute, obtuse, or rounded; base attenuate, acute, or narrowly cuneate; blade (4—) 5-16 (~20) cm long, (2-) 3-7 (-9) cm wide, dark to very dark green and dull to very lustrous above, light to medium green, brown, or yellow—brown and dull below; coriaceous; margin flat to moderately revolute, unthickened, double-serrate or sinuate to shallow—lobed and serrate or serrulate, up to 50% entire from the base; major teeth or lobes (7-) 9—17 (-22) mm apart at mid-leaf, up to 3 mm deep, irregu- lar; secondary teeth (l-) 3-5 (-8) per cm, 0.1-1.2 mm deep, irregular; abaxial surface glabrous to sparsely pubescent, moderately to densely glandular; adaxial surface and veinlets glabrous to sparsely pubes- , cent or moderately villous, moderately to very densely glandular, , slightly to moderately resin-coated; major veins and vein axils a near the base glabrous to tomentose; pubescence yellowish to brownish; glands small to large, bright or pale yellow to brown. Lateral veins 37-17, (4-) 5-8 (-l4) mm apart at mid—leaf, slightly to strongly as- icending, sometimes branching once again, especially near the base, ’terminating in major teeth at the margin; cross veins between lateral “veins poorly to well developed. Petioles (2-) 6—12 (-23) mm long, 297 l-l.5 (~3) mm in diameter, glabrous to sparsely pubescent, moder- ately to densely glandular. Stipules ovate to elliptic, the apex acute, ca. 5 mm long, ca. 1.5 mm wide, green to light brown, mod— erately villous to velutinous, the hairs yellowish, moderately glandular, the glands yellowish. Pistillate inflorescences borne in racemose groups of 3-5 on short non—divergent to strongly diverging branchlets, these generally subtended by leaves, produced during the previous growing season, erect, ovate to elliptic, at anthesis 2-3 (-4) mm long, ca. 2 mm in diameter, on peduncles 0.2-4 mm long, l—l.5 mm in diameter; staminate catkins borne in one or more racemose clusters of (2-) 3-5 at the end of the main branch above the pistillate inflorescences, the lowermost usually subtended by small leaves, produced during the previous growing season, pendent before and during anthesis, at anthesis (3-) 3.5-11 cm long, 3-9 mm in diameter, on peduncles 1-7 mm long, 0.8—1.5 mm in diameter; floral bracts 1-2 (-3) mm high, (1.5-) 2-3 (-3.5) mm wide. Staminate flowers 3 per bract; perianth of 4 parts, these ovate, elliptic, or obovate, the apex acute to rounded, 1.3-1.5 mm long, 0.4-1.1 mm wide, the mar- gin lined with minute to moderately large glands; stamens 4, opposite and free of or basally adnate to the perianth parts, appearing equal to or longer than the perianth, the filaments 1.2-1.4 mm long, the anthers 1.1-1.6 mm long and 1.1-2 mm in diameter, the thecae separate for 35-65% of their length. Infructescences ovoid to ellipsoid, (ll-) 13-25 (-29) mm long, (8-) 9-15 mm in diameter, on peduncles 0.2- 5 mm long, 1.5—2 mm in diameter; scales 3.5—5 mm long, 3.5-5.5 mm wide at the apex, l-l.8 mm wide at the base, the apex moderately 298 to greatly thickened and flat, the terminal lobe-tip truncate and not extended to somewhat extended. Fruits narrowly winged or merely wing- margined, dark brown; bodies elliptic to obovate, 1.7-3.5 mm long, 1.2-3 mm in diameter; wings 1.5—3.5 mm long, 0.2-1 mm wide, firm; persistent styles 0.7—1.2 mm long. Alnus jorullensis is more specialized than A. acumianta, occuring in somewhat less mesic habitats. The leaves are often expanded at the apex and drawn out below, this causing the veins to rise more abruptly toward the tip, the teeth at the apex to be larger than those at mid- leaf, and the lower attenuate or narrowly-cuneate margin to appear entire for a considerable distance above the base. The tree is often \ scrubby in appearance, although it becomes quite large with massive i spreading limbs. This species is closely related to Alnus acuminata, from which it is most likely derived. It shares with it such unique character- istics as the unusual transverse constrictions in its bark and the heavy accumulation of leaf glands. The leaves also often tend toward the ovate form of A, acuminata. It is more specialized, however, in leaf shape, density of the glands, and habitat, as well as other char- acters, and it is quite distinct as a species. 5a Alnus jorullensis H.B.K. var. jorullensis Alnus jorullensis H.B.K., Nov. Gen. Sp. P1. 2: 20. 1817; Alnus acuminata 5 jorullensis (H.B.K.) Regel, Mem. Soc. Nat. Mosc. 13(2): 149. 1861; Alnus jorullensis a typica Regel, Bull. Soc. Nat. Mosc. 299 38(3): 425. 1865. Type: Humboldt 9 Bon land 513;, "crescit in aridis, arenosis montis ignivomi Mexicani, Volcah de Jorullo, altit. 630 hex." (P, holotype; MSCI, photograph of type). 41223 jorullensis var. exigua Fernald, Proc. Amer. Acad. 40: 27. 1904. Type: 4, Egggg s.n., Guanajuato, mtns. of Santa Rosa, April, 1901 (CH, holotype). Alnus jorullensis var. liebmanni Callier, Mitt. Deutsch. Dendr. Ges. 27: 165. 1918. Type locality: "Mexico" (original material not seen). Spreading trees up to 15 (~20) m in height; trunk up to ca. 0.6 m in diameter; bark gray-brown to dark brown, usually corky; young stems medium brown to dark red-brown, dull to slightly lustrous, not glaucous to heavily glaucous; internodes glabrous, sparsely pubescent, or velutinous, moderately to densely glandular; nodes and stems bearing inflorescences very densely glandular; hairs yellowish to brownish; glands medium to large in size, yellowish to brownish. Lenticels of twigs circular to elliptic, 0.5-1.5 mm long, O.3~0.7 mm wide, whitish to yellowish, moderately prominent; leaf scars 1-2 mm high, 1.5—2.7 mm Wide. Buds with stalks 1-3 mm long, 1-2 mm in diameter, sparsely pubescent to moderately villous; bodies 2-7 mm long, 1.5-3 mm in diameter; scales sparsely to moderately pubescent, densely glandular. Leaves narrowly elliptic, oblong, or obovate (rarely ovate); apex acute, obtuse, or rounded; base attenuate, acute, or cuneate; blade (4-) 5-15 (~20) cm long, (2—) 3—7 (8.5) cm wide, light to medium brown or yellow-brown below; margin flat to moderately revolute, double-serrate or sinuate to shallow-lobed and serrate or serrulate, K--- _M.._'_-‘ "" r. . 300 up to 50% entire from the base; major teeth or lobes 9-17 (~22) mm apart, up to 1.5 mm deep, irregular; secondary teeth (l-) 3-5 per cm, 0.1-0.5 mm deep, irregular; abaxial surface glabrous to sparsely pubescent; adaxial surface and veinlets glabrous to sparsely pubes- cent, densely to very densely glandular; pubescence yellowish to brownish; glands medium to large in size, crowded, bright yellow (occasionally brownish or pale yellow). Lateral veins 8-10, (4-) 5-8 (~14) mm apart at mid-leaf, moderately to strongly ascending, sometimes branching once again near the base; cross veins between lateral veins usually poorly developed. Petioles (2-) 6-12 (-23) mm long, l-l.5 (-3) mm in diameter, glabrous to sparsely pubescent, moderately to densely glandular. Pistillate inflorescences borne on non-diverging branches, at anthesis 2—3 mm long, ca. 2 mm in diameter, on peduncles 0.2-4 mm long, 1.2-1.5 mm in diameter; staminate catkins borne in clusters of 3-5, at anthesis 3.5-ll cm long, 3-9 mm in dia- meter, on peduncles 1-7 mm long, 0.8-1.5 mm in diameter; floral bracts 1-2 (~3) mm high, (1.5-) 2-3 (~3.5) mm wide. Staminate flowers with 4 perianth parts, these usually obovate, rounded at the apex, 1.2-1.4 mm long, 0.6-0.9 mm wide, the margin lined with very minute glands; Stamens free or basally adnate to the perianth parts, usually appearing longer than the perianth, the filaments 1.2—1.5 mm long, the anthers 1.1-1.6 mm long and 1.1-2 mm in diameter, the thecae separate for 35- 452 of their length. Infructescences (ll-) 13-25 mm long, 9—15 mm in diameter, on peduncles 0.2-2 mm long, 1.5-1.8 mm in diameter; scales 4.5-5 mm long, 4.5-5.5 mm wide at the apex, 1.5-1.7 mm wide at the base, the apex moderately to greatly thickened, the terminal 301 lobe-tip truncate and not extended. Fruits narrowly winged or merely wing-margined; bodies 2.5-3.5 mm long, 1.7-2 mm in diameter; wings 3-3.5 mm long, 0.5-1 mm wide, firm; persistent styles l-l.3 mm long. Plates 11C, 12E, 14C, 30, and 31. Distribution 238 Ecology: Central Sinaloa and Durango south and east to southern Jalisco, Michoacah, Mexico, and central Veracruz; central Oaxaca. On rocky streambanks, intermittent streams, and moist to moderately dry slopes from elevations of about 1000 to 2500 (rarely to 3000) meters. Usually associated with Pinus, Quercus, or Abies. Figure 26. Common Names: Aile, aliso. Specimens Examined: MEXICO. Distrito Federal. Cafiada de Con- treras, EsEinosa 898_(ENCB, MEXU); Desierto de los Leones, Espinosa 222 (ENCB); San Rafael, Matuda 18818 (F); Bosque de Santa Rosa, Matuda 29219 (*MEXU); Pedregal de San Angel, Rzedowski 814 (ENCB); San Angel, Salazar 8_(MEXU). Durango. Perfil de la Sierra Madre Occidental a lo largo de la carretera Durango-Mazatlah, Martin 35.21: 818 (*ENCB); per- fil de la Sierra Madre Occidental a lo largo de la carretera Durango- Mazatlah, Martin g£_§13 881 (ENCB); from El Salto S along lumber road toward Pueblo Nuevo (about 60 air mi SW of C. Durango), Maysilles 1891 (MICH); N slopes of Cerro Huehueto (Huehuento), S of Huachicheles, about 75 mi W of C. Durango, Maysilles 8911 (MEXU, MICH); Laguna del PrOgreso, 34 road mi N of railroad at Coyotes, nggi11 g 8888 (NY). Guanajuato. Steep rocky mountainsides ca. 3 km E of Santa Rosa in I humid oak forest, McVaugh 24221 (MICH); 30 km a1 E de San LUIS de la Plate 30. Holotype of Alnus 302 'orullensis H.B.K. (photograph courtesy of John H1‘EEEEZETT‘ Plate 30. 304 Plate 31. Representative specimen of Alnus jorullensis H.B.K. var. jorullensis. 305 MICHIGAN STAT! 232803 UNIVERSITY HERBARIUM V’ , 2!; PLANTS OF MEXICO Alnus jorullensis HEX. arr/ MICHOM‘AN: 8 kilometer. north of Uruapan along the ruadalda; alavatlon 2,000 meters. Tree, 5 meter- high; trunk 15 cm. in dis- er; bark amoth with :ranaveue con- atrictlona. Occasional. John J. Furlow November 28, 1971 No. 330 Baal-Wuhan "an MICHIGAN STATE l'NIVEISITY Plate 31. Figure 26. 306 Distribution of Alnus jorullensis H.B.K. jorullensis. var. , wwrw-r no. Figure 26. 308 Paz, Rzedowski 9082 (ENCB). Guerrero. Cerro Azul, Distr. Mina, Hinton 14943 (*NY, WTU); Petlacala, Distr. Mina, Hinton 15406 (MICH, *NY, UC, WTU). Hidalgo. San Vicente, Fisher 46133 (RM); San Vicente, Fisher s.n., Aug. 16, 1937 (F, NY, RM); Jacala, Kenoyer 641 (*MO); Omitlan-Huasca, Miranda 4472 (*MEXU); hills, Cuyamaloya, 8000 ft, Pringle 19188 (F, MEXU, MSC, *MO, NY, UC); ca. 3 mi S of Tepiji del Rio, M8 Gregory E (MEXU, MICH, UC). Jalisco. Northern slopes of Nevado de Colima, McVaugh 19188 (MICH); Nevado de Colima above the sawmill called Piedra Ancha and just E of the first great cafion W of the sawmill site, McVaugh 11818 (*MEXU, MICH); N slopes of the Nevado de Colima, W summit of the N ridge, McVaugh 18881 (MICH); Sierra de Manantlan (15~20 mi SE of Autlan), near Aserradero El Cuarton, McVaugh 18888 (MICH); near El Carmen (ca. 40-50 km W of Ayutla), McVaugh fl (MICH); Real Alto, trail to Poso Hedionda, WE (F, MICH, M0, *NY, UC); camino de Atenquique a1 Nevado de Colima, Rzedowski 12882 (*ENCB, MEXU, MSC); mountains E of Mamantlan, about 15 mi SSE of Autla’n by way of Chante, Wilbur é Wilbur % (MICH). Me’xieo. Foothills of Ixtaccihuatl, 2222.3;EL’ Jan. 5, 1899 (MICH); Pantoja, Distr. Temascaltepec, Hinton 8848 (NY); La Carbonera, cerca de Rio Frio, Madrigal é.!£li.§;§;’ June 28, 1961 (MEXU); Ixtaccihuatl, Pur us 1128 (F, NY, UC); Ixtaccfhuatl, Purpus 888;, in 1903 (UC). Michoacan. 13 km W of Patzcuaro on a dry hillside with pines and oaks, Furlow 818 (MSC); 13 km W of Pdtzcuaro along the road to Uruapan, Furlow 819 (MSC); 8 km N of Uruapan along the roadside, Furlow 889 (MSC); Tancitaro, Distr. Uruapan, Hinton 18811 (NY, WTU); steep mountainside NW of Aguililla, ca. 5-7 km S of Aserradero Dos AguaS, MEXEEED.22§§§ 309 (ENCB, MICH); Mt. Tancitaro, Nelson 8811 (NY). Morelos. Near Tres Marias, Cuernavaca Railway, Dudley §;E;’ Jan., 1899 (DS); Campo Tur— ista km 60, carrt. Mex. Cuernavaca, Gallegos Harking 481 (MEXU); 2% mi S of Tres Cumbres on old hwy. 95 to Cuernavaca, Gibson 1988 (*MEXU); 1 km al N de Cuajumulco, Palacios 98, §;EL’ Feb. 15, 1965 (*ENCB, MSC); carretera nueva Mexico-Cuernavaca, km 61, Palacios ELEL! Nov. 14, 1964 (ENCB); a1 S de Tres Cumbres, km 54 carretera Mexico-Acapulco, Palacios ELEL! Dec. 31, 1964 (ENCB); km 60 carretera Mexico-Cuernavaca, Palacios s.n., Feb., 1954 (ENCB); 10 km al N de Cuernavaca, Rzedowski 18480 (ENCB); Coajomulco, 5 km a1 S de Tres Marias, 2252.2: 189 (ENCB). Oaxaca. Cerro San Felipe, Conzatti 4914 (*MEXU); mountains along rt. 175, 12 km N of Ixtlah de Juagez on the road to Valle Nacional, 818g_1998 (MICH); Loma del encino, E1 Carrizal, Yolox, EEX.EEE.Z§§ (*ENCB, MEXU); summit ridge, Sierra de San Felipe, 7000 ft, Pringle 19149 (DAO, *ENCB, MICH, MSC, UC). Puebla. 38 km W of the city of Orizaba on a steep hillside With pines, Furlow 889 (MSC); Campo Experimental, San Juan Tetla, MEX E§2.l2§2,(ENCB: MSC); Mt. Orizaba, Purpus 8998 (NY, UC); La Cruz de Sn. Miguel, Honey, XEllEHE Martinez 819 (ENCB); ca. 4.5 km E Rio Frio, Mun. Tlahuapan, HEEEE.§9£ (ENCB). Sinaloa. 7 mi W of Santa Rita along a side road E of the Los Hornos to Surutato road, Breedlove 18818 (*MICH); along road to Surutato, 3 mi SE of Los Ornos, Breedlove_9 Thorne 18814 (MICH); San Ignacio, Gonzalez Ortega 1181 (MEXU); El Batel, 70 km NE of Mazatléh, Pitekla 198 (*UC). Zacatecas. Sheltered east-facing Valley, steep mountainsides near summits ca. 20 km westward toward Tlaltenango from the road junction S of Jalpa, McVaugh 25638 (*MICH). The name Alnus jorullensis, like A, acuminata, has been often 310 i l 1‘ misapplied to various Latin American alders. In recent years the ‘ concept of this species has become somewhat less confused, but it is still difficult to distinguish A. jorullensis var. jorullensis from the other species using present keys. Standley (1920) in his key to the species of élBEE in IESEE 888 Shrubs 88 Mexico separates A, JEEELT lensis from all the other species in the first couplet by stating that its leaves are ”densely covered beneath with yellow wax glands" while the other species have "leaves without glands beneath or the glands remote and inconspicuous." As explained above, this second lead really fits none of the Latin American species of the genus, even though the glands of 8. jorullensis var. jorullensis are usually much larger, more crowded, and brighter yellow than those of the other species, and many have doubtless been confused by this when trying to use Standley's key. The type of 81888 jorullensis (Plate 30) shows a very distinctive elliptic to obovate leaf shape compared to the usually ovate form in A. acuminata. Other unique characters usually present include the heavy accumulation of large, bright-yellow leaf glands, the ascending lateral veins and smooth wavy margins of the leaves, and the scrubbier form of the trees. The type of 81888 jorullensis was said, by Humboldt, Bonpland, and Kunth (1817), to have been collected on Volcah de Jorullo, Micho- acah. This locality today, however, does not seem a likely habitat for 81888, with its elevation of only 630 meters, semi—arid climate, and vegetation including such plants as acacias and palms. On a trip to the site in 1971, I was unable to locate any alders, though I col- lected A, jorullensis var. jorullensis at nearby Patzcuaro, in a forest 311 of Pinus at an elevation of about 2000 meters. 3 41888 jorullensis var. jorullensis occurs in relatively warmer habitats than the other Mexican alders, in the pine-oak zone, at elevations generally below 2500 meters (described by Goldman, 1951, as the "arid lower tropical subzone"). Where its range overlaps with that of the following variety, 8. jorullensis var. firmifolia, hybridi- zation appears to occur, resulting in a swarm of intermediate plants. It does not reach Central America as does var. firmifolia. 5b. Alnus jorullensis var. firmifolia (Fern.) Furlow, comb. 88 stat. nov. Alnus firmifolia Fernald, Proc. Amer. Acad. 43: 61. 1907. Type: Pringle 10040, "Federal District, mountains about Cima Station, ‘ alt. 9800 ft., 30, August, 1905" (GH, holotype; DAOI, FI, MICHI, MSCI, NYI, UC!, U81, Wlsz, isotypes). Spreading trees up to 20 m in height; trunk up to 1.8 m in diameter; bark brown, smooth to corky, often broken by deep transverse con- strictions encircling the stem; young stems light to dark red-brown, dull to slightly lustrous, rarely glaucous; internodes mostly glabrous (occasionally sparsely pubescent), moderately glandular; nodes and stems bearing inflorescences densely glandular; glands small, brown to dark brown. Lenticels of twigs circular to elongate, 0.5-1 mm long, 0.3-0.7 mm wide, yellowish to brownish, somewhat prominent; leaf scars l-l.8 mm high, 1.7—3.5 mm wide. Buds ellipsoid; stalk 1-2 mm long, 0.7-2 mm in diameter, more or less glabrous; body 3-4 mm long, 1.5-3 mm in diameter; scales glabrous. Leaves elliptic, 312 elliptic-oblong, or obovate (rarely ovate), the apex usually obtuse or rounded (but occasionally acute), the base acute to broadly cuneate; blade (3.5-) 4.5—ll (~13) cm long, (2.5-) 3.5- 5.5 (~7) cm wide, medium to dark green or brown below; margin flat, double-serrate, serrate, or serrulate, up to 25% entire from the base; major teeth (7-) lO-l7 mm apart, up to 3 mm deep, very ir— regular; secondary teeth (3-) 5—8 per cm, 0.2-1.1 mm deep, slightly uneven to irregular; adaxial surface and veinlets moderately to rather densely glandular; leaf pubescence yellowish to brownish; leaf glands small to medium in size, yellowish, brownish, or dark brown. Lateral veins 7-10 (~15), (4-) 6~lO (~11) mm apart at mid-leaf, slightly to moderately ascending, sometimes branching once again near the base; cross veins between lateral veins usually well-developed. Petioles ‘ (4-) lO-lS (~18) mm long, l-l.5 (~2) mm in diameter, glabrous or sparsely pubescent, moderately to densely glandular. Pistillate inflorescences borne on branchlets diverging strongly from the main axis, at anthesis ca. 4 mm long, ca. 2.5 mm in diameter, on peduncles 0.2-1.5 mm long, ca. 1 mm in diameter; staminate catkins borne in clusters of 2—4, at anthesis (3—) 5-6.5 cm long, 7-8 mm in diameter, on peduncles 3-5 mm long, l~l.5 mm in diameter; floral bracts 1-2 (~3) mm high, (1.5-) 2-3 (~3.5) mm wide. Staminate flowers with 4 peri- anth parts, these elliptic to obovate, acute at the apex, 1.2—1.5 mm long, 0.4-0.8 mm wide, the margin lined with small to moderately large glands; stamens basally adnate to the perianth parts, appearing equal to or longer than the perianth, the filaments 0.6-0.9 mm long, the anthers 1.2-1.6 mm long, 1.3-1.7 mm in diameter, the thecae 313 separate for 40-65% of their length. Infructescences 11-28 mm long, 8-13 mm in diameter, on peduncles 0.2-5 mm long, 1.5-2 mm in diameter; scales 3.5-5 mm long, 3.5-5 mm wide at the apex, l-l.7 mm wide at the base, the apex moderately thickened and flat, the terminal lobe-tip truncate to rounded and not extended to somewhat extended. Fruits narrowly wing-margined, dark brown; body 1.7—2.2 mm long, 1.2-1.7 mm in diameter; wings 1.5—2.5 mm long, 0.2-0.7 mm wide, firm; per- sistent styles 0.7-1 mm long. Plates 3B, 4C, 11B, 14B, and 32. Distribution and Ecology: Central Durango south and east to Jalisco, northern Michoacah, Mexico, and central Veracruz; central Oaxaca; southeastern Guatemala. Along rocky streams, intermittent streams, and slopes near streams or washes from elevations of 2800 to 3800 meters (occasionally as low as 2200 meters). Usually ‘ associated with Pinus, Quercus, or Abies in open woodland associations. Figure 27. Common Names: Aile, aliso. Specimens Examined: GUATEMALA. Dept. Huehuetenango, Sierra de los Cuchumatanes, about 28 mi from Huehuetenango, slopes of Cerro Chemal, Hawkes 32.21: 1148 (*F); Cumbre de la Sierra de los Cuchu— matanes, between the first cumbre and La Pradera, Standley 81189 (*F); alpine area, vicinity of Tojquia: Sierra de los Cuchumatanes, Steyer- E35E.