STUDY OF GEOGRAPHIC VARIATION IN EUROPEAN LARCH by JOHN B. GENYS A THESIS Submitted to the School for Advanced Graduate Studies Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY 1960 Department of Forestry ,tutt , ProQuest Number: 10008627 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. uest ProQuest 10008627 Published by ProQuest LLC (2016). Copyright of the Dissertation is held by the Author. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code Microform Edition © ProQuest LLC. ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 4 8 10 6 - 1346 ACKNOWLEDGMENT S I wish to extend sincere thanks to Dr. Jonathan W. Wright who, major professor, supervised the conduct of this study. as Appreciation is also extended to other members of my Committee, Drs. William D. Baten, Allen S. Fox, John E. Grafius and Donald P. White, comments; to Dr. Terrill D. Stevens, Head of the Forestry Department of Michigan State University, guidance; for their valuable and other members of the Department for their and to the University for granting the research assistantship which allowed me to pursue the study. I am deeply indebted to the New Hampshire Forestry and Recreation Commission for granting the Fellowship under the Fox Trust Fund for For­ est Research which made this study possible. Sincere appreciation is extended to Dr. Henry I. Baldwin who established the study and helped in various phases of the measurement work. The same appreciation is extended to the New York Conservation Department for supplying their plantations for this study and particularly to E. J. Eliason for his cooperation. SUMMARY European larch consists of two varieties, Larix decidua var. decidua Mill, Larsen. and Larix decidua var. polonica (Racib.) Ostenf. & It belongs to the genus Larix Mill, which includes 13 other species or varieties. It is a coniferous tree with deciduous leaves growing naturally in central Europe, mainly in the Alps and the Carpathian Mountains and adjacent areas. Its range has been considera­ bly increased by planting. European larch was introduced to the United States at the middle of the nineteenth century and at present individual trees or small stands are found throughout the northern parts of this country east of the Mississippi. It has some economic importance only in New York and Pennsylvania, where it was planted most extensively. Rapid growth is the most important characteristic of European larch which attracted to use it in plantings. It produces merchantable yields at early ages and its durable wood is valuable for various general use. First provenance studies of European larch were undertaken at the end of the nineteenth century. These and later investigations in Europe revealed that seed origins from different parts of its range considera­ bly varied, not only in height growth, but also in stem qualities and resistance to disease. This variability was observed also in a few American studies of a limited number of origins. This study was based on the first large American test plantings of European larch which were established at the Fox State Forest, Hills­ boro, New Hampshire, York. and at the State Forest near Barkersville, New The test is known as the ,fIUFRO Larch '44" experiment because the seed lots were received from the International Union of Forest Research Organizations. It included 31 provenances of European larch collected in various parts in the natural range and a few origins from planted stands. Nursery plantings were established in 1946 and field plantings in 1958. Provenances were arranged in both experiments in unreplicated plots, except a few second-replicates in New Hampshire. Sources were arranged in rows of varying number of trees per plot in New York and in rectangular plots with about 35 trees per plot in New Hampshire. The spacing was 6 x 6 feet in New York and 5 x 5 feet or similar in New Hampshire. The writer measured and scored these American IUFRO larch planta­ tions in the summer of 1958 when the trees were 12-years old. Data on eight characteristics from these and earlier studies were the basis of this work. The general pattern of variation in European larch waso determined by using the New Hampshire data as more complete and reliable. Analyses were based on the provenance means grouped by 10 geographic and altitudinal regions. Differences between individual provenances and the within-region variations were studied on the basis of combined New Hampshire and New York data, in which case each plantation was con­ sidered as a randomized block. Results are as follows: 1. Altitudinal differences were found in seed weights. Prove­ nances from higher elevations had relatively heavier seeds than those from lower altitudes. Seed weights appeared as a poor factor in determining future height growth, because correlation between 4-year heights and seed weights was negative. Germination in European larch was completed 90 percent in 15 days. 2. Geographic variation in European larch in eight characteris­ tics studied was largely systematic. Provenances from the same geographic or altitudinal regions showed similar characteristics. Altitudinal differences were found among provenances from high and low elevations in Central and Western Austria. A break in genetical characteristics appeared between provenances from the Polish-Sudeten region and those from Slovakia. Provenances from Poland, Sudetenland and Slovakia had significantly greater 12-year heights than provenances from other regions. elevation provenances from Austria, Higher Italy and Switzerland grew slowest. The pattern of variation in diameter growth was similar to that in height growth except that Slovakian provenances did not differ signifi­ cantly in diameter growth from other provenances with smaller diameters. Analyses of variance showed that the general pattern of variation in height performance could be determined on the basis of 4-year height data or on the basis of the