§ng§ (*F); between Tojquia and Caxin bluff, summit of Sierra de los Cuchumatanes, Steyermark 89141 (*F). MEXICO. Chiapas. Paraje of Matsab, Mun. of Tenejapa, £22.1221 (NY). Distrito Federal. oad to Ajusco, Bell St Duke 16777 (MICH, UC); 1.5 km a1 SE de la 314 Plate 32. Isotype of Alnus firmifolia Fern. (= Alnus w var. firmifolia (Fern.) Furlow). 315 C. G. PRINGLL‘. PLA NT.‘E M if X ICA .‘\' ~15 “‘05 r3231; ::::-:::1 IMHO Hum Ilrmll‘ulia. I'n-rnald n. sp. y_,.,.,,r w Plate 32. ' 11 316 Figure 27. Distribution of Alnus j8£g11§3§}§_var- £E£21£2133(Fenh) Furlow. 317 n2 Figure 27. 318 Estacidn La Cima, 8588 Cisneros s.n., Oct. 1, 1966 (ENCB); Cafiada de Contreras cerca del 2d dinamo, Espinosa 849 (ENCB); La Cima, near the railroad station, Furlow 819 (MSC); ca. 2 km SSW of La Cima RR Station on either side of old highway 95, on top of Serjana de Ajusco, 11118 E£.El‘ 176 (MICH); Guardita Ajusco, May Nah AM~14 (ENCB, MEXU); mountains about Cima Station, Pringle 19949 (DAO, F, MICH, MSC, NY, UC, US, WIS); parte occidental del Pedregal de San Angel cerca del Rio Eslava, Rzedowski 1189 (ENCB); ladera NE del Cerro Ajusco, 3227 889811_11948 (ENCB, MSC). Durango. North slopes of Cerro Huehueto (Huehuento) S of Huachicheles, about 75 mi W of C. Durango, Maysilles 9114 (MICH, NY). Guerrero. Teotepec, Distr. Mina, Hinton 14194 (MICH, NY); Teotepec, Distr. Mina, Hinton 14899 (NY, UC). Hidalgo. lerca de Tezuantla, Espinosa 491 (ENCB); 4.5 km a1 E de Real del Monte, Mn. de Epazoyucan, Gonza1ez Quintero 1949 (ENCB, MSC); 2 km a1 E de l Guajolote, Mun. de Epazoyucan, Rzedowski 11411 (ENCB). Jalisco. ortheastern slopes of the Nevado de Colima, above Cafion de Leoncito, cVaugh 11848 (MICH); Nevado de Colima, EEEE.§;EL! Jan. 10, 1965 MICH); La Joya, en la ladera E del Nevado de Colima, Rzedowski 19881 ENCB, MEXU); camino de Atenquique a1 Nevado de Colima, Rzedowski 9888 (ENCB). Mexico. Paraje Provincial, Mount Popocatépetl, iffi.§fl£§§ (UC); below Ojos de Agua, Nevado de Toluca, 88118_84989 1C); Volcah de Ixtaccfhuatl, Dudley ELEL! Jan., 1899 (DS); 1 km . N de Llano Grande, sobre la ladera S del Telapon, Espinosa 848 NCB); 19 km E of Amecameca along the road to Popocatépetl, Furlow 9 (MSC); 19 km E of Amecameca along the road to Popocatépetl, rlow 321 (MSC); Nevado de Toluca, Furlow 323 (MSC); Nevado de 319 Toluca, zona este, Guzméh E, GH—1141 (ENCB); Cerro Papayo, Rio Eric, Matuda 28227 (*MEXU); Rfo Fri/o, Medellin 1. 300 (ENCB); Llano Grande, Mun. de Chalco, Perez s.n., Sept. 11, 1966 (ENCB, MSC); Popo- catépetl, Purpus 1793 (NY, UC); on Nevado de Toluca, Rose é Painter l§§4 (*US); Netzcualango, vertiente NW del Ixtaccfhuatl, Rzedowski 31§1§ (ENCB, MSC). Michoacah. Cerro Burro, 11.2 km S of Opopeo, Furlow 32g (MSC); ca. 18 mi S of Patzcuaro, E1Eg é Soderstrom 221g (MEXU, MICH, NY, *UC); Cerro Tancitaro, Leavenworth §9_9 (F, MICH, NY); Mt. Tancitaro, Nelson §§2§ (NY). Morelos. Near Tres Marias, 23111 24211 (UC); vicinia Cuernavacae, Montes Las Tres Marias, Fr6derstr6m é Hultéh 339 (NY); Yautepec, £2223_M25g3.§1 (ENCB); between Huitzilac and Tres Cumbres, Hatheway'11§4 (MO); Laguna de Zempoala, Mitastein 3g (ENCB, MSC); Tres Marias Mts., Pringle 1% (F, DAO, M0, MSC, *UC); Parque Nacional Lagunas de Zempoalas, guiram 22 (MEXU); Lagunas de Zempoala, S1323 é Gregory 1929 (*MEXU, MICH). Oaxaca. Mountains N of leah de Juarez, lumber road along ridges of sierra, departing highway 13.5 mi N of Ixtlah, Anderson §_Anderson 2412 (*ENCB, MICH); Jicinity of Cerro Zempoaltepetl, Hallberg §24 (MICH); vicinity of Ierro Zempoaltepetl, Hallberg 292 (*MICH); Sierra de San Felipe, ’ringlel1gg4g (DAO, *ENCB, MICH, MSC, UC). Puebla. La Cumbre, acapoaxtla, XEl3.§‘.lQ§l (ENCB). Veracruz. P. Orizaba, Miranda L2 (MEXU); 9 mi E of Perote, Spetzman 1412 (MEXU). From its morphology, habit, and habitats, this taxon is clearly nspecific with Alnus jorullensis H.B.K. and probably best considered Variety of that species. Morphologically, var. firmifolia differs from var. jorullensis 320 in having more irregular and obovately-shaped leaves, smaller, darker, and fewer glands on the lower leaf surface, and a larger mature size. It occurs at higher elevations in generally similar (though necessarily cooler and moister) habitats (cf. Goldman, 1951). The leaves are usually more coarsely-toothed, especially near the apex. Where the two varieties occur together, they appear to hybridize readily, producing many variable and intermediate individuals which are often difficult to assign to one variety or the other. These intermediate putative hybrids usually occur at intermediate ele- vations between the normal conditions for the two typical varieties. In the citation of specimens, those approaching one variety or the other are listed there, even though they may be intermediate. This variety occurs as far south as Guatemala, though material from only a single area in that country was seen. 6. Alnus incana (L.) Moench Alnus incana (L.) Moench, Meth. Pl. p. 424. 1794; Betula alnus B incana L., Sp. P1. 983. 1753; Betula incana (L.) L.f., Suppl. Pl. $17. 1781. 1‘. ‘ Trees or shrubs up to 25 m in height; trunks up to 3.5 dm in liameter, erect or ascending, bark light to dark gray, reddish, or *1 fiurplish-brown, smooth or broken into plates (on old individuals); ioung stems light to dark red-brown, dull to slightly lustrous, [lightly to heavily glaucous, usually without conspicuous resinous 321 coating, not differentiated into long and short shoots, without longitudinal ridges originating at the nodes; internodes sparsely pubescent to velutinous or tomentose, moderately to densely glandu- lar; nodes and stems bearing inflorescences very densely glandular; hairs yellowish to brownish to dark brown; glands small to medium in size, pale yellowish to brown; lenticels of twigs circular to ellip- tic, O.2-l.2 mm long, 0.2—0.7 mm wide, whitish to yellowish, moderately prominent; leaf scars 1-2 mm high, 1.5-2.5 mm wide, the bundle scars inconspicuous to somewhat prominent. Buds ellipsoid, acute to slightly rounded at the apex, moderately to heavily resin-coated; stalk 2-4 mm long, 1-2 mm in diameter, sparsely pubescent to velutinous, densely glandular; body 3—7 mm long, 1.5-3 mm in diameter; scales 2 (—3), stipular, equal, valvate, moderately villous to velutinous, glandular; pubescence and glands usually obscured by the resin coating. Leaves ovate to oblong-ovate, elliptic, or nearly orbicular (rarely some- jwhat obovate), the apex acute to obtuse or rounded; base attenuate, ‘broadly cuneate to acute, obtuse, or rounded, sometimes oblique; blade (2.5-) 4—9 (-11) cm long, (l-) 3-9 cm wide, medium to dark green and dull to moderately lustrous above, light to medium green and dull (sometimes slightly to heavily glaucous) below, membranaceous, char- }aceous, or somewhat coriaceous; margin flat to slightly revolute, slightly to moderately thickened, double-serrate to nearly serrulate; hajor teeth (4-) 8-13 (-l6) mm apart at mid-leaf, up to 7 mm deep, iegular; secondary teeth 3-ll (-l3) per cm, 0.3-1.2 mm deep, regular; lbaxial surface glabrous, sparsely pubescent, or somewhat pilose, sparsely to moderately glandular; adaxial surface and veinlets 322 glabrous to velutinous or tomentose, sparsely to moderately glandular, slightly to moderately resin-coated; major veins and vein axils near the base tomentose to wooly—pubescent; pubescence whitish to yellowish or brownish; glands small to medium, whitish, yellowish, or brownish. Lateral veins 7-l4, (3-) 4-9 (—10) mm apart at mid-leaf, straight to slightly ascending, sometimes branching once again, especially near the base, terminating in major teeth at the margin; cross veins be- tween lateral veins poorly to well developed. Petioles (4-) 8-17 (-25) mm long, (0.5—) 0.8-1.5 mm in diameter, moderately villous to tomen- tose, sparsely to moderately glandular. Stipules ovate, elliptic, or obovate, the apex acute to obtuse, 5—10 mm long, 2-3.5 mm wide, green, glabrous to velutinous, the hairs yellowish, moderately glandular, the glands yellowish. Pistillate inflorescences borne in racemose groups of (2-) 3-6 (-8) on short branchlets non—divergent to strongly di- vergent from the main axis, produced during the previous growing season, erect, ovate to elliptic, at anthesis 2—5 mm long, 1.2-3 ‘mm in diameter, on peduncles 0.2-2 mm long, 0.8-2 mm in diameter; Estaminate catkins borne in one or more racemose clusters of 2-4 at £the end of the main branch above the pistillate inflorescences, iproduced during the previous growing season, pendent during both 1dormancy and anthesis, at anthesis 3-7 {—10) cm long, 5-9 mm in ,diameter, on peduncles 1—18 mm long, 0.8—2.2 mm in diameter; floral 1 {bracts l-2 (-3) mm high, (1.5-) 2-3 (-3.5) mm wide. Staminate flowers 1 1? per bract; perianth of 4 parts, these elliptic to obovate, the apex rounded, 1.2-1.9 mm long, 0.4-1.3 mm wide, the margin lined with small 1 1 to medium or large glands; stamens 4, opposite and basally to nearly 323 completely adnate to the perianth parts, usually appearing somewhat shorter to equal in length to the perianth; filaments 0.4-1.2 mm long; anthers 0.8-1.4 mm long and 0.7-1.2 mm in diameter, the thecae separ- ate for 30-55 Z of their length. Infructescences ovoid to ellipsoid, (6-) 10-17 (—22) mm long, (6-) 8-11 (-l4) mm in diameter, on pedun- cles 0.2-6 (-8) mm long, (0.8-) l-l.5 (-2) mm in diameter; scales 3-5.5 mm long, 3-5.5 mm wide at the apex, 0.8-1.5 mm wide at the base, the apex thin to moderately thickened and flat, the terminal lobe-tip acute and depressed to somewhat or very extended. Fruits narrowly winged or wing-margined, brown; body ovate, elliptic, or obovate, 2.5-3.5 mm long, 1.2-2.5 mm in diameter; wings 2.5—3.5 mm long, 0.4-0.8 (-l.2) mm wide, firm to coriaceous; persistent styles 0.5-1.2 mm long. Alnus incana was one of two alders included by Linnaeus in Species Plantarum as varieties of his Betula 21223. It occurs in a circumpolar distribution, occupying habitats ranging from the lowlands to nearly subalpine conditions. Among several morphologically and geographically well-marked segments of the species are subspecies rugosa of north- eastern North America and subspecies tenuifolia of western North America. Although the typical subspecies of Europe becomes a tree, Alnus incana in the New World seldom attains a stature greater than that of a large shrub. Fernald (1945b) emphasized this in arguing for separate specific status for the North American forms. There are too many affinities between New and Old World plants, however, to con- tinue to separate them as different species. Alnus incana, especially in Europe and Asia, is extremely 324 variable (cf. Hultéh, 1971), and this has led to a profusion of species, variety, and form names applied to it. Leaf shape is usually ovate, but it ranges from this form to obovate or orbicular. Pubescence and glaucousness of the leaves are likewise quite variable, as are habit and bark characters. In spite of this diversity, however, there is usually little confusion in determining the species. Best development, as with greatest variability, is in Europe, pointing to an origin there, although this species has been greatly affected by Pleistocene glaciation, complicating the tracing of its history. 6a. Alnus incana (L.) Moench subsp. rugosa (DuRoi) Clausen £422§_incana (L.) Moench subsp. rugosa (DuRoi) Clausen, Mem. 3 Cornell Univ. Agr. Expt. Sta. 291: 8. 1949; Betula ElEEE (rugosa) DuRoi, Diss. Inaug. Obs. Bot. 31. 1771; Betlua rugosa Ehrhart, Beitr. Naturk. 3: 21. 1788; A1243 rugosa Sprengel, Syst. Veg. ed. _16, 3: 848. 1826; A1233 glutinosa 6 serrulata lusus c. rugosa Regel, jMem. Soc. Nat. Mosc. 13(2): 165. 1861; 41323 serrulata B rugosa Regel, iBull. Soc. Nat. Mosc. 38(3): 432. 1865; 41EE§_rugosa var. a typica Winkler, Pflanzenreich 19(4.61): 119. 1904. Type location: ”habitat ‘in America septentrianali” (original material not seen). Betulafé1ggs glauca Marshall, Arbust. Am., p. 20. 1785; Betula incana a glauca Aiton, Hort. Kew., p. 339. 1789; A122: glauca Michaux, 1Hist. Arb. For. Sept. 3: 320. 1813; 41223 incana a glauca Aiton f., :Hort. Kew. ed. 2, 5: 259. 1813; 41333 incana var. glauca Callier, Fedde Rep. Sp. Nov. 10: 235. 1911. Location of type unknown. 325 1 41233 incana B americana Regel, Mem. Soc. Nat. Mocs. 13(2): 155. 1861; 41223 incana var. k americana Winkler, Pflanzenreich 19(4.61): 123. 1904; 4132: rugosa var. americana Fernald, Rhodora 47: 350. 1945; A1225 americana Czerepanov, Notul. Syst. Herb. Inst. Bot. Kom. Acad. Sci. U.R.S.S. 17: 103. 1955. Type: Sartwell, ”Penn Yan in Nordamerika" (LE?, not seen). Alnus argentea hort. ex Petzold & Kirchner, Arb. Musc., p. 600. 1864, pro syn. A1233 canadensis hort. ex Winkler, Pflanzenreich 19(4.61): 119. 1904, pig gLn. £1222 oblongata hort. ex Winkler, Pflanzenreich 19(4.61): 119. 1904, 323_Willdenow, Sp. Pl. ed. 4, 4: 335. 1805, p£2_§y3. Alnus incana var. g1auca f. tomophylla Fernald, Rhodora 16: 56. ‘ 1914; Alnus incana var. tomophylla Rehder, Man. Cult. Trees and Shrubs, 1p. 147. 1927, erroneously attributed to Fernald; Alnus rugosa var. americana f. tomophylla Fernald, Rhodora 47: 353. 1945. Type: Fernald §_Wiegand 5303, Newfoundland: border of a wet thicket, Norris 1Arm, Aug. 21, 1911 (GH1, holotype). Alnus rugosa var. typica forma emersoniana Fernald, Rhodora 47: 1 __ _— ‘347. 1945. Type: Fernald §_Bartlett 14, Massachusetts: Round Pond, 1Tewksbury, Middlesex Co., Apr. 14 & Oct. 14, 1906 (GH, holotype). ‘ figflgs_rugosa var. americana forma hypomalaca Fernald, Rhodora 147: 353. 1945. Type:. Weatherby §_Weatherby 291:, New Brunswick: :Seal Cove Brook, Grand Manon, Charlotte Co.; July 24, 1941 (Gfil, holotype). Spreading shrubs up to 6 (-9) m in height; trunks up to ca. 15 cm 326 in diameter, the bark dark gray, reddish, or purplish-brown, with large, conspicuous whitish or grayish elliptic or elongate lenticels; young stems light to dark red-brown, dull; internodes sparsely pubes- cent to velutinous; lenticels of twigs 0.2-1.2 mm long, 0.2-0.7 mm wide, moderately prominent, whitish to yellowish; leaf scars 1-2 mm high, 1.5—2.5 mm wide, the bundle scars somewhat prominent. Buds acute (or sometimes slightly rounded) at the apex; stalk 2-4 mm long, 1—2 mm in diameter, sparsely pubescent to velutinous; body 3-7 mm long, 1.5-3 mm in diameter. Leaves ovate to elliptic (rarely obovate), the apex acute to obtuse; blade (2.5-) 4—9 (-11) cm long, (1.3-) 3-7.5 cm wide, medium to dark green and dull above, usually medium green and dull below, often slightly to heavily glaucous below, chartaceous; margin flat to slightly revolute, slightly to moderately thickened, double-serrate to nearly serrulate; major teeth (4-) 8-13 (-15) mm 1 lapart at mid-leaf, up to 4 mm deep, regular; secondary teeth 6-11 1(-13) per cm, 0.3—1 mm deep, regular; abaxial surface glabrous to sparsely pubescent, sparsely to moderately glandular; adaxial sur— face and veinlets glabrous to velutinous, sparsely to moderately iglandular; pubescence yellowish to brownish; glands small to medium Lin size, yellowish to brownish. Lateral veins 11-14, (3-) 4-9 (~10) ‘ apart at mid-leaf; cross veins between lateral veins well de- feloped. Petioles (4-) 8—17 (-20) mm long, 0.5-1.5 mm in diameter, ioderately villous to tomentose, sparsely to moderately glandular. Itipules 6—10 mm long, 2-3.5 mm wide. Pistillate inflorescences 327 1.2—2 (-2.5) mm in diameter, on peduncles 0.2-2 mm long, 0.8-1 mm in diameter; staminate catkins borne in one or more clusters of 2-4, at anthesis 2-7 cm long, 5-8 mm in diameter, on peduncles 1—9 (-11) mm long, 0.8-2 (-2.2) mm in diameter. Staminate flowers with 4 perianth parts, these elliptic or obovate, the apex rounded, 1.4-1.9 mm long, 1.1—1.3 mm wide, the margin lined with medium to large glands; stamens basally adnate to the perianth parts, usually somewhat shorter to equal in length to the perianth; filaments 0.4-0.9 mm long, anthers 0.9-1.2 mm long and 0.8—1.2 mm in diameter, the thecae separate for 30—45% of their length. Infructescences (6-) 10—17 mm long, (6-) 8-11 mm in diameter, on peduncles 0.2-6 (-8) mm long, l—l.5 (-2) mm in diameter; scales 3.5-4.5 mm long, 3—4 mm wide at the apex, 0.8—1.5 mm wide at the base. Fruits narrowly winged or wing-margined; body elliptic to obovate, 2.5—3.5 mm long, 1.3-2.5 mm in diameter; wings 2.5-3.5 mm long, 0.4-0.8 (—1) mm wide, firm to coriaceous; persistent styles 0.5—1 (-l.3) mm long. Plates 2A, 4B, 7A, 9A, 13D, 180, 33, and 34. Distribution and Ecology: North—central Manitoba east to south- ern Labrador, south to southeastern North Dakota, northeastern Iowa, iorthern Indiana, central Ohio, and southern Pennsylvania. Stream- >anks, lakeshores, borders of bogs, edges of swamps, wet fields and :wales, often forming fairly dense thickets. Found at elevations of early sea level along the north Atlantic coast to about 825 meters 1 the Appalachian highlands. Figure 28. Common Names: Speckled alder, tag alder, swamp alder, hazel alder, VIV 1H 328 Plate 33. Representative specimen of Alnus incana subsp. Eggggg (DuRoi) Clausen. 329 2286112 SITY HERBARIUM PLANTS ()F HINNESW Alnu! rulon (Du lat) Sauna. CLMNATLR 00.: “k. [tau-l Stat. Park. In a lav. no: wot u the edge of wood- nunr the north park ontnnce. shrub, 10 n. high. lequonl. John J. Furlow Jun:- 17. 1971 No. l Dal-mt- lubr— mud "A" Ull'llm Plate 33. 330 Plate 34. Specimen of Alnus incana subsp. £59533 (DUROi) Claug'eflr-I Holotype of Alnus incana var. glauca f. M er ' 331 “‘ faaaWMM/I/filuud uuuuuuuuuu Plate 34. Figure 28. 332 Distribution of Alnus incana subsp. rugosa (DuRoi) Clausen. Figure 28. 334 hoary alder, aulne blanchatre (Quebec). Representative Specimens: CANADA. Labrador. Paradise River, Sandwich Bay, Bishop 315 (CH); platiere pres de la route de la riv. Hamilton, rive N du lac Gabbro, Dutilly_§ Lepage 41111 (DAO); Ash- uanipi River, E side about 1 mi N of Ashuanipi, Harper 3210 (CAN); Hamilton River, Twin Falls area at Unknown River, Hustlich g Kallio 2_(CAN); Goose Bay, Schofield 181 (DAO). Manitoba. Lac Du Bonnet, Breitung Z41; (DAO); Sandilands Forest Reserve, Breitung 2251 (DAO); E end Singoosh Lake, Duck Mts., Halliday 66-1933 (CAN); Lake Winni- peg, ng££_§1p;, Aug. 28, 1889 (*NY); Hayes River system, 130 mi NE of Lake Winnipeg, Oxford House, Scoggan_§144 (CAN); Wekusko Lake, gravel roadside to Snow Lake, near portage to Berry Bay, Scoggan ézgfl (CAN). Newfoundland. SE of Tompkins, 1 mi N of St. Andrews, Codroy Village, Bassett 84g (DAO); region of Port a Port, Table Mountain, Fernald & St. John 19818 (CH); valley of the Exploits .—..——-——_— River, Grand Falls, Fernald é Wiegand 5301 (CAN, GH, NY); valley of the Exploits River, Norris Arm, Fernald §_Wiegand 5303 (CAN, *GH, NY, US); Humber Arm, Bay of Islands, Mclver's Cove, Fernald 31.31. 1646 (*GH, NY); Steady Brook (E of), banks of the Humber River, Rouleau 2729 (CAN, DAO); Glenwood (about 3 mi NW of), Salmon River shores, Rouleau 6220 (DAO). Northwest Territories. Keewatin District: Robinson Portage, Peeble é Peeble 1§_(US). Nova Scotia. Antigonish Co.: Merland, Smith 31 31. 13598 (CAN). Cape Breton Co.: by Pottle‘s Lake, North Sydney, Bissell §_Linder 21020 (GH). Cumberland Co.: Meteghan, Fernald §_Long 21016 (GH). Halifax Co.: Musquodobolt Harbour, Rouseau 35623 (CAN). Lunenburg Co.: wet thickets and swales 335 back of brackish shore of Lahave River, Bridgewater, Fernald_§_1gpg 33112 (A, CAN, *GH, MO); Aldersville, Sm1ph‘2033 (DAO). Pictou Co.: New Glasgow, 22:3 E Gorham 45.361 (DAO). Shelburne Co.: rocky shores of Deception Lake, Fernald §_1gpg'12111 (GH). Victoria Co.: Great Bras d'Or, Iona, Scamman 4211 (CH). Yarmouth Co.: Sloane Lake, Pleasant Valley, Fernald 25.31! 2191: (CH). Ontario. Algoma Distr.: N shore of Lake Superior, 10 mi SE of White River, 932w_11§§ (MSC). Bruce Co.: Stokes Bay, Krotkov 8259_(NY). Carleton Co.: Haycock Island, Shirley Bay, Ottawa River, Gillett §_Graham 2125 (DAO); Dow's Swamp, Ottawa, Minshall 2191 (*DAO). Frontenac Co.: Thousand Islands, Bicknell 3449_(NY). Glengary Co.: W of the rapids at Hydro, on Raisin River, Charlottenburg Twp., Epgp_z19_(DAO). Grenville Co.: Prescott (2 mi due N of centre-town), 2232_1§2§4_(DAO). Kenora Distr.: Moose Factory, Dutilly E Lepage 12122_(A); Lake River, 54°20, N, 22: tilly & Lepage 16718 (CAN, DAO); Lake of the Woods, Huber 988 (UC). Lanark Co.: on Jock River, EEEE.EE.E£T 549 (DAO). Manitoulin Co.: Gore Bay, Manitoulin Island, Grassl g242_(NY). Muskoka Co.: Milford Bay, Minshall 3945 (DAO). Nipissing Distr.: Algonquin Park, Macoun _s;p:, Aug. 20, 1900 (CAN). Parry Sound Distr.: damp rocky ground near Port Sydney, 22253.33: (NY). Prince Edward Co.: Weller Bay, SEEK-343;, in 1956 (DAO). Rainy River Distr.: along C.N.R. 3/4 mi N of Rainy R. Town, Garton'8214 (DAO). Renfrew Co.: Cushing Island, Ottawa R., 2% mi NW of Pembroke, Minshall 3111 (DAO). Russell Co.: Earlsbad Springs, Sgpp_3213 (DAO, RM). Simcoe Co.: just S of Hall's 90nd, Midland, Haddow 176 (DAO). Thunderbay Distr.: 24 mi 3 of Nipi- gon along rt. 17, Furlow 312 (MSC); N end of Grassy Lake, 2 mi S of 336 Cooke Point, Lake Nipigon, Garton L785 (DAO). Prince Edward Island. Prince Co.: woods just E of Wellington, Ml—fi (DAO). Queens Co.: Mount Albion, Erskine éflfl (DAO); across road from Green Gables, P.E.I. National Park, M16 (DAO). Quebec. Abitibi Co.: Harricanaw River, Bentley M (CAN). Anticosti Co.: Ile d'Anti- costi, Port Menier, Bernier _e_til_. _9_40_642_080 (DAO). Bonaventure Co.: Bonaventure Island, Gaspe, MEL“: June 12, 1961 (CAN). Chambly Co.: Longueuil, Rolland-Germain _2_94_29_ (NY). Chateauguay Co.: Saint- Chrysostome, Cleonique 11515 (DAO). Chicoutimi Co.: Riviere Ste- Marguerite, Coyouette g Brisson M (DAO). Compton Co.: Scotts— town Bog, 3 mi E of village, Calder £195 (DAO). Gaspe-Est Co.: low, wet margins of Seal Cove River, Douglastown, Collins 313g. &, Aug. 22, 1904 (CH); York, Collins e_tfl. 3.1;, Aug. 25, 1904 (*GH); gravelly banks of River Ste. Anne des Monts., Fernald 13‘. Collins _2_u (A, GH, MSC, NY). Iberville Co.: Henryville, _lglr_ie-Victorin Ea—L Ml (CAN, NY, UC). Isles de la Madeleine Co.: Beign Island, swale, fl. ml—Sfl (CH). Lac St. Jean Co.: Peribonka region NE of Lake St. John, around Lac Alex, Hustich £8 (CAN). L'Assomption Co.: L'Epiphanie et l'Assomption, w-MEEEEa—l. fl (CAN). Laval Co.: SaintvLeonard, Ricard é Boivin fl (UC). L’Islet Co.: St. Aubert, Jackson fl (DAO). Mantane Co.: Riviere Blanche, wet ground, Forbes 69_58§ (CAN, DS). Megantic Co.: Black Lake, flflil- Egg (CAN, IDAO). Mistassini Co.: Pointe Rouleau (Ile Lemoine), Rousseau _§c_ Rouleau 291 (CAN, NY). Montmorency Co.: 1% mi au N de Stoneham, Parc des Laurentides, Lemieux 94.2822 (DAO). Nouveau Quebec Region: Fort ‘George, Baldwin e_t-fl. 565 (CAN); Rupert House, Dutilly _§c_ Lepage 13595 337 (A); Great Whale River, near Sandy Point, Baldwin £1 31, £98 (WTU). Papineau Co.: a short distance N of Long Lake, about 5 mi NE of Buckingham, Jenkins 4 Parmelee £41; (DAO). Pontiac Co.: near end of Black Bay, EEEE.§.§£EEE.§EQ (DAO). Portneuf Co.: Station Agionomique, EEESIEE£§.§&£2§ (DAO). Richmond Co.: Lac Brompton, Forest 19922.(CAN)° Rimouski Co.: St. Fabien, Scoggan‘416 (CAN). Riviere du Loup Co.: Saint—Paul-de—la-Croix, Blouin 31 E1. 1114 (DAO). Saguenay Co.: Natashquan, 1£w1§_§1§§3 (CAN). Sherbrooke Co.: Rock Forest, Lamoureux 324 (CAN). Stanstead Co.: North Hatley, EEEE.§;E;! Sept. 26, 1914 (*A). Temiscouata Co.: 47025., 68040‘, Blouin £E_213 Z193 (DAO). Vandreuil Co.: Rigaud, 321.3519 (DAO). Wolfe Co.: Lac Nicolet, glilgggfleiwfi (CAN). UNITED STATES. Connecticut. Litchfield Co.: Berkshire Mountains, Barkley £1_§1. gfilgg (MEXU). Woodbury Co.: Woodbury, 912£k_§121, Aug. 25, 1908 (DAO). Illinois. Winnebago Co.: shallow bog near Sugar River, Sec. 20, Shirland Twp., £311_§.Fg11 49-382 (DAO). Indiana. Elkhart Co.: 9 about 1 mi E of Bristol, 2322.3922fi (DS, 1ND). Lagrange Co.: Pigeon River State Fish and Game Area, on the N shore of the marsh along the Pigeon River, Furlow 243 (MSC); along streams, 3 mi SE of Mongo, Palmer 49315 (NY). Lake Co.: swampy woods, Miller, EEEEE.§21L: Aug. 2, 1898 (MSC); valley of Deep River, below schoolhouse, N of Liverpool, about 2% mi NW of Hobart, Steyermark 61521 (F). Porter 1Co.: border of a tamarack swamp about 1 mi N of Mineral Springs, 12g32.9§11 (IND, NY); just E of Tremont, in the ditch beside the (road, Furlow £45 (MSC). Iowa. Allamakee Co.: margin of Yellow River near Old Stone House, 7 mi NE of Postville, Thorne 12449 338 (F, NY). Chickasaw Co.: 1 mi N of New Hampton, Thorne l1§2§ (DAO). Fayette Co.: about 8 mi SW of Oelwein, Pammel 312 (MO). Winneshiek Co.: wooded bluffs at Bluffton, Shimek §;2;, Sept. 16, 1903 (M0); near Freeport, Tolstead ELEL’ Aug. 19, 1933 (UC). Maine. Aroostook Co.: Presque Isle, Chamberlain 2222 (UC). Cumberland Co.: wet woods and thickets, Baldwin, Fernald é.£2fl§.l§fll§ (A). Hancock Co.: 3 mi E Franklin, Friesner éégg (RM, UC). Knox Co.: Camp Medomak near Washington, Steyermark £99; (F). Lincoln Co.: Ocean Point, Fassett l§§gg (*F). Oxford Co.: shore of Moose Pond, 7 mi N of Brigeton, Lambert 32329 (MEXU). Penobscot Co.: boggy woods and thickets, Milford, Fernald é.£23§.l§éli (A); swamps, Bangor, Knight §;3;J Apr. 12, Oct. 8, 1905 (*RM). Sommerset Co.: River intervals, Dead River, Fernald 2 Strong 3&2 (GH, MSC). Waldo Co.: clay bank bordering Penobscot Bay, Fassett zgfiél (UC). York Co.: York and vicinity, York Harbor, Bicknell 14:5 (NY); Cape Heddick, ME. (A, UC, US). Maryland. Mountain Lake Park, Carter s.n., Sept. 2, i190? (*NY). Massachusetts. Berkshire Co.: Hoosac R., Churchill s.n., July 26, 1915 (*MO); E Mt. Williamstown, Gilbert s.n., in 1883 (*UC); Mount Washington, Moldenke §_Moldenke lléél (NY). Dukes Co.: Martha's Vineyard, head of cove, Great Pond, Bicknell 23$; (*NY). Essex Co.: Sphagnum bog, Lymsfield, Fernald g Bartlett Z§§_(NY, US). .Hampshire Co.: thicket former in swamp, Amherst, Torrey 3:3;J June ’21, 1952 (WTU). Hampten Co.: Granville, Seymour §l_(NY). Middlesex Co.: swampy woods, Winchester, Fernald §_Bartlett 1_(*GH, NY); clay- bottomed swamp, Winchester, Fernald §_Bartlett lg_(*GH, NY); Round Pond, Tewksbury, Fernald é Bartlett l§_(CAN, NY); Round Pond, 339 Tewksbury, Fernald é Bartlett lg (CAN, F, *GH, US). Nantucket Co.: Nantucket Island, Copaum Pond, Bicknell §££§ (NY). Norfolk Co.: North Aqueduct, Wellesley, Wiegand.§;fl;v Apr. 10, 1909 (F). Michi- gan. Allegan Co.: without location, Allen s;n;, Aug. 11, 1938 (MSC). Alpena Co.: on the E shore of Long Lake, Furlow £33 (MSC). Baraga Co.: S shore of Sand Bay, ca. 2 mi ENE of Pequaming, Beaman lfili (MSC). Cass Co.: without location, Pepoon 929 (*MSC). Charlevoix Co.: SW end of Font Lake, Beaver Island, 3333 llgg (MSC). Cheboygan Co.: 11 airline mi SW of Cheboygan, Shinners léégé (DAO). Chippewa Co.: Whitefish Point, crest of foredune, Gillis 2332 (MSC). Clare Co.: S shore of Eight Point Lake, Furlow 22; (MSC). Clinton Co.: T8N, R3W, 810, 22312 swamp in valley of Maple River, Parmelee zéfll (MSC). Crawford Co.: T27N, R3W, 815, Parmelee 3222 (MSC). Dickson Co.: Norway, Wheeler s.n., Aug. 21, 1892 (MSC). Gogebic Co.: without location, Wilson l2; (DS). Gratiot Co.: Alma, 23X£§.§:E;9 Apr. 7, 1892 (DAO). Houghton Co.: Lane Linden, Farwell fl (DAO). Iosco Co.: 1 mi N of Whittemore, Hoffman § Poltorak gig (MSC). Kent Co.: without location, Fallass §;E;, June 15, 1895 (WTU). Keweenaw Co.: Isle Royale National Park, Mott Island, Bailey é Bailey 5Q§Q_(MSC); Copper Harbor, Hermann Zfigg (NY). Livingston Co.: low border of Douglas Lake near mouth of Maple River, EEEE.E§§lfl.(*NY)' Luce Co.: Deer Park, shore of Muskallonge Lake, Gillis §§§§_(MSC). Mackinac Co.: dune area between Brevort Lake and Lake Michigan, Ehlers fig§§ (DAO). Marquette Co.: thickets bordering a beaver meadow, Turin, Barlow §;EL’ Aug. 8, 1901 (MSC, NY). Mason Co.: Hamlin Lake, Ludington, lake border, Chaney 163 (NY). Mecosta Co.: Big Rapids 340 ‘1 1 1 EEEL.§SEL’ Aug. 8, 1901 (MSC, NY). Muskegon Co.: Twin Lake area, 1 T36N, RlE, SZ, Hoffman_§ Poltorak 1992 (MSC). Newaygo Co.: S side ‘ of Loda Lake, Gilly é Parmelee Egg (MSC). Ogemaw Co.: Sec. 19, NW \ quarter, Hoffman é Poltorak 229 (MSC). Ontonagon Co.: 1.5 mi E of V Silver City, edge of dunes on Lake Superior shore, Beaman 1§£2 (MSC). Roscommon Co.: TZ3N, RAW, Sl9, Parmelee é Hyypio 2§_(MSC). Saginaw ‘ Co.: Birch Run, Bartlett é Richards l2 (DAO). St. Clair Co.: near Port Huron, 22§§E.§;2;$ July 21, 1896 (*RM). St. Joseph Co.: TSS, RllW, SZZ, Parmelee EE.3lf 32_(MSC). Van Buren Co.: Pilgrim Haven, TZS, Rl7W, S32, Gillis 6328 (MSC). Wayne Co.: Nankin Lk., Chandler 342;, Nov. 5, 1916 (MSC). Minnesota. Aitkin Co.: 5 mi E of Jacob- sen, Furlow 252 (MSC). Anoka Co.: 3/4 mi due N from Little Cook Lake, edge of Carlos Avery Game Refuge, S of Linwood, Thieret §é§z ‘ (*F). Benton Co.: Mayhew Creek Valley NE of East St. Cloud, M22£g_§ §g££_1§§&l (DAO). Chisago Co.: Center City, Taylor ELEL’ July, 1892 (NY, RM). Clearwater Co.: Lake Itasca State Park, Furlow 321 *(*MSC). Cook Co.: Grand Portage, Rydberg 2§§g (NY). Hubbard Co.: Benedict, Bergman 2932 (*NY). Louis Co.: 15 mi NW of Duluth along iMinn. rt. 194, growing in the ditch beside the road, Furlow 221 (MSC); ,border of bog, bay side, Duluth, Lakela 1331 (*NY). Marrison Co.: 12 mi W of Brainerd along US rt. 210, Furlow Z§9_(MSC). Stearns Co.: St. Cloud, Campbell 3:2;9 July, 1896 (RM). New Hampshire. Cheshire Co.: Richmond, Batchelder s.n., Aug. 21, 1919 (NY). Coos Co.: shore of Lake Umbagog, Cambridge, Pease 18150 (*GH). Grafton Co.: along Batchelders Brook % mi from Baker River, Churchill s.n., Aug. 6, 1936 (MSC); Dolly Copp Campground, White Mountains Nat'l. Forest, Nelson é 341 Nelson 2&63 (RM, UC). Sullivan Co.: along rocky streams near Acworth, Palmer 3212: (A). New Jersey. Sussex Co.: Cranberry Lake, Mackenzie §;2;J Aug. 19, 1906 (RM). New York. Cayuga Co.: on the W shore of Sterling Pond, Fair Haven Beach State Park, Furlow 121 (*MSC). Chemung Co.: without location, Lugy.9613 (*RM). Chenago Co.: without location, LEEX.§2§§ (*F, RM). Essex Co.: 5 mi NW of Lake Placid along the West Branch River, Furlow 122 (MSC). Franklin Co.: 9 mi NW of Loon Lake along N.Y. rt. 99, Furlow 197 (MSC). ‘— Genesee Co.: Dryden Lake, Dudley ELEL’ Sept. 27, 1880 (DS). Greene ‘ Co.: in moist soil along roadside, Catskill, Moldenke 3362 (NY). 1 Herkimer Co.: summit of Black Bear Mountain, near inlet, Matthews 2222 (UC). Oswego Co.: sand dunes overlying Silurian shales and schists by Lake Ontario, Selkirk, Fernald £E_§l, $2232 (GH). Rock- land Co.: trail between lst and 2nd Letchworth Reservoirs, Bear Mountain-Harriman Section, Palisades Interstate Park, I£h£_1121 (NY). St. Lawrence Co.: 3 mi NW of Canton on the floodplain of the Grasse River, Furlow 12E (MSC). Sullivan Co.: shore of Lake Shandelee, Wilson ELEL’ Aug. 14, 1918 (DS). Tioga Co.: Apalachian, £3323 EZQ (NY). Tompkins Co.: near Ithaca, Barclay 542;, June 1, 1876 (NY). Ulster Co.: Modena, Barnhart 1262 (NY). Warren Co.: 2 mi N of Luzerne, Eggg_12gll (NY). North Dakota. Cavalier Co.: Walhalli, Bergman 292: (*MO). Pembina Co.: Neche, Bolley s.n., July 30, 1891 (NY). Rich- 1and Co.: Leonard, Stevens 1340 (UC). Ohio. Ashland Co.: Lauden— ville, Schaffner s.n., May 30, 1925 (OS). Ashtabula Co.: springs along lower Ashtabula R., Hicks s.n., June 18, 1932 (OS). Attowa Co.: and Point opposite Cedar Point, Gordon s.n., Sept., 1927 (OS). 342 Champaign Co.: Urbana, Demaree 11628 (UC). Columbiana Co.: Wilderness Swamp, 2 mi N of rt. 172, Herrick §;E;9 Aug. 24, 1957 (OS). Cuyahoga Co.: Dover Bay, §E23£.§;EL’ Aug. 17, 1902 (OS). Genga Co.: Genga Lake, Jennings §;2;J Aug. 22, 1903 ()3). Henry Co.: along the Maumee River about 3 mi NE of Napoleon, Weishaupt §;E;’ July 1, 1956 (OS). Lake Co.: swamp by roadside, Madison, Crofoot 2:2;a Sept. 15, 1936 (OS). Lucas Co.: NW of Whitehouse, Moseley ELEL’ Oct. 19, 1919 (M0). Mahoning Co.: peat bog E of Naylor, Weishaupt s.n., June 16, 1956 (OS). Portage Co.: near Streetsborough, Franks 33.213 ELE;’ Aug. 13, 1927 (OS). Richland Co.: Mansfield, Wilkinson 2913 (NY, OS). Stark Co.: near Louisville, Grisez §;E;’ Sept. 14, 1941 (OS). Trumbull Co.: without location, Shanks ELEL’ 1936 (08). Wayne Co.: Wooster, §ffl£1.§.22!§l.fl§2 (OS). Pennsylvania. Bedford Co.: 3/16 mi 8 of Woodvale, Berkheimer 19521 (UC). Fayette Co.: Big Meadow Run, Farmington, EEES.§;EL’ July 9, 1931 (NY). Lancaster Co.: in vicinity 'of mouth of Tucquan Creek, Heller §;E;7 Sept. 1, 1889 (US). Lehigh .Co.: meadows S of trolly tracks just SW of Trexlertown, 23353.2211 (UC). Monroe Co.: % mi E of Tannersville, E3313_11§§ (*DAO). Somer— set Co.: Mt. Davis Bog, Jennings ELEL! July 15, 1949 (DAO). Rhode Island. Providence Co.: bank E of gate, Hauterive, East Providence, Collins s.n., Oct. 17, 1906 (NY); North Providence, Olney s.n., ‘without date (*NY). Vermont. Benningham Co.: Manchester, 2§y_163 (CH). Caledonia Co.: Peacham, very common near stream, Blanchard §;2;, May 1, 1884-(*UC). Chittenden Co.: shore of Lake Champlain, lnear Buckington, Rehder ELEL’ Aug. 4, 1902 (A). Rutland Co.: Twin Mountains, W. Rutland, Eggleston 3210 (NY). Windham Co.: Jamaica, 343 Moldenke §_Moldenke 2614 (NY). Windsor Co.: 1 mi W of Bridgewater Corners along the Ottauguechee River, Furlow 191_(MSC). West Virginia. Preston Co.: Cranesville, 22Xl§.§.2§!i§.192.(NY)' Randolph Co.: Cheat Bridge, 22y1§_§_2gy1§_111§ (WTU). Tucker Co.: George Thompson Farm, Canaan Valley, Core §_Strausbaugh s.n., July 25, 1947 (DAO). Wisconsin. Barron Co.: 6.7 mi SE of Chetek (in SE corner of county), Bartlett é Grayton Z§_(NY). Brown Co.: Wise, Schuette s.n., May and later, 1880 (NY). Dunn Co.: 1 mi SW of Elk Mound, along Sunny Brook, nggg_§5_(DAO). Florence Co.: Tifler, Haynie 28151 (F). Jackson ‘7 Co.: 1 mi E of Black River Falls, Furlow 156 (MSC). La Crosse Co.: La Crosse, Paunand §_Kggh_§121, Aug. 7, 1927 (RM). Lincoln Co.: Merrill, Pammel é.£2§h.§l&.(RM)’ Marinette Co.: marsh near Green Bay at City of Marinett, Grassl 1111 (*NY)- Marquette Co.: in sandy bottom of Chaffee Creek NE of Westfield and near the Waushara County line, Bergseng s.n., Sept. 4, 1956 (DAO). Monroe Co.: swamp near Mill Bluff, w of Camp Douglas, Fassett 20589 (DAO). Sauk Co.: »swampy ground near Kilburn, Palmer 27679 (MO, UC). Shawano Co.: near Keshena, Palmer 27742 (DS). Vilas Co.: Manitowish River, 5 mi W of Boulder Junction, Schneider 1473 (F). This taxon was generally considered conspecific with A1233 incana of Europe until Fernald (1945b) argued in favor of its being given :specific status and treated it so in the 8th edition of Gray‘s Manual lgE'Botany (1950). In the former work, the author lists differences between the taxa of America and Eurasia, especially emphasizing the {fact that the Old World taxon is usually a tree while that of the New World is a shrub. He concludes by stating: ”surely no argument 344 beyond the mere facts and the plates is needed to show that we have been far astray in calling our northern Swamp Alder the same as the Eurasian A, incana.