means of 4 largest trees measured per plot. However, origins. these data gave less efficient statistical differences between Analyses of stem form were based on number of crooks in the lower stem and the number of trees with small crooks in the upper stem. The straightest stems were found in the provenances at higher elevations in Austria, Italy and Switzerland. An extremely high number of crooks were found in the provenances from lower elevations in Western Austria. Among the rapidly growing provenances the straightest stems were found in those from Slovakia. Some provenances in the plantation had many trees with J-shaped bases (basal swe^J) and some were entirely out of vertical (lean). Analysis of variance showed these characteristics were more common in the fast growing provenances from Poland and Sudetenland. However this rule did not apply to the rapidly growing provenances from Slo­ vakia, and a separate study of individual trees revealed that lean was not related to tree height. The Austrian provenances from higher altitudes had less lean or basal sweep than from lower altitudes. In 1958 the IUFRO larch plantation in New Hampshire was subjected to late snow damage and attack by the woolly larch aphid (Chermes strobilobius Kalt.). The most injury by these agents was caused in the Sudeten, Slovakian and Eastern Austrian provenances and the least in provenances from higher altitudes. nances, 3. those from Poland were only moderately attacked by the aphids* From these characteristics it appears that Polish provenances, best in cellulose production, land. Among the rapidly growing prove­ are related to provenances from Sudeten­ Slovakian provenances differed; from the above origins in several characteristics and were best in quality of wood production. Origins from the Alps grew slower, differences. in general, and showed altitudinal Iiow-elevation provenances were somewhat faster growing, had more stem deformities and were more attacked by aphid than the high-elevation provenances. 4. No origins from planted stands grew faster than the best provenances of European larch from ; native stands. The provenance (IUFRO-53) from Scotland had fewer trees attacked by the woolly larch aphid than any other provenance of European larch; (IUFRO-28) was 100 percent free from aphid. and Japanese larch Some hybrids (L. eurolepis Henry) were observed within the origin IUFRO-55. They were relatively resistant to the aphid and had greater heights than other trees in the plot. 5. Twelve-year heights in American IUFRO larch plantations and heights of similar provenances planted in Great Britain, Germany, Italy were very strongly correlated. and This indicated that data on height performance of different provenances in Europe can be trusted in the United States, and data obtained in this study are probably applicable in other regions of the northeastern United States. viii TABLE OF CONTENTS Page A C K N O W L E D G E M E N T S .............................................. 11 S U M M A R Y ......................................................... iii LIST OF T A B L E S ................................................ X LIST OF FIGURES AND ILLUSTRATIONS............................. xii CHAPTER I. .......................................... 1 II. O B J E C T I V E S ............................................ 6 III. T A X O N O M Y .............................................. 1. Distribution ..................................... 2. M o r p h o l o g y ....................................... 3. Geographic V a r i e t i e s ............................. 4. Place in the Taxonomy of Other L a r c h e s .......... 7 7 11 13 16 IV. MATERIAL AND M E T H O D S ................................. 1. Plantation Establishment and C a r e ................ 2. Range of Provenances Studied . . . . ............. 3. Plantation Arrangement and Replication .......... 4. Measuring and S c o r i n g ............................. ............................. 5. Statistical Analysis 23 23 25 32 33 33 STUDIES RELATED TO S E E D ............................... 1. Variation in Seed W e i g h t ......................... 2. Relation Between Seed Weight and TreeHeight . . . 3. Rate of Germination .......... 37 37 V. VI. VII. INTRODUCTION 39 bP EFFECT OF S P A C I N G ..................................... GENETIC DIFFERENCES IN GROWTH CHARACTERISTICS ........ 1. Twelve-Year Heights and Diameters . . . 2. Stem Straightness .......... 3. Basal Sweep and L e a n ............................. 4. Late Snow D a m a g e ................................. 5. Infestation by Woolly Larch A p h i d ................. 6. Damage by Other A g e n c i e s ......................... ............................. a. Porcupine Damage b. Larch Canker in E u r o p e ....................... 7. Summary of the European Larch Racial Test Data . . U6 U6 hi 51 55 56 59 59 60 62 ix TABLE OF CONTENTS (Cont.) VIII. Page RELATIONSHIP BETWEEN GENETICS, DISTRIBUTION AND T A X O N O M Y ................................................. 1. Relationship Between Genetics and Taxonomy . . . . 2. The Genetic Variation Patterns Summarized by R e g i o n s .............................................. a. Polish L a r c h .................................... b. Sudeten L a r c h .................................... c. Slovakian L a r c h .................................. d. Low-Elevation (550-950 Meters) Larch from A u s t r i a .......................................... e. High-Elevation (925-1900 Meters) Larch from Austria and Italy and Larch from Switzerland . 6 6 66 6 6 67 67 69 69 70 IX. LARCH PLOTS OF MISCELLANEOUS O R I G I N S ................. 72 X. APPLICABILITY OF PRESENT D A T A .......................... 1. Applicability of Present Data in Other Regions . . 2. Applicability in Future Production of Hybrids . . . a. Japanese Larch as a Crossing P a r t n e r ......... 76 76 79 81 XI. XII. FINDINGS RELATED TO CONDUCTING FUTURE PROVENANCE S T U D I E S ................................................... 