“ The differences cited by Fernald, however, all seem minor when compared with the striking similarities of the two taxa. The original material of Betula ElEEE rugosa DuRoi was a culti- vated shrub in the botanic garden of Harbke near Brunswick. The existence of the type is not known. Fernald (1945b) provides a photograph taken by Rehder at B (and presumed since destroyed) of a specimen of "Betula rugosa” distributed by Ehrhart and coming from the Harbke Gardenl. A1233 incana subsp. rugosa is often indistinguishable from the typical subspecies on the basis of herbarium material. In northeastern North America the foliage is often moderately to heavily glaucous below (A, rugosa var. americana Fern., A, incana var. glauca Ait. f.), this trait gradually disappearing to the west and the south. As noted by Steele (1961), this character is consistent throughout the geographical range of the variety, but it is difficult to use since it doesn's appear until after the leaves are fully matured late in the summer. A cut-leafed form of this subspecies, A1231 rugosa f. tomophylla, as described by Fernald (1914) from Newfoundland (Plate 34). Unlike he cut—leafed form of A, EEEEEJ this taxon is represented by only a ingle collection. Cut-leafed variants of Alnus incana in Europe are 1Termed a ”topotype" by Fernald. 345 numerous and have a very involved nomenclature (cf. Hylander, 1957). Alnus incana subsp. rugosa is easily distinguished from the other American species with which it occurs by its smooth, dark, conspicuously-lenticeled bark and by its ovate, acute, double- serrate leaves. It is predominantly a lowland shrub, more common in swamps and bogs than along running streams like most of the other species of Alnus. The largest specimen of this taxon on record grows at Holland, Michigan with a height of 17 meters, a trunk circumference of 80 centimeters, and a spread of 8 meters (Pomeroy and Dixon, 1966). l The exact western limit of the geographical range is difficult i to establish since this subspecies intergrades gradually into sub- species tenuifolia in northern Saskatchewan and Manitoba. To the south and east its range overlaps that of A132: serrulata, and in this ‘ region occurs a putative hybrid swarm of intermediate forms. Such specimens seem to be especially common in material from Massachusetts. In both areas of overlap with other taxa, A1223 incana subsp. rugosa may be difficult to determine. No attempt has been made to resolve this problem in the present key. 6b. Alnus incana (L.) Moench subSp. tenuifolia (Nutt.) Breitung A1335 incana (L.) Moench subsp. tenuifolia (Nutt.) Breitung, Amer. Midl. Natr. 58: 25. 1957; A1233 tenuifolia Nuttall, North Amer. Sylva l: 48. 1842. Type: Nuttall 313;, "on the borders of small streams within the range of the Rocky Mountains, and afterward in the valleys of the Blue Mountains of Oregon (P?, holotype; GH!, isotype). 346 A1223 incana var. virescens Watson, Bot. Calif. 2: 81. 1880, 323 Wahlenberg, F1. Lapponica, p. 250. 1812; A1323 glutinosa var. virescens Kuntze, Rev. Gen. Pl. 2: 638. 1891; A1325 tenuifolia var. a. virescens Callier in Schneider, Ill. Handb. Laubh. l: 133. 1904. Type: Watson_111 (not seen). A1223 communis Desfont. ex Kuntze, Rev. Gen. Pl. 2: 638. 1891, 232.§XE' A1235 occidentalis Dippel, Handb. Laubh. 2: 158. 1892; A1EE§ tenuifolia var. b. occidentalis Callier in Schneider, Ill. Handb. Laubh. 1: 133. 19043.élEE§ incana subspecies rugosa variety occiden- 13113 Hitchcock in Hitchcock et al., Vasc. Plts. Pac. Northwest 2: 73. 1964. Type: Diecks, Purpus, ”in Nordwest—Amerika heimische" (not seen). A1333 densiflora Muller, Madrofib 5, 152. 1940. Type: A1122 21A, Nevada, Storey Co.: southwest of Virginia City on the Jumbo Can- yon Road, September 3, 1937 (US!, holotype). Spreading shrubs or trees up to 9 (-12) m in height; trunks us- ually several, ascending, up to 30 cm in diameter; bark light gray to dark brown, smooth, speckled with moderately-prominent to in- conspicuous round to elliptic lenticels; young stems often moderately to heavily glaucous; lenticels of twigs 0.2—0.7 mm long, usually in— conspicuous; leaf scars 1—2 mm high, 1.8-2.5 mm wide, with moder— ately-prominent bundle scars. Buds ellipsoid to obovoid, usually somewhat rounded at the apex, lightly to moderately resin-coated; stalk 1-3 mm long, 1-2 mm in diameter, glabrous to velutinous; body 4-7 mm long, 1.5-3 mm in diameter. Leaves ovate to elliptic; apex 347 acuminate, acute, or obtuse; base usually broadly cuneate to rounded; blade (30) 4—8.5 (—9.5) cm long, 2.5-8 cm wide, usually medium to dark green and dull to moderately lustrous above, light to medium green or brownish and dull below, membranaceous to chartaceous; margin flat, not thickened, double-serrate; the major teeth often more or less rounded, (5-) 7-10 (-16) mm apart at mid-leaf, (2-) 4-7 mm deep, regular; secondary teeth (3-) 4—9 per cm, 0.5-1.2 mm deep, regu- lar; abaxial surface glabrous to sparsely pubescent; adaxial sur- face and veinlets glabrous to sparsely pubescent, slightly or not at all resin—coated. Lateral veins 8—13, (3-) 5-7 (-9) mm apart at mid leaf, usually branching once again, especially near the base; cross veins between lateral veins poorly to well developed. Petioles (4-) 7-18 (-25) mm long, O.8~l.5 mm in diameter, glabrous to moder- ,ately villous or velutinous, moderately to densely glandular. Stip— ules elliptic to obovate, the apex rounded, 5-7 mm long, 2—3 mm wide, green to light brown, glabrous to moderately villous. Pistillate inflorescences borne in groups of (2_) 3-5, at anthesis 2.5—4 mm long, 1.5-2.8 mm in diameter, on peduncles 0.2—1 mm long, 0.8—2 mm in diameter; staminate catkins borne in groups of 3-5 at the end of the main branch above the pistillate inflorescences, this branch mod— erately to strongly divergent from the main axis, at anthesis 4-10 cm long, 7-9 mm in diameter, on peduncles 2—18 mm long, 1-1.5 mm in diameter. Staminate flowers with 4 perianth parts, these obovate, 1.6-2 mm long, 1.2_1.6 mm wide, the margin lined with very minute glands; stamens almost completely adnate to the perianth parts, sually appearing shorter than the perianth, the filaments 0.4-1.1 348 mm long, the anthers 0.8-1.1 mm long and 0.8—1.1 mm in diameter, the thecae separate for 45-55% of their length. Infructescences 9-18 (-20) mm long, (5-) 8-13 mm in diameter, on peduncles 0.2-3 (~6) mm long, 0.8-1.5 mm in diameter; scales 4-5.2 mm long, 3.7-5 mm wide at the apex, (0.8-) 1.2-1.5 mm wide at the base, the apex mod- erately thickened, the terminal lobe-tip acute to rounded and some- what extended. Fruits narrowly winged or merely wing-margined, brown; body mostly elliptic or obovate, 2.5—3.5 mm long, 1.2-2.5 mm in diameter; wings 3-3.5 mm long, 0.5-1.2 mm wide, firm to coriaceous; persistent styles 0.5—1.2 mm wide, firm to coriaceous; persistent styles 0.5—1.2 mm long. Plates 9B, 12D, and 35. Distribution and Ecology; Southern Alaska east to Mackenzie District (as far north as the Mackenzie Delta), south to south-central California, southern Arizona, and central New Mexico. Streambanks, lake shores, wet fields and meadows, bog and muskeg margins, and moist slopes at elevations from about 100 meters in Alaska to over 3000 Occurring singly or forming fairly meters in Colorado and Arizona. dense thickets. Often associated with Populus, Salix, Abies, Pinus, or Pseudotsuga. Figure 29. Common Names: Thinleaf (or thin-leafed) alder, mountain alder, alder, river alder. Representative Specimens: CANADA. Alberta. Lac la Biche, Cody i Gutteridge 6909 (DAO); by Athabasca R., Ft. McMurray, Cody and utteridge 7224 (DAO); Cold Lake, French Bay E of peninsula along the lberta—Saskatchewan border, Dumais §_Rankin 1233 (*CAN); about 20 mi Plate 35. 349 Lower left: isotype of Alnus tenuifolia Nutt. (= AHNS incana subsp. tenuifolia Nutt.) Breitung). Upper riflfi: specimen of Alnus incana subsp. tenuifolia (Nutt-) Breitung. 350 W4. WWI ~. . W, WMWI W)“ }W “£77., $7554.... 2» 74.an 4L__1,lr/V/ M ,mffia— *bflfi" 11.04% M ...l_ ('ou.. Nlrruu. _ Mal Ivy Elba Dunn-I. Infill. Plate 35 . 351 Figure 29. Distribution of Alnus incana subsp. tenuifolia (NuttJ Breitung in North America. I70 "£6? I"3: ‘ 3M? ‘;60 "E!?'PO “9'! We. l" . .1“ m ‘ {02/ . .- o' " - _1 Va ./ - ul' . 4' ' x o \ Figure 29. . at.” A . u. . —..-=- .—~' 353 SE of Smith, FurlOW'111 (*MSC); Jasper National Park, about 7 mi SW of the N park entrance, Furlow 119 (*MSC); Saskatoon Mountain near Beaverlodge, §£gfl_191 (DAO); Edmonton, Macoun A Herriot 3121, Aug. 29, 1906 (CAN, NY); Red Deer, fléWatson m (CAN, RM, UC, WTU); base of E slope of Caribou Mountains, 1332.1121 (CAN, NY); river valley adjacent to Fort Saskatchewan, Turner 3113 (*DAO). British Columbia. Dawson Creek, Breitung 1111 (*DAO); 4 mi NNE of Nelson, Calder A Savile 9395 (*DAO, UC); Stikine River, near mouth of Clearwater R., Cooper §_Andrews 397 (F); Prince George, ‘mtig-cq _ Eastham 11640 (CAN); ca. 20 mi E of Vanderhoof, Hitchcock A Muhlick 22109 (DS, NY, UC, WTU); Merritt, bank above Nicola River, McCabe 4438 (*UC); 25 mi S of Ft. St. James, McCabe 7589 (UC, WTU); Queen Charlotte Isl., Newcombe s.n., Apr. 25, 1901 (*F); Liard Hotsprings, Porsild & Porsild 22028 (CAN); N slopes of Peace R. valley, vicinity of Hudson Hope, Raup A Abbe 3699 (*CAN, NY); along the Moyie River 6 mi N of Kingsgate, Wg§g£_111§ (CAN, NY, UC, WTU). Northwest Terri- tories. Mackenzie District: Yellowknife, N end of Kam Lake, 9291 2§fl1 (DAO); Snowdrift, at E end of Great Slave Lake, 9291.12122 (DAO); near Basworth Creek, Norman Wells, 9291.9 Gutteridge Z111 (*DAO, F); Fort Simpson, EEQX.§.ME£ES.§99§ (DAO); Great Slave Lake, Hardisty Island, 9291 Q McCanse 1298 (DAO); Mackenzie River 4 mi E of Trout ‘ River, Eggy é Spicer.111z1 (DAO, rUC); Liard River, 2 mi above Blue 1 Bill Creek, 42 mi N of Fort Liard, mg Spicer 11932 (*DAO, UC); ‘ top of Peel River bank, Fort McPherson, EESE§.§;EL9 July 29, 1957 1 (DAO); flat alluvial island in delta at junction of Mackenzie E Channel with Kugwallet Bay of Arctic Ocean, Lindsey 706 (CAN). Saskatchewan. 354 South shore of Lake Athabasca, E of William River, vicinity of Little Gull Lake, shore of small lake to the E, éEEEE 319-62 (DAO); Meadow Lake Forest Reserve, 20 mi S of Meadow Lake, Breitung 1191 (DAO); along the Beaver River, near mouth, Churchill River system, Ile La Crosse, Breitung w (DAO); Lazy Edward Bay, Cree Lake, HE (DAO, RM); bank of William R., vicinity of William Pt., Lake Athabasca, 5332 1111 (CAN, NY). Yukon Territory. Whitehorse, Anderson 9119 (CAN); Mayo, Anderson 9111 (CAN, MSC); Bear Creek area about 8 mi E of Dawson, Calder 1 Billard 1119 (DAO, RM, UC); Ten Mile Creek at mi 813, Alaska Highway on Teslin Lake, Calder 1 Gillett 11111 (DAO); Dawson, Eastwood 11 (*CAN); Dawson, Eastwood 19 (*CAN); Fort Sel- kirk, Gorman 1991 (CAN, NY); Rampart House on the Alaska-Yukon border, 19§2_119 (*DAO); St. Elias Mountains, on E side of Alsek River, £315- £23 11:11 (CAN). UNITED STATES. Alaska. Mendenhall, Anderson 131 (NY); Fairbanks, Anderson_1119 (NY); Moose Pass, Anderson 1991 (CAN, RM); Hope, Anderson 1111 (CAN, RM); floodplain of the Chena River at mi 3.2 on Chena Pump Road, Anderson 1121, Sept. 6, 1971 (MSC); Kenai River near mouth of Cooper Creek, Kenai Peninsula, Calder 1111 (DAO); foot of Chugash Mts., Anchorage, Dutilly EEHA1._11119 (*DAO, US); in meadows at Moose Pass, Nelson_§ Nelson 1911 (RM); bank of Naknek River, King Salmon, Schofield 1191 (DAO, WTU). Arizona. Apache Co.: Greer, Fulton 1111 (*ARIZ); 3 mi N of Alpine, 123123 229 (MSC); bottom of Canyon de Chelly, % mi above Monument Canyon, Goodman 1 Payson 1111 (*WTU). Graham Co.: Riggs Flat, Graham Mts., Shreve 5251 (*ARIZ). Navajo Co.: head of Tsailee Canyon, 999g11§§ 9 (ARIZ). Pima Co.: Rincdh Mts., Manning Trail, Blumer 3420 (ARIZ, UC); Mt. Lemmon, Knoblock 1111 (MSC). California. Alpine Co.: Car- son Spur, Hansen 199 (UC). Amador Co.: Sierra Nevada, Hansen 1_(UC). El Dorado Co.: ca. 2 mi S of Meyers, Upper Truckee River Basin, Crampton 1199 (*UC); about 1 mi E of Pyramid Ranger Station, Robbins 1111 (*UC). Fresno Co.: Paradise Valley, Clemens 1111, July 18, 1910 (NY, UC). Glenn Co.: Plaskett Meadows, 133% (*UC). Humboldt Co.: Croghan Hole on Trinity Summit, IEEEX.12£§R (UC, WTU). Lassen Co.: Mt. Lassen Volcanic National Park, 1 mi below Drakesbad, 1111 s.n., Aug. 24-26, 1925 (RM). Mendocino Co.: Spruce Grove, Hemphill 11:1 (UC). Mono Co.: margin of Lake George, 11111-1111 (UC). Nevada Co.: Sierra Nevada at Donner Pass, Greene 1111, Oct. 14, 1874 (JEPS). Placer Co.: 4 mi S of Truckee on Truckee—Tahoe Road, 11121_1111 (*UC). Plumas Co.: Spanish Creek, off hwy. 24 just N of Quincy, fifl (WTU). Shasta Co.: 1 mi N ofMt. Shasta City, 1111' 1991 (UC). Siskiyou Co.: shore of Castle Lake, Bacigalupi 1199 (JEPS). Trinity Co.: 1 mi S of the Race Track, South Fork Mountain, 325331.12919 (UC). Tuolumne Co.: Kennedy’s Meadow, E fork of middle fork Stanislaus River, 91111 111 (JEPS). Colorado. Boulder Co.: Boulder, Moseley 1111, in 1896 (*RM); lower Boulder CaHon, Osterhout Zflgfl (*NY, RM); Allan's Park, Ramaley 991 (*RM); along streams at mouth of Gregory Canyon W of Boulder, Robbins 111 (*UC). Clear 3prings Co.: Idaho Springs, Rydberg 1111, Aug. 26, 1895 (NY). )enver Co.: Rocky Mountains, Denver, Kuntze 1111, Sept., 1874 (NY). iagle Co.: near Walcott, Palmer 11111 (*NY). Fremont Co.: Lake -reek and Grape Creek, Brandegee 919_(UC). Gunnison Co.: Crested utte, 8 mi NE of Gothic, Booth 49c334 (WTU). Huerfano Co.: Cuchara 356 Valley, near La Veta, Rydberg 1_Vreeland 1111 (NY). La Plata Co.: between Mancos and Durango, 11211.199 (NY). Larimer Co.: 2 mi below Bear Lake, Rocky Mountain National Park, Furlow 191 (MSC). Mineral Co.: Rio Grande National Forest, Wagon Wheel Gap, Murdoch 1191 (UC). Ouray Co.: Red Mountain road S of Ouray, Underwood 1 Eglhll119 (*NY). Park Co.: 2.5 mi N of Grant on Guinella Pass Road, Fambrough 119_(US). Saguache Co.: by stream along hwy. 114 ca. 10 mi 5 of hwy 50, 13111111Wi11ey fl (DAO, NY). Summit Co.: near Breckenridge, Mackenzie 199 (NY, RM). Idaho. Ada Co.: along the river, Boise, 91A£1_111 (NY, RM, UC). Adams Co.: along Big Creek, 13 mi E of Cascade, Payette National Forest, 1111§_1111 (WTU). Bonner Co.: Priest Lake, Nelson 1_Nelson 1911 (NY, UC). Bonneville Co.: 17 mi SW of Victor, Targhee National Forest, Furlow 277 (MSC). Butte Co.: mouth of the W fork of Little Cottonwood, Craters of the Moon National Monument, 1111£_11111 (ARIZ). Custer Co.: borders of Redfish Lake, Sawtooth Range, Thompson 11111 (WTU). Elmore Co.: Saw- tooth Primitive Area, along stream 7 mi S of Spangle Lakes, 111117 5211.1 Muhlick 19111 (NY, WTU). Fremont Co.: Big Falls, on river )ank, Christ 1111 (NY). Idaho Co.: 6 mi SW of Lolo Pass, Clearwater lational Forest, Furlow 111 (MSC); about 1 mi above Selway Falls, Eflhgy_1119 (ARIZ, DAO, NY, WTU). Latah Co.: Rowlands Park, NE of bscow, Christ 1.1151'1191 (NY). Lemhi Co.: 13 mi S of Lost Trail 'ass, Salmon National Forest, Furlow 111 (MSC). Lewis Co.: Cot- onwood Canyon, Mulford 1111) June 14, 1892 (NY). Nez Perces Co.: ewiston, Christ 1991 (NY). Owyhee Co.: lower Sawpit Creek, ca. 1 i S of Silver City, Baker 8235 (WTU). Peroe Co.: on creek bank 357 near Zaza in the Craig Mountains, S_t. SEEM (NY, UC). Teton Co.: creek bank, Victor, Merrill EWilcox w (RM). Victor Co.: Moose Creek, Victor, Christ _SEE (NY). Montana. Flathead Co.: along Logan Creek 10 mi NE of Lake McDonald, Glacier National Park, Furlow 213 (MSC); Columbia Falls, Williams E (*MSC, NY). Gallatin Co.: Bozeman, BlankenshiE s_.n_._, Apr. 17, 1899 (NY, RM, WTU). Glacier Co.: NE of Lower Two-Medicine Lake, Bartlett _6: Grayson fl (NY). Lewis and Clark Co.: Helena, Butler 16 (NY). Mineral Co.: 5 mi SE of state bound- ary between Lookout Pass and Saltese, Bartlett 1 Grayson l_0_3_§_ (NY). Missoula Co.: Missoula, M1 (*NY); 6 mi NE of Arlee, Furlow 172 (MSC). Park Co.: Livingston, along Yellowstone River, ME (UC). Silverbow Co.: Melrose, Rydberg 16_11 (NY). Nevada. Douglas Co.: Edgewood, Lake Tahoe, at California line, WEE (*UC). Elko Co.: streambank, Charleston, Holmgren l_63_4 (NY). Lyon Co.: Carson Creek Cafion, Brandegee ELL, without date (*UC); Carson City, Watson 1022 (US). Ormsby Co.: King‘s Cafion, BEE (MSC, NY, RM, UC). Storey 1 Co.: 2-31/2 mi SW of Virginia City, mgfl (A, NY). Washoe Co.: 8 ’ mi W of Reno Hot Springs, Archer fl (*DAO); Hunter Creek Canyon, Kennedy 21;, Apr. 5, 1901 (*UC). New Mexico. Catron Co.: upper Willow Creek drainage, Mogollon Range, 60 mi NW of Silver City, 18 mi L E of Mogollon, _Bflgjfi (WTU). Colfax Co.: vicinity of Ute Park, Standley £51.11 (NY). Rio Arriba Co.: at the Boy Scout Camp in Gallinas Canyon, about 12 mi SW of Las Vegas, Hinckey $91 (NY). San Juan Co.: 7 mi NE of Washington Pass near E base of Chuska ‘Mts., Watson s_.n_., Aug. 27, 1958 (*ARIZ). San Miguel Co.: low woods along Pecos River, Pecos, Drouet é Richards 3312 (*F). Santa Fe’ Co.: 358 Pecos River National Forest, Standley 5923 (NY). Socorro Co.: Water Cafion, Magdalena Mts., Herrick §_Herrick 12 (F). Taos Co.: 1 mi NE of Tres Ritos, along road to Durea Canyon, Terrill §_B£gwn_2§1§ (US). Oregon. Coos Co.: Coos Bay, Sheldon 5117721 (UC). Crook Co.: Straw- berry Range, Blue Mountains, between Mitchell and Prineville, Mason 2229 (*UC). Grant Co.: banks of John Day River, Prairie City, Ferris g Duthie 199 (RM). Harney Co.: opposite Tumtum Lake, 7.6 mi 3 of Fields, Ferris 11§§1 (UC). Josephine Co.: glade on Sand Ridge trail back of Oregon Caves, Beattie 211g (UC, WTU). Klamath Co.: SW corner, N wall, Redblanket Canyon, Crater Lake National Park, Bakg£_191§_(NY). Morrow Co.: along Willow Creek at Tons, Lunnell §;E;’ July 16, 1903 (RM). Wasco Co.: Dufur, Abrams 2211 (RM). R. & Blue Mts., Nuttall s.n., without date (GH). Utah. Cache Co.: along Logan River, Red Bank Public Camp, Maguire gégg (RM, UC). Salt Lake Co.: Murry, gggg§_§121, Apr. 11, 1917 (*UC); near Salt Lake City, Stokes E1213 May 15, 1900 (NY); Lambs Canyon, Wasatch Mountains, Vickery 111$ ’(ARIZ, UC, WTU). Sevier Co.: head of Salina Canyon, £222§_2&12 (MSC, NY, UC). Summit Co.: 1.2 mi below Soapstone Junction, Unitah Mts., Erdman 329 (NY). Wasatch Co.: along bank of small stream W of Strawberry Res., Foster Zé§_(NY). Washington. Chelan Co.: 'lower end of Lake Chelan, Sudworth £131, Oct. 2, 1904 (US). Kittitas Co.: Ellensburg, E1me£_£1§_(NY). Spokane Co.: base of Mt. Carleton (Mt. Spokane), Kreager 11g (NY, UC, WTU). Stevens Co.: 28 mi E of (Colville on hillside above Little Pend Oreille Lake, Beattie 11§§1_(NY). Yakima Co.: Surveyor's Creek, Toppenish, Heidenreich 1é_(WTU). Wyoming. Albany Co.: Medicine Bow National Forest, W of Eagle Rock, 359 Holliday §£_(RM); Jelm, Nelson §9§§_(NY, RM). Fremont Co.: along shore of Bull Lake, Porter 2212 (RM, WTU). Park Co.: Yellowstone National Park, Soda Butte Creek, Nelson g_Nelson §§§§ (NY, RM, WTU). Platte Co.: along the bank of North Laramie River about 20 mi SW of Wheatland, West 1 (RM). Sweetwater Co.: Black Rock Creek, Teton Forest Reserve, Tweedy 345 (NY). Teton Co.: Two Ocean Lake, marshy margin of the lake, Churchill s.n., July 20, 1958 (MSC). Uinta Co.: banks of Bear River, 1 mi S of Evanston, Goodman 4496 (RM). This taxon differs from subsp. rugosa mainly in the coarser teeth of its leaves, its larger, more tree-like habit, its lighter bark with less—conspicuous lenticels, and its montane riparian habitat. Although variable, it is not as much so as subsp. rugosa. In the northern part of its range, it occurs at nearly sea level, but it is , found primarily in the mountains throughout most of its range. ‘ In 1940, Muller described what he thought was a new species of Alnus, A, densiflora, and a new section, section Pycnantha, from a specimen collected in the Sierra Nevada Mountains of Nevada, noting that the staminate flowers were very ”compact" in the inflorescences. EThe type of this species (Allen 213) is typical A, incana subsp. tenuifolia, the crowded staminate catkins being merely the un- expanded next season's inflorescences, although it is clear from €Muller's description that he thought them to be in anthesis (in Sep- :tember). Hitchcock, in his recent treatment of this taxon (Hitchcock 21.31., 1964), Views all of Alnus incana in North America as belonging to one subspecies and the European material to another, the eastern and 360 western North American segments being treated as varieties. From the amount of divergence among these taxa, however, especially between the eastern and western New World forms, it seems better to make them all of the same rank. The only other taxa with which A1233 incana subsp. tenuifolia could be confused in its normal range are 11235 rhombifolia and A. oblongifolia, with which it occurs sympatrically. In leafless material these species can usually be distinguished from A. incana by the buds, which are more rounded at the apex and more completely covered by the bud scales in the latter. Flowering material can be determined by the number of stamens and by the fact that the stamens are shorter than the perianth only in A. incana subsp. tenuifolia. Several floras (including Gleason, 1963, and Gleason and Cron- quist, 1963) state that the range 0f.élflE§ incana subsp. tenuifolia extends east to Minnesota and North Dakota. No material of this taxon was seen from that area, however, all specimens being referred (to subsp. rugosa. Although no material of A. incana subsp. tenui- folia was seen from Mexico, it might be expected there in the area ,adjacent to its range in Arizona and New Mexico. Gleason (1963) and Gleason and Cronquist (1963) state that this taxon exists in Baja California. 7. Alnus serrulata (Aiton) Willdenow Alnus serrulata (Aiton) Willdenow, Sp. Pl. ed. 4, 4(1): 336. 1805; 3etula serrulata Aiton, Hort. Kew., p. 338. 1789; Betula alnus ser- rulata Michaux, Fl. Bor. Amer. 2: 181. 1803; Alnus serrulata a vulgaris 361 Spach, Ann. Sci. Nat. ser. 2, 15: 206. 1841; A1225 glutinosa 5.§SE£ET 1313 Regel, Mem. Soc. Nat. Mosc. 13(2): 164. 1861, in part; A1nu§ glutinosa 6 serrulata lusus a. genuina Regel, Mem. Soc. Nat. Mosc. 13(2): 164. 1861; fiflfli serrulata a genuina Regel, Bull. Soc. Nat. Mosc. 38(3): 432. 1865; Alnus rugosa var. B serrulata Winkler, Pflan- zenreich 19(4.61): 120. 1904; A1223 incana var. serrulata Boivin, Le Nat. Canad. 94: 651. 1967. Type: "mat. of Pennsylvania. Cult. 1769 by Peter Collinson, Esq.” (BM7, not seen). A1235 carpinifolia Desfont. ex Spach, Ann. Sci. Nat. ser. 2, 15: 206. 1841, 252.313.; A1223 carpinifolia Desfont. ex Hartig, Vollst. Naturgesch. Forstl. Kulturpfl., p. 336. 1851, p£2_§yg. A1233 serrulata 8 macrophylla Spach, Ann. Sci. Nat. ser. 2, 15: 206. 1841; Alnus macrophylla Desfont. ex Spach, Ann. Sci. Nat. ser. ‘ 2, 15: 206. 1841, pro syn.; Alnus macrophylla Desfont. ex Hartig, Vollst. Naturgesch. Forstl. Kulturpfl., p. 336. 1851, 232.211‘ élEE§.EEE£i Tuckerman, Amer. Jour. Sci. ser. 2, 45: 32. 1843, 322 Bongard, Mem. Akad. Sci. St. Petersb. ser. 4, 2: 44. 1833; A1223 £2253 Desfont. ex Spach, Ann. Sci. Nat. ser. 2, 15: 206. 1841, 233 313.; A122§_£EE£E Desfont. ex Hartig, Vollst. Naturgesch. Forstl. Kulturpfl., p. 336. 1851, p£2_§yn. A1233 latifolia Desfont. ex Hartig, Vollst. Naturgesh. Forstl. Kulturpf1., p. 336. 1851, 33313. Alnus glutinosa 6 serrulata lusus b. obtusifolia Regel, Mem. oc. Nat. Mosc. 13(2): 165. 1861; Alnus serrulata 6 obtusifolia egel, Bull. Soc. Nat. Mosc. 38(3): 433. 1865; Alnus rugosa var. obtusifolia Winkler, Pflanzenreich 19 (4.61): 120. 1904. 362 Type: "gesehen vom Ohio und in kultivirten Exemplaren" (not seen). '11235 americana hort. ex Petzold & Kirchner, Arb. Musc., p. 597. 1864, 223 Hartig, Vollst. Naturgesh. Forstl. Kulturpf1., p. 337. 1851. é1ppg noveboracensis Britton, Torreya 4: 124. 1904; Alpp§_§g£- rulata var. vulgaris forma noveboracensis Fernald, Rhodora 47: 358. 1945. Type: Britton 312;, New York, Grant City, Staten Island (NY!, holotype). Alnus undulata hort. ex Winkler, Pflanzenreich 19(4.61): 119. 1904, pro syn. Alnus serrulata var. subelliptica Fernald, Rhodora 47: 358. 1945. Type: Fernald_§ Bartlett 16, ”Massachusetts: sandy swamp, Tewksbury, April 14 and Oct. 14, 1906" (GHl, holotype; CANI, Fl, NY!, US:, W181, isotypes). Alnus serrulata Var. subelliptica forma emarginata Fernald, Rhodora 47: 359. 1945. Type; Bissell é Weatherby_(Weatherby 2031), "Con— necticut: open, rather sphagnous swamp, Rainbow, Windsor, Hartford Co., Sept. 16, 1906 and April 6, 1907" (GH, holotype). Alnus serrulata var. subelliptica forma mollescens Fernald, hodora 47: 359. 1945. Type: S1: John 2681, "New York: wet hollow. Suffolk Co., July 25—Aug. 3, 1920” (GH, holo- iverbed, Southampton. ype)- Alnus serrulata var. subelliptica forma nanella Fernald, Rhodora 7: 360. 1945. Type: Fernald §_Lewis 14596, "Virginia: Ram Hole wamp, Seward Forest, near Triplett, Brunswick Co., June 22 and Sept. 3, 1944” (GHl, holotype). Compact shrubs up to 10 m in height; trunks up to 15 cm in 363 diameter, ascending; bark light gray, smooth to slightly rough, the lenticels inconspicuous; young stems light brown to dark red-brown, dull, often slightly to moderately glaucous, without noticeable resinous coating, not differentiated into long and short shoots, without longitudinal ridges; internodes glabrous to velutinous, moderately to densely glandular; nodes and branchlets bearing in- florescences very densely glandular; hairs whitish to yellowish or brownish; glands small to medium in size, yellowish to dark brown; lenticels of twigs circular to elliptic, 0.3-0.6 mm long, 0.2-0.4 mm wide, yellowish, inconspicuous; leaf scars 0.7-3 mm high, 1.5—2.5 mm wide, the bundle scars moderately prominent. Buds ellipsoid to ob— ovoid, slightly rounded to rounded at the apex, moderately to heavily resin-coated; stalk 1.5-3 mm long, l—l.5 mm in diameter, sparsely to ‘moderately pubescent, densely glandular; body 3-6 mm long, 2-3 mm in diameter; bud scales 2, stipular, equal, valvate, moderately pubes- cent, glandular; pubescence and glands usually obscured by the resin coat. Leaves usually elliptic or obovate (rarely ovate); apex obtuse to rounded (rarely acute); base broadly cuneate (sometimes rounded); blade (4—) 509 (-14.5) cm long, (2-) 3.5-6.5 (-7.5) cm wide, medium to dark green and dull to moderately lustrous above, light to medium green or green-brown and dull below, chartaceous to somewhat coriaceous; argin flat to slightly revolute, slightly to moderately thickened, errulate to somewhat double-serrate; major teeth (when present) 5-) 7-18 (-22) mm apart at mid-leaf, less than 2 mm deep, irregular; econdary teeth (5-) 7—11 (-15) per cm, 0.1-0.8 (-1) mm deep, regular 0 slightly uneven; abaxial surface glabrous to sparsely pubescent, 364 moderately to densely glandular; adaxial surface and veinlets glabrous to moderately villous, moderately to densely glandular, slightly to moderately resin-coated; major veins and vein axils near the base tomentose to wooly-pubescent; pubescence whitish to yellowish; glands small to medium in size, whitish to yellowish or brownish. Lateral veins 8-11, (3-) 4—8 (-10) mm apart at mid-leaf, straight or slightly ascending, sometimes branching once again, especially near the base, terminating in teeth at the margin; cross veins between lateral veins poorly (to rarely well) developed. Petioles (2-) 6-15 (-22) mm long, 0.5-1.5 (-2) mm in diameter, glabrous, moderately villous, or tomentose, sparsely to moderately glandular. Stipules elliptic to obovate, the apex obtuse to rounded, 2.5-5.5 mm long, 1.6—3 mm wide, green to light brown, glabrous to moderately villous, ‘the hairs yellowish, moderately glandular, the glands pale yellow. Pistillate inflorescences borne in racemose clusters of (2-) 3-5 on short non-strongly-divergent branchlets, produced during the pre- vious growing season, erect, ovate to elliptic, at anthesis 3-6 mm long, 1.5-2.5 mm in diameter, on peduncles 0.5—2 (-3) mm long, 0.8-1 mm in diameter; staminate catkins borne in one or more racemose clusters of 3-5 at the end of the main branch above the pistillate inflorescences, this branch usually strongly divergent, bending harply away from the main axis, produced during the previous growing eason, pendend during dormancy and anthesis, at anthesis 3-8.5 cm ong, 4—10 mm in diameter, on peduncles 1.5-8 mm long, 0.5-1.2 mm n diameter; floral bracts 1—2 (m3) mm high, (1.5-) 2-3 (—3.5) mm ide. Staminate flowers 3 per bract; perianth of 4 parts, these 365 elliptic to obovate, the apex obtuse to rounded, 0.7—1.1 mm long, 0.3-1.1 mm wide, the margin lined with minute glands; stamens 4, opposite and basally adnate to the perianth parts, usually ap- pearing much longer than the perianth, the filaments 0.6-0.9 mm long, the anthers 0.8-1.1 mm long and 0.8-1 mm in diameter, the thecae separate for 15-4OZ of their length. Infructescences ovoid to ellipsoid, 10-17 (~22) mm long, (6-) 8-11 mm in diameter; on peduncles 0.2-5 (-8) mm long, 0.8-1.2 mm in diameter; scales 3-4.5 mm long, 3—4 mm wide at the apex, 0.8-1 mm wide at the base, the apex moderately thickened, flat, the terminal lobe-tip acute and somewhat to very extended. Fruits narrowly winged or wing-margined, brown; body obovate, 2.2-3.3 mm long, 1.2-2 mm in diameter; wings 2-2.5 mm long, 0.2-0.5 mm wide, firm to coriaceous; persistent styles 0.7-1.4 mm long. Plates 20, 4D, 8D, l5C, 18E, 19A, 36, and 37. Distribution 22d Ecology: South-central Quebec east to southern Nova Scotia, southwest to northern Ohio and Indiana, central Missouri, and eastern Oklahoma, south to the Gulf of Mexico and northern Florida. Streambanks, edges of sloughs, swampy fields, margins of bogs, and lake shores from near sea level to elevations of about 750 meters in the Appalachian and Ozark highlands. Figure 30. Common Names: Smooth alder, common alder, hazel alder (erroneously), :ag alder, red alder. Representative Specimens: CANADA. Nova Scotia. Yarmouth Co.: iutler's Lake, Gavelton, Fernald 21 1. 21021 (GH); border of Eel lrook, Eel Brook, Jack 3412 (A, UC). Lunenburg Co.: outlet of Wallace 366 Plate 36. Specimen of Alnus serrulata (Ait.) Willd. H010tYPe 0f Alnus serrulata var. subelliptica Fern. 367 W FLORA OF MASSACHUSETTS, as.“ COUNTY .../4 wwmjflw, JanAawEQEfibvaaw¢tha Wen—MM gufififl&mu .QaquMfl%m¢ l “I" Plate 36. 368 serrulata (Ait.) Willd. Plate 37. Representative specimen of Alnus No./.9/ ,7 Ma; 19: y Julia: A. Steyermark, Contact Plate 37. Figure 30. 370 Distribution of Alnus serrulata (Ait.) Willd. 371 Figure 30. 372 Lake near Italy Cross, 2221y_§1§ (DAO). Queens Co.: Ponhook Lake, §21£h_g£_313 19111 (DAO). Quebec. Chambly Co.: Chambly, Cleonigue 2991 (DAO). Lotbiniere Co.: Lotbiniere, rivages du Saint-Laurrent, M35127Victorin 31.31: §§111 (*CAN). Richelieu Co.: Sorel, M35137 Victorin §_Rolland—Germain £1811 (DAO). Riviere du Loup Co.: Sainte-Rita, Blouin 31.31._111§ (DAO). Temiscouata Co.: Lac Naud, Blouin fig. 7_00§ (*DAO). UNITED STATES. Alabama. Baldwin Co.: in a swamp, Perdido, Blanton 1281 (NY, RM); E shore of Mobile Bay at Battles Warf, Brinker 118 (M0). Choctaw Co.: 1.9 mi S of Butler on hwy. 17, 11113‘1§1§2_(WIS). Cullman Co.: Cullman, Eggert 3:3;r June 21, 1897 (M0, NY). Franklin Co.: vicinity of Russellville, 13923_4§_(M0). Lee Co.: Auburn, 23353.8 (NY). Russell Co.: 11 mi W of Phoenix City, Furlow 146 (MSC). Arkansas. Benton Co.: Benton, Demaree 11212 (NY). Garland Co.: near Hot Springs, Chg§g_2§21 (F, NY, RM). Hot Springs Co.: Magnet Cove, Demaree 16620 (*NY). »Independence Co.: Batesville, Demaree 26789 (RM). Izard Co.: Calico Rock, Demaree 23562 (NY, UC). Lincoln Co.: Star City, Demaree ‘14611 (NY). Little River Co.: Ashdown, Palmer 8161 (M0). Nevada ‘00.: swamp margin, Bluff City, Demaree 14182 (NY). Pike Co.: banks ’of Little Mo. River, New Hope, Demaree 2142 (*NY). Pope Co.: Nogo, Merrill 41 (NY). Pulaski Co.: Little Rock, EEEEE.§;EL’ Apr. 20, 1860, Aug., 1860 (*NY); along Broadie Creek, near Little Rock, Merrill ‘Lfléi (*MO). Sebastian Co.: Arkansas National Forest, near Mans- field, Palmer 1211: (NY, MO). Sevier Co.: West Otis, Brinkley 19: (F). Yell Co.: Ola, creek bottoms, Demaree 11111 (M0). Connecticut. New Haven Co.: Waterbury, Cooke St., Lucian Ef.£ (NY); Spindle Hill 373 Road, Waterbury, Lucian E, 142 (NY). New London Co.: Norwich, Setchell 313;, Mar. 19, 1883, Aug. 22, 1883 (UC). Delaware. Sussex Co.: moist thicket near Georgetown, Britton 11_(NY); 4 mi S of Mil- ford, near the W shore of Hudson‘s Pond, Furlow 141 (*MSC). New Castle Co.: region about Newark, Tidestrom 1104 (MICH). District of Columbia. Near Takoma Park, prgg_111_(NY); at edge of natural woods in U.S. Nat. Arboretum, Mazzeo 111g (*WTU); common at Terra Cotta, Tidestrom 4111 (NY). Florida. Clay Co.: Hammock, Peters Creek, Mp DeWinkeler _97_O6 (MICH, MO, NY). Escambia Co.: bottom— land forest along Pine Barren Creek near US 29, Redfearn é.§£21.£112 (GH). Gadsden Co.: along a very small stream 19 mi W of Tallahassee, Godfrey 2112; (CH, NY). Jefferson Co.: 1 mi E of Lloyd, Godfrey 22116 (*NY). Liberty Co.: near Horseford, Palmer 18218 (MO, NY). Wakulla Co.: near the Sopchoppy River, about 5 mi N of Sopchoppy, Godfrey fl (GH, *WTU). Georgia. Ben Hill Co.: 10 mi N of Fitz- gerald and 12 mi 3 of Abbeville, Wilbur p Webster 172_8 (NY). Clark Co.: just S of Athens, Duncan 1811 (RM, *UC). DeKalb Co.: on and about Stone Mountain, Eflflll.§fll;9 Aug. 1-6, 1895 (NY). Randolph Co.: banks of small creek S of Cuthbert, Harper 1111 (A, F, CH, MO, NY). Nilcox Co.: 11.5 mi 3 of Abbeville on US rt. 129, Furlow 142 (MSC). Illinois. Johnson Co.: Tunnel Hill, Palmer 12116 (MO). Pope Co.: :reekbank, Belle Smith Spring, SE of McCormick, EXEEE.§£§§Z (WIS); lear Dixon Springs, 12§E§_1122§ (MO, NY). Indiana. Brown Co.: on :he N bank of Salt Creek just E of Belmont, Furlow_11§ (MSC). DuBois 10.: 1 mi N of Bretzville along Ind. rt. 64, Furlow 116 (MSC). 'ackson Co.: in the wet woods about % mi S of Chestnut Ridge, Beam ‘1 l l l 374 2111 (IND, NY). Lake Co.: Gary, Umbach 1112 (WTU). Laporte Co.: Michigan City, Babcock 3121, June 22, 1872 (F); Long Beach, 3 mi NE, woods along hwy. US 12 just S of state line, Bartlett 1 Richards 1111 (MICH); valley of State Line Creek, 11y23'1112 (*F). Porter Co.: in wet woods near Tamarack stop along the South Shore Traction Line, Diaflfl (1ND, NY). Kentucky. Boyd Co.: Ashland, Demaree £11 (NY). I Clinton Co.: Spring Creek 3 of Albany, EEEEE.§ Hodgdon 1221 (F, NY). Fayette Co.: Lexington, 112£1_§121, 1831 (NY). Greenup Co.: South Portsmouth, Demaree 11356 (UC). Harlan Co.: near Harlan Court House, Kearney 1_(MSC, NY). Lincoln Co.: 3 mi SW of Crab Orchard, Wharton 1411_(MICH). McLean Co.: Livermore, Palmer 11191 (*MO). Powell Co.: 2% mi N of Walterville, bank of Black Creek, Wharton 1141 (MICH, NY). Rowan Co.: 2 3/4 mi S of Farmers, Wharton 1221 (MICH). Louisiana. Lincoln Co.: boggy branch near Ruston, Williams s.n., Mar. 6, 1950 (DAO, *WTU). Natchitouches Co.: Natchitouches, Palmer 1111 (M0). ‘Rapidos Co.: 11 mi SW of Alexandria, Webster §_Wilbur 1111_(NY). Sabine Co.: around shore of ”Little Lake", Hodge's Gardens, S of Many, 13212 E Solymosy 1112 (*DAO). Maine. Penobscot Co.: river thicket, scarce, Orono, Fernald 1111 (A). Sagadahoc Co.: Bowdoinham, Norton 1911 (WTU). Waldo Co.: in front of Tsuga Lodge, N shore of Megunticock Lake, Friesner 15111 (NY, 7' - > - am. am. 2 .E vs > - u m; 3. mm. E o o m - m - 3. 3. MN .3 Va 3 - S - mm. mm. NN Va V: m - > - 8. mm. a w w o s - - mm. om. cm .5 s .5 s m - mm. as. 3 Va .5 o» m; on - mm. mm. 2 Va .3 2 - E - Na. :. 2 um Va 2 - S m; 1:. Nm. 2 v: a V: m .3 my 8. om. 2 VS a > - > - as. i. 1: o w o m; z m; 3. we. 2 o a o m; z m; mm. 8. NH o a o s z m; 3. N“. S o a s s z m; 3. em. 3 s a - s - - Na. 8. a m m o m; > - ms. 3. w m m o - - - mm. 3.. a v: m > m; > as am. mm. o m m > - > - No. S. m m m > - - - aw. S. s m m > - - - mm. 3. m m m u - - - Nw. E. N m m - m - - ow. cm. H fig “6: 12% £33 fiwfi 23; “81.2 $33-32 “85:2 HOHOanMHHHD @Hn—NWMNV HOHOfl>QHUHD wHQMWHNV IUHU< UMH®U< UHHMUJVIHOCMUUQ UCDOQEOU NQHmm UHU< @UHNOHU m 551m .505 UwUMOHUCD W ofiaacmwfipm m .owmfiHOM mDCH< CH wwcsanoo UHHocwca wHQDHOmIHocmusn mo mcoHuomou uoHoo was meHm> w M .MN anmH . - my - om - so. me. «m m am > - m - mm. mm. mm m am > - u - Nw. so. Nm am am m - - - Ho. mm. Hm - - o my > - MN. HN. om m m m - - - Nw. cs. as - - > - - - mo. mm. ms xm » m - - - ma. wN. as u - > - - - es. me. as m s o s m - No. me. ms m m m m» m - No. «N. as m m 1 ms 0 - om. wN. ms xm am o - > - OH. am. Na w a 1 my 1 - on. as. as w a - - - - so. Ha. oa » s - - - - om. ma. om w m m - - - - so. am. mm a. w m o my 0 - ea. am. mm s s o - - - om. mm. mm s s m my m - ma. mm. mm s w m m» m my mm. mm. sm » s o - - - as. aw. mm - - w my as - om. we. Nm - - o - o - 0N. mm. Hm xm am my - ow - «a. ss. cm s s o my » - OH. Nm. aN s w o my m my so. am. mm sun um: u: as u: as u: fig Lamas Loam: .mum3-cao< hBasz uoHOmeHuHD manomw> uoH0w>muuHD «H933, I©Ho< oHuwo< owuwo - .> - om. mm. Hw 0 ms H mH mm. cs. ow 0 ms 0 - om. mm. as Hm . Hm - «H. mm. ma > - z - 0H. Hw. an > - m - mm. mm. on m ms o - om. oH. ms - - - - mm. mm. as > - > - am. ow. ms > - > - as. on. Na 0 ms 1 H» mm. as. HR > ms > - we. mm. OR o - m - so. we. as m - - - as. mo. we > - - - wm. No. No Hm - Hm - mH. me. so Hm - Hm - sH. mm. mo 0 - H - ma. 0N. so m - - - Ha. aN. mo 1 m» - - cm. wN. Ne w - o - OH. am. He H ms 1 - mm. mm. om m m» m . mm. aw. mm o - a» - mm. as. mm s - > - on. on. am > - > - OH. NH. mm m m» m - No. ma. mm sHQ uoz uanH HHmHH HHwHH HHwHH Hmumz Hmumz-eHu< Hmpesz uwHOH>MHUHD @Hflwwfixw HwHOH>mHuHD wHflme> IHVMUAN Ufluwnvda UHqu n > ummOH n .cmewISOHHo% n 0% m “Mafia H xm meQHDQ u m ”muofiumfl>wunpw HoHoo H H o - > - i... am. Ha - - Hm - Hm - aH. am. co m m .H m; .H - E. Hm. ow - - o - - - NN. .3. mm - - > - - - MK. S. g - - o l o - No. as. ow - - o - o - ...H. 3. mm - - o - o - No. as. aw - - 8H m; o - S. cm. mm - - H - > - as. we. % .39 Ha: EwHH HHmHH EwHH EwHH H311: 31.3-30... H352 uwHOfixfiwhuHD wHfimmH> uwHOH>MHuHD OHflfiwH> IUMU< UHH00< UHH¢U .mmmam mHQHxHD . .¢amoqm—o «amamu .mmmDm mHQHmHD .¢-osq~o ¢h¢JDmmmm .¢a~o~n~o ¢h<2u£DU¢ ¢EHHHm mHmZMJADmOfi .¢qu~woo ¢u40h~2mmh .m¢> mHWZMJJDmOfl o¢aHoaooo GHJOLHDZWH .mmmDm ¢Z¢Dzu .QaHoaooo «mowam .mmmsm ¢2¢02H .<~—oqmoo ¢2¢UZH .mmmDm ¢2<02H .¢-osqoo ¢H¢mm¢4u .m<> ¢H¢2HZDU¢ o¢qHO~noo A.¢.mv «HQZHESUQ .¢-oqmoo .¢q—Oadoo QFJDVOM«2m324¢qoeo.Hmoqoo~«OHO OO~N om~m O¢~m On~m om~m Guam OOHN Doom omom Ohom ooom omom oqom omom ONON o—om Doom Chad OOH— omuu ovna Ona— omau can" 00"— Good owou OPOH OOOH OmoH owoa Ono" ONOH OHOH 000“ .q usoucflgm 468 ‘1 no.8 .0060 60.8.8.8.mofio .836 693.360.360.09- 3.8.3 60.380.098.86. 28.3 60.8.8.3 60.86 8 mo.Ho.oo.oo.Ho.oo.mo.mo.no.no.no.mH.so.oo.mo.mH.no.mo.mHo mo.Ho.o.mH.so.cH.so.so.>o.m..so.so.Hoo so.po.po.so.oo.so.mo.po.oo.oo.¢o.no.mo.oo.mo.mo.mo.no.oo.oo.oo.ao- so.oo.mH.Ho.no.mo.mo.Ho.mo.oo.Ho.Ho.Ho.oo.Ho.mo.oo.oo.oo.oo.oo.oHo so.mo.oo._o.so.no.mo.mo.oo.mm.Ho.Ho.Ho.oo.no.no.mo.mo.Ho._o.po.mHo oo.mo.co.Ho.cH.mo.oo.Ho.wo.sv.mo.Ho.Ho.mH.so.mH.mo.mo.Ho.Ho.mo.vHo oo.mH.oo.mo.aH.mo.mo.Ho.mo.qo.oo.so.mo.mH.oo.mH.Ho.Ho.Ho.mo.Ho.mHo oo.HH.oo.mo.qo.oo.mo.o_.om.qo.mo.so.oo.oo.oo.Ho.Ho.mo.mo.>o.Ho.NHo oo.ao.oo.mo.qH.mo.mo.no.oo.m_.mo.mo.oo.q..mo.so.Ho.mo.mo.mH.oo.HHo so.mo.oo.mo.ho.mo.mo.mo.m_qu.mo.oo.mo.oo.mo.mH.mo.mo.mo.mH.mH.oHo oo.wo.wo.mo....mo.mo.Ho.Ho.oo.mo.oo.mo.aH.mo.mo.Ho._o.Ho.Ho.Ho.ooo qo.sm.NH.mo.oo.mo.no.Ho.mH.m_.oo.mo.mo.mH.mo.m..no.mo.mo.qH.oo.moo ao.om.m~.mo.m_.no.mo.mo.mH.mH.oo.no.mo.mH.oo.Ho.mo.mo.mo.aH.oo.soo no.mo.oo.mo.so.mo.mo.oo.vm.no.mo.oo.mo.mH.no.m..mo.mo.mo.Ho.no.ooo co.o«.oo.mo.mo.mo.mo.Ho.mo.cH.no.oo.mo.cH.no.qH.qo.mH.qo.Ho.oo.moo so.oo.o.mo.qH.qH.mo.mH.mo.mo.mo.m..oo.Hoo 00mm OOVN Omfim ONVN oovm om¢m OQVN omam ONQN OHQN oo¢m comm ommm Ohmm 000m omom oqmm omnm ommm OHMN comm Comm ommm Obwm 00mm ommm O§NN omwm ONNN Guam 00mm Comm om—m Oh~m .AmmwcHucoov q uSOHEHHm 469 000.000«ccc.hccqccc.bcc.cocomcc«cocomccqccc.cccacoc.mcc~ccc.mccgoon Hdcohccammc.mmc~dec.cmc«mHc.mcc.cmc.occ«mcc.cccqcuc.mccswocohccuoHc Hac.hccqhoccmmcqcmfl.ch.cnc.mochHc.ccothc.ccc~mmc.occqmoc.boc«mHc m~c.>ccqwmnowficqm¢H.mmc.mwc."ccqwgc.moc«¢cc.ccc~mmc.mccsocc.mccq¢«c HHcowccumhc.HNcamqn.«qcawcc._cosfiHo.mccumcc.ccc«mmcomccacacoficcqch vac.cccamhc.mmc~cmn.mnc«chomcc«occomccqmmc.mccammc.mcc~mcc.occuNHO m~c.mccnmhc.o~cshNH.HmcamHc.mcc~mcc.mccsmccocccsch.mcc~HHc.occ«HHc vuc.occqchcocmcth~.cmogcmcomccgmcc.mccsVOO.Hcc«oHc.mccq®ccomccqc~c nmoooccqmvcowucacoc.mmcqucomccqmcconcc.ccc.cco~cco.ccc~€cc.nccqmcc cmc.wccnmhcocmoqmo~.mmcgmccoaccqmcc.Hcchcc.cccquc.occ~mcc.mccawcc adcooccawwoommchMH.mnchqc.mcc.booonccsncc.cccabaooccc.wcc.mcc«boo m~c.wccgvhc.¢mcacoc.omcqm~0.mcc~c~c.ncc.bHc.Hocxodc.mccqbcc.mccaocc QHO.HHc~mhc.chqmc~.owc.c~c.mccam.c.mcc~¢cc.~cc«<~c.¢cc~cHcooccqmcc "Hc.bccqcoc.coovcc~mccc"ccahcc.mccnmfico¢ccuomc.ccc.ch.ccc~occ c~coccc~mccouccahcc.mcc«ficc.mco~NHc.mcc.~Hc.©ccquc.ccc.mcc.occ.mcc accomccsmcoowccsocc.qccancc.~00thcomcc.c~o.mccsmmc.ccc«mmc.ccc~vcc 000.500.ficc.ncoqocc.mcc«©cc.mccquc.mccach.cccacc~.ccc.cmc.ccc.ncc mHc.bccacco.mccqmcc.vcc.mo.m.o.o~o.moo.Hoo.moo.qoo.mmo.moo.moo.noo.Hoo.ooo.oHo soo.moo.ooo.omo.mHo.oHo.ooo.Hoo.qoo.qoo.omo.moo.oHo.moo.doo.ooo.ooo moo.qoo.oHH.mmo.mHo.mHo.aoo.ooo.moo.moo.nmo.moo.qHo.qoo.Hoo.ooo.moo moo.soo.moo.omo.mHo.oHo.moo.ooo.qoo.woo.mHo.qoo.HHo.qoo.Hoo.ooo.soo ooo.~oo.ooH.oqo.mHo.oHo.Hoo.ooo.qoo.moo.mmo.ooo.ooo.moo.Hoo.ooo.ooo moo.moo.oso.omo.oHo.soo.moo.ooo.moo.moo.omo.moo.oHo.moo.1oo.ooo.moo ooo.ooo.mwo.omo.omo.mHo.moo.ooo.moo.moo.mmo.moo._Ho.moo.Hoo.ooo.qoo moo.ooo.s¢o.moo.mHo.oHo.qoo.Hoo.ooo.qoo.smo.moo.oHo.qoo.Hoo.ooo.moo oHo.moo.mmH.omo.mHo.oHo.moo.Hoo.soo.soo.mmo.qoo.mHo.moo.Hoo.ooo.moo HHo.qoo.omH.omo.omo.oHo.ooo.ooo.moo.moo.mmo.moo.mHo.moo.Hoo.ooo.Hoo cmum cb~m OOHn cmfim cwum Oman cmnn cuqn comm coco cwcn choc coca cmcm cqcm cncm cmcn c—cn coca comm cwom crow comm cmmm ccmm cmmm cmom chN 000m comm cmmm chmm comm cmmm .AWMWCHumMWV q “soucHum 471 coo.moo.ao- oo-.oo-.oHo oom.ooo.mHo oom.ooo.oo.moo.mHo.mHo.mmo.mHo.eoo.moo.coo.noo.w_o.moo.oHo oqo.qoo.moo.aoo.moo.Hoo.>mo.mmo.oqo.oHo.moo.ooo.moo.moo.mHo.m_o.mmo.mmo.moo.noo.moo.qoo.mao.mHo.moo ono.oHo.oHo.moo.soo.moo.sHo.mHo.omo.smo.moo.moo.qoo.moo.omo.mdo.moo moo.soo.ouo.moo.oHo.moo.mHo.oHo.omo.mHo.moo.ooo.moo.noo.omo.oHo.Loo cmmm Odom comm cmfim omHum Oh§m ooqm cmqm O§¢m cmqm cmflm c~§m OOVm comm comm Obmm comm comm cqmm cmmm cmmm c—mm comm comm comm cbmm comm comm cqmm cmmm ommm chm comm cmum . Amacficmmv 1H H533 ..HnH 472 moo {mmwfi 2¢ouzcmt amz~m2¢4 Hm¢w >h~mmw>HZD wH¢Hm chmruuz yccdcmh¢m H2¢Am QZQ >Z¢Hcm kc .Hmmo moo Bcamcm .7 zmco wm moo mhz¢um¢cu> 2¢Hm¢mcm mumxfi 92¢ Am¢mc¢qohmmo mczq¢ mo ¢X¢H 2¢c~mm2¢ ark hccuofiqchocuuosumcudzuccdumm¢rcqoflu¢x¢8qomoumo«cudmmq¢hqoocmu>mm mquuczummmqmoumOZuccmc«Qm324duhmH4m¢mc~£m324¢um4Hh ccdc c 09‘ cmcw Ohcq coc¢ cmoq cvo¢ cmc§ cmcfl c~o¢ coca .AwMWCHummmv q usoucflum 473 Printout 5. Listing of ALNUSD, the character list for taxonomic data matrix ALNUSM. 474 Printout 5. 1000 1010 1020 1030 1040 1050 1060 1070 1080 1090 1100 1110 1120 1130 1140 1150 1160 1170 1180 1190 1200 1210 1220 1230 1240 1250 1260 1270 1280 1290 1300 1310 1320 1330 1340 1350 1360 1370 1380 1390 1400 1410 1420 1430 1440 1450 1460 1470 1480 1490 1500 1510 FILE ALNUSD- CHARACTERS FOR NEW WORLD SPECIES OF ALNUS- 001:0:HABIT 001:1:SHRUBBY 001:3:ARBORESCENT 002300LEAVES USUALLY BORNE 002:1:ON WELL‘MARKED SHORT SHOOTS 002:3:DIRECTLY ON THE MAIN STEM 003:0:YOUNG STEMS 003:1:WITH CONSPICUOUS LONGITUDINAL RIDGES 003:3:WITHOUT CONSPICUOUS LONGITUDINAL RIDGES 004:0:BUD APEX 004:1:ACUTE OR ACUMINATE 004: 3.9ROLNDED 005: 0: WINTER BUDS OOSJIaNAKED OR COVERED BY 2 OR 3 EQUAL STIPULAR SCALES OOSJSJCOVERED BY 4 OR MORE UNEQUAL SCALES 006:0;BUDS 006:1:WITH A THICK: LUSTROUS: RESINOUS COATING 00613.91!” THOUT A THICK RESINOUS COATING 007:0:STIPULE APEX 007:1:ACUTE OR ACUMINATE 007: 3: ROUNDED 008:0:LEAF SERRATIONS 008a13REGULAR IN SIZE 008:3:IRREGULAR IN SIZE 009:0:LEAF LOBES OR MAJOR TEEHI 009:1:REGULAR IN SIZE 009:3:IRREGULAR IN SIZE 010: O:LEAVES 010:1:GLAUCOUS BELOW 010.3:NOT GLAUCOUS BELOW 011:0:MAJOR VEINS ON LOWER SURFACE OF LEAF 011:1:PROMINENT 011:3:NOT PROMINENT 012:0:LATERAL VEINS 012:1)NEARLY STRAIGHT 012:3:ASCENDING 013:01LATERAL VEINS TERMINATING 013:1:IN TEETH OR LOBES ON THE MARGIN 013:3:BY BRANCHING AND CONNECTING WITH OTHER VEINS 014:0:CROSS VEINS BETWEEN THE LATERAL VEINS 014:1:STRONG (PROMINENT) 014:3IWEAK (OBSCURE) 015:0:PISTILLATE CATKINS 015:1:SOLITARY IN THE AXILS OF LEAVES OISJSJBORNE IN RACEMOSE CLUSTERS 01610:BRANCHLET BEARING PISTILLATE CATKINS 016:1:STRAI GHT 016:3:BENT SHARPLY AWAY FROM MAIN AXIS OI7JOJBRANCHLET BEARING STAMINATE CATKINS 017:1:STRAIGHT 017:3:BENT SHARPLY AWAY FROM MAIN AXIS 475 Printout 5 (Continued). 1520 1530 1540 1550 1560 1570 1580 1590 1600 1610 1620 1630 1640 1650 1660 1670 1680 1690 1700 1710 1720 1730 1740 1750 1760 1770 1780 1790 1800 1810 1820 1830 1840 1850 1860 1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020 2030 018:0:TERMINAL LUBE OF CONE SCALE 018:1:ELONGATE OR EXTENDED 018:3:NOT EXTENDED 01910:APEX 0F CONE SCALE 019:1:FLAT 019:3:BENT SHARPLY DOWNWARD 020:0:FRUIT WING 020:1:TH1N (MEMBRANACEOUS) 020539C0REACE0US OR WOODY 021:0:BARK 021:1)LIGHT GRAY OR BROWN 021:2:BROWN 021:3:DARK BROWN OR RED 021:4:BLACK 022:0:BARK 022:1:SMO0TH 022:2:50MEWHAT TEXTURED: BUT UNBROKEN 022:3:BROKEN UP INTO SMALL PLATES 022:4:DEEPLY FISSURED 023:0:LENTICELS 0F TRUNK 023:1:CIRCULAR 023:2:ELLIPTIC 023:3:ELONGATE 023:4:L1NEAR 024:0:LENTICELS 0F TRUNK 024:1:SMALL 024:2)0F MODERATE SIZE 024:3:LARGE 024141VERY LARGE 024:0;LENTICELS 0F TRUNK 025:1:INCONSPICUOUS 025:2;OBVIOUS 025:3:VERY PROMINENT 026:0:STEMS 026:1:GLABROUS 026:2:SPARSELY PUBESCENT 026:3:PUBESCENT 026:4:DENSELY PUBESCENT 027:0:STEM PUBESCENCE 027:1:SHORT 027:2:OF MODERATE LENGni 027:3:LONG OZBIOISTEM SURFACE 028:1:WITHOUT GLANDS 028:2:SPARSELY GLANDULAR 028:3:MODERATELY GLANDULAR 028:4:DENSELY GLANDULAR 029,0:LENGTH 0F BUD STALK 029:1:LESS THAN 1/4 THE LENGTH OF THE BUD 029:2:1/4 T0 3/4 fiiE LENGTH OF THE BUD 029:3:MORE THAN 3/4 THE LENGTH OF THE BUD 030:0:STIPULES .n.,—_— . "—.""-.:-:;-.' _— ' 476 Printout 5 (Continued). 2040 030:1:0VATE 2050 030:2:ELLIPTIC 2060 030:3:080VATE 2070 031:0:LEAVES 2080 031:1:0VATE 2090 031:2:ELLIPTIC 2100 031:3:030VATE 2110 032:0:LEAF BASE 2120 032:1:ATTENUATE 2130 032:2:ACUTE 2140 032:3:CUNEATE 2150 032:4:ROUNDED 2160 032:5:TRUNCATE 2170 032:6)CORDATE 2180 033:0:LEAF MARGIN 2190 033:1:DISTANTLY SERRULATE 2200 033:2:SERRULATE 2210 033:3:SERRATE 2220 033:4)DOUBLE'SERRATE 2230 033)5:52RRATE-DENTICULATE 2240 033:6:SERRATE-LOBED 2250 034:0:LEAF LOBES 2260 034:1:LESS THAN 1/4 AS DEEP AS WIDE 2280 034:2:1/4 T0 1/2 AS DEEP AS WIDE 2290 034:3:1/2 T0 2 TIMES AS DEEP AS WIDE 2300 034:4:MORE THAN 2 TIMES AS DEEP AS WIDE 2310 035:0:LEAF MARGINS 2320 035:1:NOT THICKENED 2330 035:2:SLIGHTLY THICKENED 2340 035:3:GREATLY THICKENED 2350 036)0:LEAF MARGINS 2360 036:1:FLAT 2370 036:2)SLIGHTLY REVOLUTE 2380 036:3:STRONGLY REVOLUTE 2390 037:0:LEAF APEXES 2400 037:1:ACUM1NATE 2410 037a23ACUTE 2420 037:3:OBTUSE 2430 037:4:ROUNDED 2440 037151TRUNCATE 2445 038:0:LEAVES 2450 038:1:LESS THAN 1 1/2 TIMES AS LONG AS WIDE 2460 0383211 1/2 T0 2 1/2 TIMES AS LONG AS WIDE 2470 038:3:MORE THAN 2 1/2 TIMES AS LONG AS "IDE 2480 039:0)UPPER SURFACE OF LEAVES 2490 039:1:LIGHT GREEN 2500 039:2:MEDIUM GREEN 2510 039:3:DARK GREEN 2520 039:4:VERY DARK GREEN 2530 040:0:LOWER SURFACE OF LEAVES 2540 040:1:LIGHT GREEN 2550 040:2:MEDIUM GREEN 477 Printout 5 (Continued). 2560 2570 2580 2590 2600 2610 2620 2630 2640 2650 2660 2670 2680 2690 2700 2710 2720 2730 2740 2750 2760 2770 2780 2790 2800 2810 2820 2830 2840 2850 2860 2870 2880 2890 2900 2910 2920 2930 2940 2950 2960 2970 2980 2990 3000 3010 3020 3030 3040 3050 3060 3070 040:3:DARK GREEN 040:4:LIGHT BROWN 040:5:MEDIUM BROWN 040:6:DARK BROWN O41: O:LEAF BLADES 041:1:MEMBRANACEOUS 041:2:CHARTACEOUS 041:3:COREACEOUS 042:0:LOVER SURFACE OF LEAF BLADES 042:1:GLABROUS 042:2:SPARSELY PUBESCENT 042:3:PUBESCENT 042:4:DENSELY PUBESCENT 043:0:PUBESCENCE ON LOWER SURFACE OF LEAF BLADES 043: 1: SHORT 043:2:OF MODERATE LENGTH 043: 3: LONG 044:0:VEINS ON LOWER SURFACE OF LEAVES 044:1:GLABROUS 044:2:SPARCELY PUBESCENT 044:3:PUBESCENT 044:4:[EmHHflJr PUBESCENT 045:0:PUBESCENCE ON LEAF VEINS 045: 1: SHORT 045:2:OF MODERATE LENGTH 045:3:LONG 046:0:PUBESCENCE ON LOWER SURFACE OF LEAVES 046: 1: WI TI 5“ 046:2:YELLOWISH 046: 3: BROWNI SH 046: 4: DARK BROWN 047:0:TIPS OF MAJOR TEETH 047:1:WITHOUT PROMINENT GLANDS 1147:2:SLIGHTLN'(HANNDULAR 047:3:GLANDULAR 048:0:LOWER SURFACE OF LEAF BLADES 048:1:HHHDHDUT GLANDG 048: 2: SPARSELY GLANDULAR 048:3:MODERATELY GLANDULAR 048:4:DENSELY GLANDULAR 049:0:LENGTH OF LEAF BLADES 049:1:LE55 THAN 5 TIMES THE LENGTH OF DRE PETIOLE 049:2:BETWEEN 5 AND 10 TIMES THE LENGTH OF THE PETIOLE 049:3:MORE THAN 10 TIMES THE LENGTH OF THE PETIOLE 050:0:PETIOLES 050: 1: GLABROUS 050: 2: SPARSELY PUBESCENT 050:3:PUBESCENT 050:4mEWmHHHJYIPUBESCENT 051:0:PETIOLE PUBESCENCE 051: 1: SHORT 051:2bCHVPMIDERATE LENGTH 478 Printout 5 (Continued). 3080 3090 3100 3110 3120 3130 3140 3150 3160 3170 3180 3190 3200 3210 3220 3230 3240 3250 3260 3270 3280 3290 3300 3310 3320 3330 3340 3350 3360 3370 3380 3390 3400 3410 3420 3430 3440 3450 3460 3470 3480 3490 3500 3510 3520 3530 3540 3550 3560 3570 3580 3590 051:3:LONG 052:0:PISTILLATE AND STAMINATE CATKINS 052:1:BORNE ON THE SAME BRANCHLET 052:2:BORNE 0N WELL SEPARATED BBANCHLETS 052:3:BORNE 0N CLOSE BRANCHLETS 053:0:MATURE INFRUCTESCENCES (COMES) 053:1:0V01D 053: 2: ELLIPSOID 053:3:OBOVOID 053:4:CYLINDRICAL 054:0:CONE SCALE APEXES 054:1:NOT THICKENED 054:2:MODERATELY THICKENED 054:3:GREATLY THICKENED 055:0:STAMENS 055:1:LONGER THAN THE PERIANTH 055:2:APPROXIMATELY EQUAL IN LENGTH TO THE PERIANTH 055:3:SHORTER THAN THE PERIANTH g 056:0:HABITAT IN 056:1:SUPERHUMID ZONE 056:2:HUMID ZONE 056:3:MOIST'HUMID ZONE 056:4:DRY'HUMID ZONE 057:0:GROWING 057:1:IN STANDING 0R RUNNING WATER 057:2:BESI DE STANDING OR RUNNING WATER 057:3:IN AREAS OF INTERMITTENT STANDING OR RUNNING WATER 057:4:0N DRY GROUND 058:0:FLOVERING FROM 058:1:MARCH T0 JUNE 058:2:JULY T0 OCTOBER 058:3:NOVEMBER T0 FEBRUARY 059:0:PISTILLATE CATKINS FORMED 059:1:THE SEASON PREVIOUS T0 FLOWERING 059:2:THE SEASON 0F FLOWERING WITH THE NEW LEAVES 059:3:THE SEASON 0F FLOWERING AFTER THE NEW LEAVES 060:0: TRLNKS 060:9:Mo HIGH 061:0:TRUNKS 061:9:CM- IN DIAMETER 062:0:0NE YEAR'S GROWTH HAVING 062:9:NODES 063:0:LENTICELS 0N YOUNG TWIGS 063:9:MICROMETERS LONG 064:0:BUD STALKS 064:9:MM0 LONG 065:0:BUDS 065:9:MM0 LONG 066:0:BUDS 066:9:MM-WIDE 067:0:STIPULES 067:9:MMo LONG 479 Printout 5 (Continued). 3600 3610 3620 3630 3640 3650 3660 3670 3680 3690 3700 3710 3720 3730 3740 3750 3760 3770 3780 3790 3800 3810 3820 3830 3840 3850 3860 3870 3880 3890 3900 3910 3920 3930 3940 3950 3960 3970 3980 3990 4000 4010 4020 4030 4040 4050 4060 4070 4080 4090 4100 4110 068:0:STIPULES 068:9:MM; WIDE 069: OaLEAF LOBES 069:9:MM0 APART 070:0:LEAF LOBES 070:9:MMo DEEP 071:0:LEAF SERRATIONS 071:9:PER CM- 072:O:LEAF SERRATIONS 072:9:MICROMETERS DEEP 073:0:LEAVES 073:9:MM0 LONG 074:0:LEAVES 074:9:MM0 WIDE 075:0:LEAVES WITH 075:9:PAIRS OF LATERAL VEINS 076:0:LATERAL VEINS OF LEAVES BRANCHING AGAIN 076:9:TIMES 077:0:LATERAL VEINS AT MID'LEAF 077:9:MM- APART 078:0:PETIOLES 078:9:MM- LONG 079:0:PISTILLATE CATKINS 079:9:MM- LONG AT ANTHESIS 080:0:PEDUNCLES 0F PISTILLATE CATKINS 080:9:MM0 LONG AT ANTHESIS 081:0:PEDUNCLES OF PISTILLATE CATKINS 081:9:MICROMETERS WIDE AT ANTHESIS OSEIOISTAMINATE CATKINS 082:9:MM0 LONG 083:0:STAMINATE CATKINS 083:9:MM0 IN DIAMETER 084:0:PEDUNCLES OF STAMINATE CATKINS 084:9:MMo LONG OBSIOISTAMENS 085:9: 086:0:FILAMENTS 086,9:MICROMETERS LONG 087:0:ANTHERS 087:9:M1CROMETERS LONG 088:0:ANTHERS DIVIDED AT APEX 088:91PER CENT OF TOTAL LENGTH 089:0:MATURE INFRUCTESCENCES (CONES) 089:9:MM0 LONG 090:0:MATURE INFRUCTESCENCES (CONES) 090:9:MM0 IN DIAMETER 091:0:PEDUNCLES OF MATURE INFRUCTESCENCES (CONES) 091:9:MM0 LONG 092:0:PEDUNCLES OF MATURE INFRUCTESCENCES (CONES) 092:9:MICROMETERS IN DIAMETER 093:01CONE SCALES 093:9:MM- LONG 480 Printout 5 (Continued). 4120 4130 4140 4150 4160 4170 4180 4190 4200 4210 4220 4230 4240 4250 4260 094:0:CONE SCALES 094:9:MM0 WIDE AT VIDEST POINT 09 5: 0: F301 TS 095:9:MICROMETERS LONG 096:0:FRUIT BODIES 096:9:MICROMETERS WIDE 097: O: FRUI T WINGS 097:9:MICROMETERS WIDE 098:0:PERSISTENT STYLES 0N FRUITS 098:9:MICROMETERS LONG 099:0:LATITUDE 0F RANGE 099:9:DEGREES 100:0:HABITAT ELEVATION 100:9:DECAMETERS '99:O:END 0F DESCRIPTOR LIST LITERATURE CITED LITERATURE CITED Abbe, E.C. 1935. Studies in the phylogeny of the Betulaceae. 1. Floral and inflorescence anatomy and morphology. Bot. Gaz. 97: 1-670 . 1938. Studies in the phylogeny of the Betulaceae II. Extremes in the range of variation of floral and inflorescence morphology. Bot. Gaz. 99: 431-469. Acosta-Solis, M. 1939. Principal timbers used in the Sierra del Ecuador. Tropical Woods 57: 3-5. Adams, R.P. 1972. Chemosystematic and numerical studies of natural populations of Juniperus pinchotii Sudw. Taxon 21: 407-427. . 1974. On "numerical chemotaxonomy” revisited. Taxon 23: 336-338. and B.L. Turner. 1970. Chemosystematics and numerical of natural populations of Juniperus ashei. Taxon 19: 728-751. Aiton, W. 1789. Hortus Kewensis. London. Vol. 3. Aiton, W.T. 1813. Hortus Kewensis, ed. 2. London. Vol. 5. Alston, R.E. 1966. Chemotaxonomy or biochemical systematics. 22. T. Swain (ed.). Comparative Phytochemistry. New York. Pp. 33- 56. . 1967. Biochemical systematics. IE_T. Dobzhansky, M.K. Hecht, and N.C. Steere (eds.). Evolutionary Biology, vol. 1. and B.L. Turner. 1959. Applications of paper chroma- tography to systematics: recombinations of parental biochemical components in Baptisia hybrid populations. Nature 184: 285- 286. and B.L. Turner. 1963. Biochemical Systematics. Engle- wood Cliffs, N.J. 404 pp. Anderson, E. and E.C. Abbe. 1934. A quantitative comparison of spe- cific and generic differences in the Betulaceae. Jour. Arn. Arn. Arb. 15: 43-49. 481 , -. .e — _v~_=.-'-""'”’ 482 Ascherson, P.F.A. 1864. Flora der Provinz Brandenburgh. Berlin. Vol. 1. Axelrod, D.I. 1939. A Miocene flora from the western border of the Mojave Desert. Carnegie Inst. Wash. Publ. 516, pp. 1-129. . 1952. A theory of angiosperm evolution. Evolution 6: 29"60. 4 and P.H. Raven. 1972. Evolutionary biogeography viewed from plate tectonic theory. ‘In_J.A. Behenke (ed.). Challenging Biological Problems. New York. Pp. 218-236. Bailey, I.W. 1911. The relation of the leaf trace to the formation of compound rays in the lower dicotyledons. Ann. Bot. 25: 225- 241. . 1912. The evolutionary history of the foliar ray in the wood of the dicotyledons and its phylogenetic significance. Ann. Ball, P.W. 1964. Alnus. EE.T°G' Tutin, V.H. Heywood, N.A. Burges, D.H. Valentine, S.M. Walters, and D.A. Webb (eds.). Flora Europaea. Cambridge. Vol. 1, p. 59. Bartlett, A.S. and E.S. Barghoorn. 1973. Phytogeographic history of the Isthmus of Panama during the past 12,000 years (a history of vegetation, climate, and sea-level change). EE.A° Graham (ed.). Vegetation and Vegetational History of Northern Latin America. Amsterdam. Pp. 203-299. Bartlett, H.H. 1909. Notes on Mexican and Central American alders. Proc. Amer. Acad. 44: 609-612. Bate-Smith, E.C. 1963. Usefulness of chemistry in plant taxonomy as illustrated by flavenoid constituents. EE.T' Swain (ed.). Chemical Plant Taxonomy. New York. Pp. 127-139. Benson, L. 1962. Plant Taxonomy: Methods and Principles. New York. 494 pp. Berry, E.W. 1922. Late Tertiary plants from Jancocata, Bolivia. In E.B. Matthews (ed.). Contributions to the Paleobotany of Peru:—- Bolivia, and Chili. The Johns Hopkins University Studies in Geology, No. 4. Baltimore. Pp. 207-221. 1923. Tree Ancestors: a Glimpse into the Past. Balti- more. 270 pp. . 1926. Tertiary floras from British Columbia. Bull. Can. 483 Boivin, B. 1967. Enumeration des plantes du Canada. VII. Resume statistique et regions adjacentes. Nat. Canad. 94: 625-655. Bond, G. 1955. Evidence for fixation of nitrogen by root nodules of alder (Alnus) under field conditions. New Phytol. 55: 147-153. . 1963. The root nodules of non-leguminous angiosperms. In P.S. Nutman and B. Mosse (eds.). Symbiotic Associations. London. Pp. 72-91. , W.W. Fletche, and T.P. Ferguson. 1954. The development and function of the root nodules of Alnus, Myrica, and Hippophae. Plant and Soil 5: 309-323. Bongard, H.G. 1833. Observations sur la vegetation de 1'isle de Sitcha. Mem. Akad. Sci. St. Petersb. ser. 4, 2: 44. Boubier, A.M. 1896. Recherches sur l'anatomie systematique des Betulacees-Corylacees. Malpighia 10: 349—436. Braun, B.L. 1951. Plant distribution in relation to the glacial boundary. Ohio Jour. Sci. 51: 139—145. . 1955. The phytogeography of unglaciated eastern United Bot. Rev. 21: 297-375. States and its interpretation. Brehm, B.G. 1966. Taxonomic implications of variation in chromato- Brittonia 18: 194-202. graphic pattern components. Annotated catalogue of the vascular flora of 1957a. 1—72. Breitung, A.J. Amer. Midl. Natrl. 58: Saskatchewan. . 1957b. Plants of Waterton Lakes National Park, Alberta. 71: 39-71. Canad. Fld.-Natr1. An undescribed species of Alnus. Torreya 4: 124. Britton, N.L. 1904. Brown, D.M. 1941. The vegetation of Roan Mountain, a phytosocio- logical and successional study. Ecol. Mon. 11: 61-97. Brown, R.W. 1937. Additions to some fossil floras of the western United States. U.S. Geol. Surv. Prof. Paper 186-J. Cain, A.J. and C.A. Harrison. 1960. Phyletic weighting. Proc. 2001. Soc. Lond. 135: 1-31. Calder, J.A. and R.L. Taylor. 1968. Flora of the Queen Charlotte Islands. Pt. 2. Cytological Aspects of the Vascular Plants. Can. Dept. of Agr. Mon. 4. Ottawa. Uber die in Schlesien vorkommenden Formen der Callier, A. 1892. _ ._ . -. .. —-g 3,”;— \.-‘— -—~;_ .— 484 Gattung Alnus. Jahresbericht der Schlesischen Gesellschaft fur Vaterlandische Cultur 69(2): 72-85. . 1904, 1912. Gattung Alnus. IE_C.K. Schneider. Illus- triertes Handbuch der Laubholzkunde. Jena. Vol. 1, pp. 119- 136; vol. 2, pp. 857-891. . 1911. Diagnoses formarum novarum generis Alnus. Fedde Repert. Sp. Nov. 10: 225-237. . 1918. Alnus Formen der europaischen Herbarien und Garten. Mitt. Deutsch. Dendr. Gesel. 27: 39-185. Camin, J.H. and R.R. Sokal. 1965. A method for deducing branching sequences in phylogeny. Evolution 19: 311-326. Camp, W.H. and C.L. Gilly. 1943. The structure and origin of species. Brittonia 4: 323-385. Chamberlain, C.J. 1927. Hayfever in the Pacific North West. Ann. Otol. Rhinol. Laryngol. 36: 1083-1092. Chaney, R.W. 1936. The succession and distribution of Cenozoic floras around the northern Pacific basin. In Essays in Geo- botany in Honor of W.A. Setchell. Berkeley. Pp. 55-85. . 1944. Tertiary centers and migration routes. Ecol. Mon. 17: 143. . 1957. Miocene floras of the Columbia Plateau. Pt. I. Composition and Interpretation. Carnegie Inst. Wash. Publ. 617, pp. 1—134. and D.I. Axelrod. 1957. Miocene floras of the Columbia Plateau. Pt. II. Systematic Considerations. Carnegie Inst. Wash. Publ. 617, pp. 135-237. Chiba, S. 1962. Studies on the breeding of Betula and Alnus Species. (1) On the differences of morphological characters and chromosome numbers between Alnus hirsuta and Alnus hirsuta var. microphylla. Jour. Jap. For. Soc. 44: 237-243. Clarkson, R.E. 1960. Notes of the distribution of Alnus crispa in eastern North America. Castanea 25: 83-86. Clausen, R.T. 194g. gheckligt 03 the vascular plants of the Cayuga Quadrangle 42 -43 N, 76 -77 W. Mem. Cornell Univ. Agr. Expt. Sta. 29].: 1‘87. Contandriopoulos, J. 1964. Recherches sur la flore endemique de la corse et sur ses origines. (II). Revue General de Botanique 71: 361-384. 485 Coon, N. 1963. Useful Plants for Healing. Great Neck, N.Y. 272 pp. C00per, W.S. 1923a. The recent ecological history of Glacier Bay, Alaska. I. The interglacial forests of Glacier Bay. Ecol. 4: 93-128. . 1923b. The recent ecological history of Glacier Bay, Alaska. II. The present vegetation cycle. Ecol. 4: 223-246. . 1931. A third expedition to Glacier Bay, Alaska. . 1939. A fourth expedition to Glacier Bay, Alaska. Ecol. 20: 130-155. Crawford, D.J. and R.D. Dorn. 1974. "Numerical chemotaxonomy" and other aspects of chemosystematics. Taxon 23: 331-335. Crocker, R.L. and B.A. Dickson. 1957. Soil development on the reces- sional moraines of the Herbert and Mendenhall glaciers, south- eastern Alaska. Jour. Ecol. 45: 169-185. Crocker, R.L. and J. Major. 1955. Soil development in relation to vegetation and surface age at Glacier Bay, Alaska. Jour. Ecol. 43: 427-448. Crovello, T.J. 1968. Key communality cluster analysis as a taxonomic tool. Taxon 17: 241-258. . 1969. Effects of change of characters and of number of characters in numerical taxonomy. Amer. Midl. Natrl. 81: 68-86. . 1970. Analysis of character variation in ecology and systematics. Ann. Rev. Ecol. Syst. 1: 55-98. Culpeper, N. 1653. The Complete Herbal and English Physitian Enlarged. London. 398 pp. Czerepanov, S. 1955. Systema generis Alnus Mill. S. Str. Generumque affinium. Notulae Systematicae ex Herbario Instituti Botanici Nomine V.L. Komarovi Academiae Scientiarum U.R.S.S. 17: 90-105. Davis, G.L. 1966. Systematic Embryology of the AngiOSperms. New York. 528 pp. Demcker, R. 1909. Neue Geh81ze. Mitt. Deutsch. Dendr. Gesel. 18: 323-326. Desfontaines, R.L. 1829. Catalogus Plantarum Horti Regnii Parisiensis, ed. 3. Paris. 484 pp. 486 Dippel, L. 1892. Handbuch der Laubholzkunde. Berlin. Vol. 2. Dixon, D. 1961. These are the champs. Amer. Forests 67(1): 40-46; 48-50. Dorman, F. 1924. Zur Kenntnis der Hautdrfisen und der Harzexkretion von Alnus viridis. S.B. Akad. Wiss. Wien 133: 585-612. Doyle, J.A. 1969. Cretaceous angiosperm pollen of the Atlantic coastal plain and its evolutionary significance. Jour. Arn. Arb. 50: 1-35. Dressler, R.L. 1954. Some floristic relationships between Mexico and the United States. Rhodora 56: 81—96. DuRietz, G.E. 1930. The fundamental units of biological taxonomy. Svensk. Bot. Tidskr. 24: 333—428. DuRoi, J.P. 1771. Dissertatio Inauguralis Observationes Botanicas Sistens. Helmstadii. 62 pp. Edlin, H.L. 1964. A modern sylva or a discourse of forest trees. 11. A1der--A1nus spp. Quart. Jour. Forest. 58: 302-310. Ehrhart, B. 1753. Oekonomische Pflanzen-Historie. Ulm. Vol. 2. Ehrhart, F. 1788. Beitrage zur Naturkunde. Hannover und Osnabruk. Vol. 2. Ellison, W.L., R.E. Alston, and B.L. Turner. 1962. Methods of pre— sentation of crude biochemical data for systematic purposes, with particular reference to the genus Bahia (Compositae). Amer. Jour. Bot. 49: 599-604. Endlicher, S. 1842. Genera Plantarum Secundum Ordines Naturales Deposita, supplementum secundum. Vindobonae. 114 pp. Erdtman, G. 1969. Handbook of palynology. New York. 486 pp. Farris, J.S. 1967. The meaning of relationship and taxonomic procedure. Syst. 2001. 16: 44-51. . 1970. Methods for computing Wagner trees. Syst. 2001. 19: 83-92. , A.G. Kluge, and M.J. Eckardt. 1970. A numerical approach to phylogenetic systematics. Syst. 2001. 19: 172—189. Fernald, M.L. 1904a. The green alders of New England. Rhodora 6: 162-163. 487 . 1904b. Synopsis of the Mexican and Central American species of Alnus. Proc. Amer. Acad. 40: 24-28. . 1907. Diagnoses of new spermatophytes from Mexico. Proc. Amer. Acad. 43: 61-68. . 1913. Alnus crisPa (Ait.) Pursh var. mollis Fern. n. comb. Rhodora 15: 44. . 1914. A cut-leafed alder. Rhodora 16: 56. . 1945a. Alnus criSpa (Ait.) Pursh, forma stragula. Rhodora 47: 144. . 1945b. Eastern North American representatives of Alnus incana. Rhodora 47: 333-361. . 1950. Gray's Manual of Botany, ed. 8. New York. 1632 PP- Fish, F. and P.G. Waterman. 1973. Chemosystematics in the Rutaceae. II. The Chemosystematics of the Zanthozylum/Fagara complex. 3 Taxon 22: 177-203. ' Flake, R.H., E. Van Rudloff, and B.L. Turner. 1969. Quantitative study of clinal variation in Juniperus virginiana using ' terpenoid data. Proc. Nat. Acad. Sci. 64: 487-494. Forsaith, C.C. 1920. Anatomical reduction in some alpine plants. Fryxell, P.A. 1967. The interpretation of disjunct distributions. Taxon 16: 316-324. Gaertner, J. 1791. De Fructibus et Seminibus Plantarum. Stut- gardiae. Vol. 2. Gandoger, M.M. 1919. Sertum plantarum novarum. Bull. Soc. Bot. Giannasi, D.E. and C.M. Rogers. 1970. Taxonomic significance of floral pigments in Linum (Linaceae). Brittonia 22: 163-174. Gillis, W.T. 1971. The systematics and ecology of poison-ivy and the poison-oaks (Toxicodendron, Anacardiaceae). Rhodora 73: 72—159; 161-237; 370-443; 465-540. Gleason, H.A. 1963. The New Britton and Brown Illistrated Flora of the Northeastern United States and Adjacent Canada. New York. V0].- 3, pp. 36-37. 488 and A. Cronquist. 1963. Manual of the Vascular Plants of Northeastern United States and Adjacent Canada. New York. 810 pp. Goldman, B.A. 1951. Biological investigations in Mexico. Smithson- ian Institution Publ. 4017. 476 pp. Graham, A. 1972a. Outline of the origin and historical recog- nition of floristic affinities between Asia and eastern North America. IE_A. Graham (ed.). Floristics and Paleofloristics of Asia and Eastern North America. Amsterdam. Pp. 1-18. . 1972b. Some aspects of Tertiary vegetational history about the Carribbean basin. Memorias de Symposia, I Congreso Latinoamericano V Mexicano.de Botanica. Mexico, D.F. Pp. 97- 117. . 1973. History of the arborescent temperate element in the northern Latin American biota. IE_A. Graham (ed.). Vege- tation and Vegetational History of Northern Latin America. Amsterdam. Pp. 301-314. Gram, K., C.M. Larsen, C.S. Larsen, and M. Westergaard. 1942. Con- tributions to the cytogenetics of forest trees. II. Alnus studies. Kongelige Veterinaer-og Landbohojskole Aarsskrift. 1941: 44-58. Gray, A. 1842. Notes of a botanical excursion to the mountains of North Carolina. Amer. Jour. Sci. 42: 1-49. Hall, J.W. 1952. The comparative anatomy and phylogeny of the Betulaceae. Bot. Gaz. 113: 235-270. Hammen, T. Van der. 1972. Historia de la vegetacion y el medio am- biente del norte Sudamericano. Memorias de Symposia, I Congreso Latinoamericano V Mexicano de Botanica. Mexico, D.F.‘ Pp. 119- 134. Hanover, J.W. and R.J. Hoff. 1966. A comparison of phenolic con- stituents of Pinus monticola resistant and susceptible to Cronartium ribicola. Physiologia Plantarum 19: 554-562. Hanover, J.W. and R.C. Wilkinson. 1970. Chemical evidence for introgressive hybridization in Picea. Silvae Genetics 19: 17‘220 Hara, H. 1948. Notes on the mountain area of Central America. Jour. Jap. Bot. 22: 6. Hardin, J.W. 1957. A revision of the American Hippocastanaceae. Brittonia 9: 145-171; 173~195. 489 Harshburger, J.W. 1903. An ecological study of mountainous North Carolina. Bot. Gaz. 36: 241-258; 368-383. Hartig, T. 1851. Vollstandige Naturgeschichte der Forstlichen Kulturpflanzen Deutschlands. Berlin. 584 pp. Hawker, L.E. and J. Fraymouth. 1951. A re-investigation of the root- nodules of species of Elaeagnus, Hippophae, Alnus, and Myrica, with reference to the morphology and life histories of the causative organisms. Jour. Gen. Microbiol. 5: 369-386. Hegnauer, R. 1964. Chemotaxonomie der Pflanzen. Basel. Vol. 3. Heilman, P.E. 1966. Changes in the distribution and availability of nitrogen with forest succession on north slopes in interior Alaska. Ecol. 47: 825—831. Heiser, C.B., J. Soria, and D.L. Burton. 1965. A numerical taxonomic study of Solanum species and hybrids. Amer. Natrl. 99: 471-488. Hemsley, W.B. 1882. Botany. EE_F.D. Goodman and O. Salvin (eds.). Biologia Centrali-Americana. London. Vol. 55. Heusser, C.J. 1960. Late Pleistocene Events of North Pacific America. New York. 308 pp. Heywood, V.H. 1966. Phytochemistry and taxonomy. In T. Swain (ed.). Comparative Phytochemistry. New York. Pp. 1-20. Hickey, L.J. 1973. Classification of the architecture of dicoty- ledonous leaves. Amer. Jour. Bot. 60: 17-33. Hill, J. 1756-1757. The British Herbal. London. 536 pp. Hillis, W.E. 1962. The distribution and formation of polyphenols within the tree. IE_W.E. Hillis (ed.). Wood Extractions. New York. Pp. 59-131. Hitchcock, C.L., A. Cronquist, M. Ownbey, and J.W. Thompson. 1964. Vascular Plants of the Pacific Northwest. Seattle. Vol. 2. Hjelmqvist, H. 1948. Studies on the floral morphology and phylogeny of the amentiferae. Bot. Notiser. Suppl. 2(1): 1-171. Hoar, C.S. 1916. The anatomy and phylogenetic position of the Betulaceae. Amer. Jour. Bot. 3: 415-435. Hooker, W.J. 1838. Flora Boreali-Americana. London. Vol. 2. House, H.H. 1924. Annotated list of the ferns and flowering plants of New York State. New York State Mus. Bull. 254: 1-759. 490 Hsiao, J. 1973. A numerical taxonomic study of the genus Platanus based on morphological and phenolic characters. Amer. Jour.‘ Bot. 60: 687-694. Hultéh, E. 1937a. Flora of the Aleutian Islands. Stockholm. 397 pp. . 1937b. Outline of the History of Arctic and Boreal Biota During the Quaternary Period; their Evolution During and After the Glacial Period as Indicated by the Equiformal Progressive Areas of Present Plant Species. Stockholm. 168 pp. . 1941-1950. Flora of Alaska and Yukon. Lunds Universi- tets Arsskrift, N.F., 37-45: 1-1341. . 1963. Phytogeographical connections of the North Atlan- tic. EE.A° L6ve and D. L6ve (eds.). North Atlantic Biota and their History. New York. Pp. 45-72. . 1967. Comments on the flora of Alaska and Yukon. Arkiv f6r Botanik, ser. 2, 7(1): 1-147. . 1971. The circumpolar plants. II. Dicotyledons. Kunglia Svenska Vetenskapsakademiens Handlingar, ser. 4, 13(1): 1-463. Humboldt, F.H.A. von, A.J. Bonpland, and C.S. Kunth. 1817. Nova Gen- eris et Species Plantarum. Paris. Vol. 2. Hurd, R.M. 1971. Annual tree-litter production by successional for- est stands, Juneau, Alaska. Ecol.:52: 881-884. Hylander, N. 1957. On cut-leaved and small-leaved forms of Alnus glutinosa and A, incana. Svensk Botanisk Tidskrift 51: 437,453, Iltis, H.H. 1959. Studies in the Capparidaceae. VI. Cleome sect. Physostemon: taxonomy, geography, and evolution. Brittonia 11: 123-162. . 1966. The western element in the eastern North Ameri— can flora and its phytogeographical implications. Amer. Jour. Bot. 53: 635. Ives, J.D. 1963. Field problems in determining the maximum extent of Pleistocene glaciation along the eastern Canadian seaboard-- a geographer's point of View. IE_A. Lave and D. L6ve (eds.). North Atlantic Biota and their History. New York. Pp. 337-354. Jackson, R.C. and T.J. Crovello. 1971. A comparison of numerical and biosystematic studies in Haplopappus. Brittonia 23: 54-70. Jaretzky, R. 1930. Zur Zytologie der Fagales. Planta 10: 120-137. 491 Johannsen, D.A. 1940. Plant Microtechnique. New York. 523 pp. Johnson, F.D. 1968a. Disjunct populations of red alder in Idaho. IE_J.M. Trappe, J.F. Franklin, R.F. Tarrant, and C.M. Hansen (eds.). Biology of Alder. Portland. Pp. 1-8. . 1968b. Taxonomy and distribution of northwestern alders. In J.M. Trappe, J.F. Franklin, R.F. Tarrant, and C.M. Hansen .TEds.). Biology of Alder. Portland. Pp. 9-22. Judd, W.A. 1974. The Systematics of Lyonia (Ericaceae) in North America. M.S. Thesis, Michigan State University. Katz, M.W. and A.M. Torres. 1965. Numerical analyses of ceSpitose zinnias. Brittonia 17: 335-349. Kelley, A.P. 1950. Mycotrophy in Plants. Waltham, Mass. 223 pp. Kellogg, A. 1869. Report of Albert Kellogg, M.D., Aid United States Coast Survey, to George Davidson, Assistant United States Coast Survey on the Botany of Alaska. IE_B. Peirce. Report of the l Superintendent of the United States Coast Survey showing the ' Progress of the Survey During the Year 1867, 40th Congress, 2nd session. House Executive Document no. 275. Pp. 318-324. Kjellman, F.R. 1883. Vega. EXp. Vet. Ivktt. 2: 52. Kledka, A. and V. Vukolov. 1935. Srounévaci studie o mykorrhize dfevin (uEitkovych, okrasnych a ovacnych). Sbornik Ceskoslav Akad. Zemed. 10: 443-457. Kluge, A.G. and J.S. Farris. 1969. Quantitative phyletics and the evolution of anurans. Syst. Zool. 18: 1-32. Koch, K.H.E. 1869-1873. Dendrologie, ed. 2. Erlangen. 3 vols. Koch, W.D.J. 1837. Synopsis Florae Germanicae et Helveticae. Francofurti ad Moenum. Pt. 2. Kodama, A. 1967. Karyological studies on root nodules of three species of Alnus. Bot. Mag. (Tokyo) 80: 230-232. Komorov, V.L. 1929. F1. Penins. Kamtsch. . 1936. Flora S.S.S.R. Moskva-Leningrad. Vol. 5. Koshy, T.K., W.F. Grant, and W.H. Brittain. 1972. Numerical chemo- taxonomy of the Betula caerulea complex. Symp. Biol. Hung. 12: 201-211. Kryshtofovich, A.N. 1921. Evolution of the Tertiary flora in Asia. New Phytol. 28: 303-312. 492 Kuntze, O. 1891. Revisio Generum Plantarum. Wurzburg. Pp. 638-640. Lamarck, J.B.A.P.M. de and A.P. de Candolle. 1805. Flore Francaise, ed. 3. Paris. Vol. 3. La Motte, R.S. 1952. Catalogue of the Cenozoic Plants of North America Through 1950. New York. 381 pp. Lange, J. 1887. Conspectus florae Groenlandicae, pars secunda. Med. om Gr6nl. 3: 280. Lanjouw, J. and F.A. Stafleu. 1964. Index Herbariorum. Part I. The Herbaria of the World, ed. 5. Utrecht. 251 pp. Lawrence, D.B. 1958. Glaciers and vegetation in southwestern Alaska. Amer. Scient. 46: 89-122. Ledebour, K.F. von. 1850. Flora Rossica. Stuttgartiae. Vol. 3. Leopold, E.B. and H.D. MacGinitie. 1972. Development and affinities of Tertiary floras in the Rocky Mountains. 13 A. Graham (ed.). Floristics and Paleofloristics of Asia and Eastern North America. Amsterdam. Pp. 147—200 Lepage, E. 1950. Une variété nouvelle de l'Alnus crispa (Ait.) Pursh. Nat. Canad. 77: 44-46. Li, H.L. 1971. Floristic Relationships between Eastern Asia and Eastern North America. Philadelphia. 429 pp. . 1972. Eastern Asia-eastern North America species pairs in wide-ranging genera. IE_A. Graham (ed.). Floristics and Paleofloristics of Asia and Eastern North America. Amsterdam. Pp. 64-78. Linnaeus, C. 1753. Species Plantarum. Holmiae. Vol. 2. Linnaeus, C. 1871. Supplementum Plantarum Systematis Vegetabilum, ed. decimae tertiae. Brunsvigas. 467 pp. Livingstone, D.A. 1955. Some pollen profiles from arctic Alaska. Ecol. 36: 587—600. Ljunger, A. 1959. A1-och alféradling. Skogen 46: 115-117. Loddiges, C. 1826. Botanical Cabinet. London. Vol. 12. London, J.C. 1838. Arboretum et Fruticetum Britannicum. London. 8 vols. L6ve, A. 1959. Origin of the arctic flora. Publ. McGill Univ. Mus. 1: 82-95. 493 . 1967. The evolutionary significance of disjunctions. Taxon 16: 324-333. and D. L6ve. 1965. Taxonomic remarks on some American alpine plants. Univ. Colo. Stds. Ser. Biol. 17: 1-43. and D. L6ve. 1967. The origin of the North Atlantic flora. Aquilo Ser. Bot. 6: 52-66. L6ve, D. 1959. The postglacial development of the flora of Manitoba: a discussion. Can. Jour. Bot. 37: 547-585. Ludwig, C.F. 1783. Die Neuere Wilde Baumzucht in einem Alphabetis- chen und Systematischen Verzeichnisse Aufgestellt. Leipzig. 70 pp. Macbride, J.F. 1936. Flora of Peru, pt. 2. Field Mus. Natrl. Hist. Marshall, H. 1785. Arbustum Americanum. Philadelphia. 174 pp. Martinez, M. 1959. Plantas Medicinales de Mexico. Mexico, D.F. 657 pp. Masamune, G. 1934. Plantae vasculares a Prof. I. Namikawa anno 1920 in Kamtschatka lectae, auctore Yushun Kudo. Jour. Jap. Bot. 10: 487~526. Masui, K. A study of the ectotrophic mycorrhiza of Alnus. Mem. C011. Sci. Kyoto Imp. Univ. ser. B, 2: 189-209. Matuda, E. 1953. Plantas Asiaticas en Mexico. Mem. Congr. Cient. Mex. IV Centenario Univ. Mex. 6: 234-248. Mayr, E. 1947. Ecological factors in speciation. Evolution 1: 263~288. . 1965. Numerical phenetics and taxonomic theory. Syst. Zool. 14: 73~97. McVaugh, R. 1952. Suggested phylogeny of Prunus serotina and other wide-ranging phylads in North America. Brittonia 7: 317-346. McVean, D.N. 1953a. Alnus Mill. lE_Biologica1 flora of the British Isles. Jour. Ecol. 41: 447-466. . 1953b. Regional variation of Alnus glutinosa (L.) Gaertn. in Britain. Watsonia 3: 26-32. . 1955. Ecology of Alnus glutinosa (L.) Gaertn. 11. Seed distribution and germination. Jour. Ecol. 43: 61-71. .- ..—...- “'53.: Ill. l..l. .. I Jili- 494 . 1956a. Ecology of Alnus_g1utinosa (L.) Gaertn. III. Seedling establishment. Jour. Ecol. 44: 195-218. . 1956b. Ecology of Alnus_g1utinosa (L.) Gaertn. IV. Root system. Jour. Ecol. 44: 219-225. Metcalfe, C.R. and L. Chalk. 1950. Anatomy of the Dicotyledons. Oxford. Vol. 2. Meyer, C.A. 1831. Verzeich. Pfl. Kaukas., p. 43. I Michaux, A. 1803. Flora Boreali-Americana. Paris. 2 vols. Michaux, F.A. 1813. Historie des Arbores Forestiers de'l Amerique Septentrionale. Paris. Vol. 3. . 1859. The North American Sylva. Philadelphia. Vol. 2. Miller, P. 1754. The Gardner's Dictionery, Abridged, ed. 4. London. 1759. The Gardner's Dictionery, ed. 7. London. . 1768. The Gardner's Dictionery, ed. 8. London. Mirbel, C.F.B. de. 1827. Description de quelques especes nouvelles de la famille des amentacees. Mem. Mus. Nat. Hist. Nat. 14: 462-474. Mitchell, W.W. 1968. On the ecology of the Sitka alder in the sub- alpine zone of south-central Alaska. IE_J.M. Trappe, J.F. Franklin, R.F. Tarrant, and C.M. Hansen (eds.). Biology of Alder. Portland. Pp. 45-56. Moench, K. 1794. Methodus Plantas Horti Botanici et Agri Marburgen- sis. 780 pp. Morishima, H. 1969. Phenetic similarity and phylogenetic relation- ships among strains of Oryza perennis, estimated by methods of numerical taxonomy. Evolution 23: 429-443. and H.l. Oka. 1960. The pattern of interspecific vari- ation in the genus Cryza: its quantitative representation by statistical methods. Evolution 14: 153-165. Morse, L.E. 1968. Construction of identification keys by computer. Amer. Jour. Bot. 55: 737. . 1971. Specimen identification and key construction with time-sharing computers. Taxon 20: 269~282. . 1974a. Computer-assisted storage and retrieval of the data of taxonomy and systematics. Taxon 23: 29-43. 495 . 1974b. Computer programs for specimen identification, key construction, and description printing using taxonomic data matrices. Publ. Mus. Mich. St. Univ. Biol. Ser. 5: 1-128. , J.A. Peters, and P.B. Hamel. 1971. A general data format for summarizing taxonomic information. Bioscience 21: 174-181. Muller, C.H. 1940. New and otherwise noteworthy plants of the Southwest. Madrofio 5: 152-158. Munz, P.A. and I.M. Johnston. 1925. Miscellaneous notes on plants of southern California. IV. Bull. Torr. Bot. Cl. 52: 221-228. Murai, S. 1963. Phytotaxonomical and geobotanical studies on so— called genus Alnus in Japan (II). Comparative studies on all species, including shrubby species. Bull. Gov. For. Expt. Sta. Jap. 154: 21-72. . 1964. Phytotaxonomical and geobotanical studies on gen. Alnus in Japan (III). Taxonomy of whole world species and distribution of each sect. Bull. Gov. For. Expt. Sta. Jap. 171: 1-107. . 1968. Relationships of allied species between north- western USA and Japan on the genus Alnus. IE_J.M. Trappe, J.F. Franklin, R.F. Tarrant, and G.M. Hansen (eds.). Biology of Alder. Portland. Pp. 23-36. Neal, J.L., Jr., J.M. Trappe, K.C. Lu, and W.B. Bollen. 1968. Some ectotrophic mycorrhizae of Alnus rubra. IE_J.M. Trappe, J.F. Franklin, R.F. Tarrant, and G.M. Hansen (eds.). Biology of Alder. Portland. Pp. 179-184. Newton, M., B.A. E1 Hassan, and J. Javitkovski. 1968. Role of red alder in western Oregon forest succession. IE_J.M. Trappe, J.F. Franklin, R.F. Tarrant, and G.M. Hansen (eds.). Biology of Alder. Portland. Pp. 73-84. Nuttall, T. 1842. The North American Sylva. Philadelphia. Vol. 1. Nyman, C.F. 1881. Conspectus Florae Europaeae. Orebro Sueciae. Opiz, P.M. 1839. Oekonomische Neuigkeiten und Verhandlungen 1839: 521-527. . 1852. Seznam Rostlin Kveteny Ceske. Prague. Vol. 10. Payne, R.C. and D.E. Fairbrothers. 1973. Disc electrophoretic study of pollen proteins from natural populations of Betula populifolia in New Jersey. Amer. Jour. Bot. 60: 182-189. 496 Persoon, C.H. 1805-1807. Synopsis Plantarum. Paris. 2 vols. Petermann, W.L. 1849. Deutschlands Flora mit Abbildungen. Leipzig. 668 pp. Pettet, A. 1965. Studies on British pansies. III. A factorial analysis of morphological variation. Watsonia 6: 141-163. Petzold, E. and G. Kirchner. 1864. Arboretum Muscoviense. Gotha. 830 pp. Philippi, R.A. 1895. Plantas nuevas Chilenas de las familias que corresponden a1 Tomo IV de la obre de Gay. Anal. Univ. Chile 91: 514. Pomeroy, K.B. and D. Dixon. 1966. These are the champs. Amer. Forests 72(5): 14-35. Porsild, A.E. 1939. Contributions to the flora of Alaska. Rhodora 41: 199-254; 262-301; 441-483. Poucques M.L. de. 1949. Etudes caryologiques sur les Fagales. 1. Le genre Alnus. Bull. Mus. Hist. Nat. Paris 21(2): 147-152. Pouzar, Z. 1964. Nomenclatural remarks on some generic names of Phanerogams validly published by Filip Maximillian Opiz. Preslia 36: 337-342. ‘ Pursh, F. 1814. Flora Americae Septentrionalis. London. 2 vols. Raup, H.M. 1936. Phytogeographic studies in the Athabaska-Great Slave Lake region. I. Catalogue of vascular plants. Jour. Arn. Arb. 17: 241-315. . 1947. Some natural floristic areas in boreal America. Ecol. Mon. 17: 221-234. Raven, P.H. 1972. Plant species disjunctions: a summary. Ann. Mo. Bot. Gard. 59: 234-246. and D.I. Axelrod. 1972. Plate tectonics and Australasian paleobotany. Science 176: 1379-1386. Ray, J. 1682. Methodus Plantarum. Amsterdam. 166 pp. Record, S.J. and R.W. Hess. 1943. Timbers of the New World. New Haven. 640 pp. Regel, E. 1861. Monographische Bearbeitung der Betulaceen. Mem. Soc. Imp. Nat. Mosc. 13(2): 59-187. 497 . 1865. Bemerkungen uber die Gattungen Betula und Alnus nebst Beschreibung einiger neuer Arten. Bull. Soc. Imp. Nat. Mosc. 38(3): 388-434. . 1868. Betulaceae. EE.A'P° de Candolle, A. de Candolle, and C. de Candolle. Prodromous Systematis Naturalis Regni Vegetabilis. Paris. Vol. 16, pt. 2. Rehder, A. 1927. Manual of Cultivated Trees and Shrubs. New York. 930 pp. . Bibliography of Cultivated Trees and Shrubs. Jamaica Plain, Mass. 825 pp. Reichardt, H.W. 1854. Verzeichniss allen von Herrn J. Ch. Neumann in Bohmen gesammelten Pflanzen. Verhandl. Zool. Bot. Ver. Wein 4: 253-284. Reiners, W.A., I.A. Worley, and D.B. Lawrence. 1971. Plant diversity in a chronosequence at Glacier Bay, Alaska. Ecol. 52: 55-69. Rhoades, A.M., W.P. Bemis, T.W. Whitaker, and S.G. Carmen. 1968. A numerical taxonomic study of Cucurbita. Brittonia 20: 251- 266. Richardson, J. 1823. Bot. Appendix. 12_J. Franklin. Narrative of a Journey to the Shores of the Polar Sea of North America, 1819-1822. London. Pp. 729-768, no. 374. Ridgway, J.E. and J.J. Skvarla. 1969. Scanning electron microscopy as an aid to pollen taxonomy. Ann. Mo. Bot. Gard. 56: 121-124. Rodriguez-Barrueco, C. and G. Bond. 1968. Nodule endophytes in the genus Alnus. IE_J.M. Trappe, J.F. Franklin, R.F. Tarrant, and G.M. Hansen (eds.). Biology of Alder. Portland. Pp. 185-192. Ruprecht, F.J. 1845. Beitrage zur Pflanzenkunde des Russichen Reiches. St. Petersburgh. Vol. 2. Rydberg, P.A. 1897. Rarities from Montana. Bull. Torr. Bot. Cl. 24: 188-192. Rzedowski, J. 1965. Relaciones geograficas y posibles origenes de la flora de Mexico. Boletin Soc. Bot. Mex. 29: 121-177. Saito, A. 1970. Electrophoretic comparison of soluble pollen proteins in Alnus in relation to inter- and intraSpecific identification. Jour. Jap. For. Soc. 52: 291-295. Sargent, C.S. 1896, 1902. The Silva of North America. Boston. Vols. 9 and 14. 498 Schalin, I. 1968. Germination analysis of gray alder (Alnus incana) and black alder (Alnus glutinosa) seeds. 22. J.M. Trappe, J.F. Franklin, R.F. Tarrant, and G.M. Hansen (eds.). Biology of Alder. Portland. Pp. 107-113. Schlechtendal, D.F.L. von. 1832. De plantis Mexicanis a G. Schiede M. Dre. collectis nuntium adfert D.F.L. de Schlechtendal. Lin- naea 7: 136-144. Schrank, F. von Paula von. 1789. Baiersche Flora. Mfinchen. 2 vols. Sharp, A.J. 1953a. Generic correlations in the flora of Mexico and eastern Asia. Jour. Tenn. Acad. Sci. 28: 188. . 1953b. Notes on the flora of Mexico: world distribution of the woody dicotyledonous families and the origin of the modern vegetation. Jour. Ecol. 41: 374-380. . 1972. The possible significance of some exotic distri- butions of plants occurring in Japan and/or North America. 12_A. Graham (ed.). Floristics and Paleofloristics of Asia and Eastern North America. Amsterdam. Pp. 61-64. Scora, R.W. 1966. The evolution of the genus Monarda (Labiatae). Evolution 20: 185-190. Smith, C.R. and R.K. Koehn. 1971. Phenetic and cladistic studies of biochemical and morphological characteristics of Catostomus. Syst. 2001. 20: 282-297. Smith H.H. 1923. Ethnobotany of the Menomini Indians. Bull. Publ. Mus. City of Milwaukee 4: 1-174. . 1928. Ethnobotany of the Meskwaki Indians.. Bull. Publ. Mus. City of Milwaukee 4: 175-326. . 1932. Ethnobotany of the Ojibwe Indians. Bull. Publ. Mus. City of Milwaukee 4: 327-525. . 1933. Ethnobotany of the forest Potawatomi Indians. Bull. Publ. Mus. City of Milwaukee 7: 1-230. Smith, I. '(ed.). 1960. Chromatographic and electrOphoretic tech- niques. Vol. 1. Chromatography, ed. 2. .New York. 617 pp. Smith, J.H.G. 1968. Growth and yield of red alder in British Columbia. IE_J.M. Trappe, J.F. Franklin, R.F. Tarrant, and G.M. Hansen (eds.). Biology of Alder. Portland. Pp. 273-286. Sneath, P.H.A. and R.R. Sokal. 1973. Numerical Taxonomy, the Prin— ciples and Practice of Numerical Classification. San Francisco. 573 pp. 499 Sokal, R.R. 1961. Distance as a measure of taxonomic simi- larity. Syst. Zool. 10: 70—79. . 1962. Typology and empiricism in taxonomy. Jour. _ATheoret. Biol. 3: 230-267. and C.D. Michner. 1958. A statistical method for evaluating systematic relationships. Univ. Kans. Sci. Bull. 38: 1409-1438. and P.H.A. Sneath. 1963. Principles of Numerical Taxonomy. San Francisco. 359 pp. Solbrig, O.T. 1970. The phylogeny of Gutierrezia: an eclectic approach. Brittonia 22: 217-229. ' Solereder, H. 1908. Systematic Anatomy of the Dicotyledons. Oxford. Vol. 2. Soria, J. and C.B. Heiser. 1961. A statistical study of relation- ships of certain Species of the Solanum nigrum complex. Econ. Bot. 15: 245-255. Spach, A.E. 1841. Revisio Betulacearum. Ann. Sci. Nat. ser. 2, Sprengel, C. 1826. Caroli Linnaei,...Systema Vegetabilium, ed. 16. Gottingae. Vol. 3. Stafleu, F.A., C.E.B. Bonner, R. McVaugh, R.D. Meikle, R.C. Rollins, R. Ross, J.M. Schopf, G.M. Schulte, R. de Vilmorin, and E.G. Voss (eds.). International Code of Botanical Nomenclature. Utrecht. 426 pp. Standley, P.C. 1920. Betulaceae. £2_Trees and Shrubs of Mexico, Part 1. Contr. U.S. Nat. Herb. 23(1): 167-168. and J.A. Steyermark. 1952. Betulaceae. IE_Flora of Guatemala. Fieldiana: Botany 24(3): 359-364. Starker, T.J. 1939. A new alder. Jour. For. 37: 415-416. Steele, F.L. 1961. Introgression of Alnus serrulata and Alnus rugosa. Rhodora 63: 297-304. Stern, K.R. 1961. Revision of Dicentra (Fumariaceae). Brittonia 12: 1-570 . 1962. The use of pollen morphology in the taxonomy of Dicentra. Amer. Jour. Bot. 49: 362-368. 500 Takhtajan, A. 1969. Flowering Plants: Origin and Dispersal. Edin- burgh. 310 pp. Tarrant, R.F. 1968. Some effects of alder on the forest environ- ment. .£2.J°M° Trappe, J.F. Franklin, R.F. Tarrant, and G.M. Hansen (eds.). Biology of Alder. Portland. P. 193. Taylor, R.J. 1971. Intraindividual phenolic variation in the genus Tiarella (Saxifragaceae): its genetic regulation and application to systematics. Taxon 20: 467-472. 1966. Taximetrics as applied to the genus Lithgphragma Taylor, R.L. Amer. Jour. Bot. 53: 372-377. (Saxifragaceae). Thorarinsson, S. 1963. The Svinafell layers plant-bearing inter- glacial sediments in Oraefi, southeast Iceland. IE_A. Lfive and D. L6ve (eds.). North Atlantic Biota and their History. New York. Pp. 337-389. Comparative anatomy of the Moraceae and their Tippo, O. 1938. 100: 1-99. presumed allies. Bot. Gaz. Tisdale, E.W., M.A. Fosberg, and G.E. Poulton. 1966. Vegetation and soil development on a recently glaciated area near Mount Robson, British Columbia. Ecol. 47: 517-523. Torrey, J. 1859. Botany of the boundary. IE_W.H. Emory. Report on the United States and Mexico boundary survey made under the direction of the Secretary of the Interior. 34th Congress, lst session. Senate Executive Document no. 135. Vol. 2, pp. 861-863. Tournefort, J.P. 1700. Institutiones rei herbariae, editio altera. Parisiis. Vol. 1. Mycorrhizal hosts and distribution of Ceno- Trappe, J.M. 1964. Lloydia 27: 100-106. coccum grandiforme. Tucker, J.M. 1974. Patterns of parallel evolution of leaf form in New World oaks. Taxon 23: 129-154. Tuckerman, E. 1843. Observations on some interesting plants of New England. Amer. Jour. Sci. ser. 2, 45: 27-49. Turner, B.L. 1969. Chemosystematics: recent developments. Taxon 18: 134-151. . 1972. Chemosystematic data: their use in the study of Ann. Mo. Bot. Gard. 59: 152-164. disjunctions. Turrill, W.B. 1962. Alnus viridis subsp. crisEa. Curtis's Bot. Mag. 173: 382. ~ ".... ‘Q—r. ah..- - - 1:1 I 11.1. 501 Ugolini, F.C. 1968. Soil development and alder invasion in a recently deglaciated area of Glacier Bay, Alaska. lfl.J°M' Trappe, J.F. Franklin, R.F. Tarrant, and G.M. Hansen (eds.). Biology of Alder. Portland. Pp. 115-140. Van Alstine, R.E. 1969. Geology and mineral deposits of the Poncho Springs NE quadrangle, Chaffee County, Colorado. U.S. Geol. Surv. Prof. Paper 626: 1-52. Vavilov, N.I. 1951. The Origin, Variation, Immunity, and Breeding of Cultivated Plants. New York. 366 pp. Veldman, D.J. 1967. Fortran Programming for the Behavioral Sciences. New York. 406 pp. Villars, D. 1788. Historie des Plantes de Dauphine. Grenoble. Vol. 3. Vines, R.A. 1960. Trees, Shrubs, and Woody Vines of the Southwest. Austin. 1104 pp. Wagner, W.H., Jr. 1961. Problems in the classification of ferns. Recent Advances in Botany 1: 841-844. . 1962. A graphic method for expressing relationships based upon group correlations of indexes of divergence. 12. L. Benson. Plant Taxonomy: Methods and Principles. New York. Pp. 415-416. . 1969. The construction of a classification. 12. Systematic Biology. Washington, D.C., National Academy of Science. Pp. 67-90. Wahlenberg, G. 1812. Flora Lapponica. Berolini. 550 pp. Wanscher, J.H. 1934. The basic chromosome number of the higher plants. New Phytol. 33: 101-126. Watson, 8. 1880. Botany of California. Boston. Vol. 2. Weimarck, G. 1972. On "numerical chemotaxonomy." Taxon 21: 615-619. Weston, R. 1770. Botanicus Universalis et Hortulanus. London. Vol. 1. Wetzel, G. 1927. Chromosomenzahlen bei den Fagales. Berichte der Deutsch. Bot. Gesells. 45: 251-252. . 1928. Chromosomenstudien bei den Fagales. Berichte der Deutsch. Bot. Gesells. 46: 212-214. 502 . 1929. Chromosomenstudien bei den Fagales. Bot. Archiv 25: 257-283. Wherry, E.T. 1960. Intermediate occurrences of Alnus criSpa. Castanea 25: 135. Whiffin, T. and M.W. Bierner. 1972. A quick method for computing Wagner Trees. Taxon 21: 83-90. Whittaker, R.H. and W.A. Niering. 1965. Vegetation of the Santa Catalina Mountains, Arizona: a gradient analysis of the south slope. Ecol. 46: 429-452. Wilkinson, R.C., J.W. Hanover, J.W. Wright, and R.H. Flake. 1971. Genetic variation in the monoterpene composition of white spruce. Forest Sci. 17: 83-90. Willdenow, K.L. 1805. Caroli a Linne Species Plantarum, ed. 4. Berolini. Vol. 4, pt. 1. Williamson, R.L. 1968. Productivity of red alder in western Oregon and Washington. In J.M. Trappe, J.F. Franklin, R.F. Tarrant, and G.M. Hansenfzeds.). Biology of Alder. Portland. Pp. 287-292. Winkler, H. 1904. Betulaceae. Das Pflanzenreich 19(4.61): 1-149. Wodehouse, R.P. 1935. Pollen Grains. New York. 574 pp. Wolfe, J. 1969. Neogene floristic and vegetational history of the Pacific Northwest. Madrofio 20: 83-110. ,. 1972. An interpretation of Alaskan Tertiary floras. EE_A. Graham (ed.). Floristics and Paleofloristics of Asia and Eastern North America. Amsterdam. Pp. 201-233. . 1973. Fossil forms of amentiferae. Brittonia 25: 334“355 o Wood, A. 1855. Class-book of Botany. Boston. Woodworth, R.H. 1929. Cytological studies in the Betulaceae. II. Corylus and Alnus. Bot. Gaz. 88: 383-399. . 1930. Cytological studies in the Betulaceae. III. Parthenogenesis and polyembryony in Alnus rugosa. Bot. Gaz. 89: 402-409. . 1931. Polyploidy in the Betulaceae. Jour. Arn. 503 Worthington, N.P, R.H. Ruth, and E.B. Matson. 1962. Red Alder, its Management and Utilization. U.S. Dept. of Agr. Misc. Publ. 881. Washington. 44 pp. Zumaglini, A.M. 1849. Flora Pedemontana. Vol. 1. .....- -__,__:. ...~..~.--—-—--' "1mmm