1. Need of R e p l i c a t i o n ................................. 2. Applicability of Early Height Measurements . . . . 3. Efficiency of Statistical Analyses by Using Different S a m p l e s ............ .................. .. literature cited ............................. 83 83 8 I4. 85 86 X LIST OF TABLES Table 1. Page Nomenclature of European Larch V a r i e t i e s ................ Ill 2. Natural Distribution of the Genus Larix .................. 17 3. Key to the Species and Varieties of L a r c h e s .............. 19 4. Location of IUFRO Larch Provenance Studies in North America and E u r o p e ........................................ 2ii 5. 6 Larch P l a n t a t i o n s ........................................ 26 . Description of Methods Used in Establishment of the American IUFRO Larch Plantations ........................ 28 7. 8 Description of Sites Used in Establishment of the IUFRO Origin Data, IUFRO Larch Provenances from Native Stands Tested in the United S t a t e s ............................... 30 . Origin Data, IUFRO Larch Provenances from Planted Stands Tested in the United S t a t e s ............................... 31 9. 10. 11. 12. Description of Measurings and Scorings Made in the American IUFRO Larch P l a n t a t i o n s ........................ 3 J 4 Provenances of European Larch from Native Stands Classi­ fied by Geographic Regions and Elevation ............... 3 6 The Geographic Pattern of Variation in European Larch ...................... With Respect to Seed Weight 38 Seed Weights and 4-Year Heights in 27 Corresponding Provenances ............. ................................. h 1 13. Average 12-Year Heights and Diameters at Breast Height of Interior Trees Compared with Those of Border Trees . . 14. Geographic Variation in European Larch with Respect to 12-Year Heights and Diameters, New Hampshire Data . . . . bl Geographic Variation in European Larch with Respect to 12-Year Heights and Diameters, Combined New York and New Hampshire D a t a ............................................ I4 8 15. xi LIST OF TABLES (Cont.) Table 16. 17. 18. 19. 20. 21. 22. 23. Page Differences in 12-Year Heights Among Provenances of Larch, Based on Two Randomized Blocks Planted in New Hampshire and One Block in New Y o r k ...................... Il9 Geographic Variation in European Larch with Respect to Stem Form, New Hampshire D a t a ............................. 50 Geographic Variation in European Larch with Respect to Percentage of Trees with 5° or More Lean, and Percentage of Trees Exhibiting Basal Sweep, New Hampshire Data . . . Sh Geographic Variation in European Larch with Respect to Percentage of Trees Exhibiting Bare Branches in the Inner Portion of Crowns as a Result of the Late Snow Damage in 1958 56 Geographic Variation in European Larch with Respect to Percentage of Trees Heavily Infested by the Woolly Larch A p h i d ....................................................... 59 Summary of the Data of European Larch Racial Test (IUFRO) at the Fox State F o r e s t ................................... 63 Classification of Origins by Statistically Proven D i f f e r e n c e s ................................................. 65 Correlation Between Average Heights of IUFRO Larch Provenances Tested in Northeastern United States (New Hampshire and New York) and the Same Provenances Tested in Europe.................................................... 73 xii LIST OF FIGURES Figure 1. Page European larch near the State Forest Nursery in Gerrish, New Hampshire, 42 years after planting .................... 3 2. Distribution of the genus Larix ............................. 8 3. Natural distribution of European larch in the two large and several small areas .......................................... 9 4. 5. 6. European larch in its natural range at La Val Bondasca in the Italian A l p s .......................................... 10 Open grown European larch near the State Forest Nursery in Gerrish, New Hampshire ..................................... 12 Graphic representation of the phylogenetic relationship in ........................... 22 The IUFRO larch plantation at the Fox State Forest, Hills­ boro, New Hampshire, in 1958, 10 years after planting . . . 29 the genus Larix after Sukachev 7. 8 . Sudeten larch was heavily attacked by the woolly larch aphid (Chermes strobilobius Kalt.) 9. ....................... Graphic summary of the data by characters and regions . . . 58 6]j. 10. Polish larch (IUFRO-38) in New H a m p s h i r e .................. 68 11. European larch from planted stands in Scotland (IUFRO-53) 7h . 1 I. INTRODUCTION Larch (Larix Mill.) is a coniferous genus containing 10 species and a few varieties that are distributed in the cooler regions and mountains of the northern hemisphere. European larch (Larix decidua Miller), from the mountains in C e n ­ tral Europe, is one of the most important larch species. It has a relatively small natural range which has been considerably increased by planting. European larch has been present in England since 1629 (51), in Germany since 1700, and in Norway since 1740 (76 ). It is the third most commonly planted species in Great Britain, occupying an area of 132,000 acres (51). Its plantings in West Germany total about 85,000 acres (75). Rapid growth was the main reason for its being highly regarded for planting. In Great Britain the 50-year height for site class I is 80 feet and the corresponding volume (over bark) is 7,100 cubic feet per acre (82). European larch is used widely for fence posts, pit- props, boat-, pier-, bridge-, wagon- and river-defense works (51). This wood is highly valued in Europe for its durability. sound for several centuries in Venice, It stayed Italy, when used under water. Some churches built of this material in Poland three to five hundred years ago are still standing (34). European larch was introduced into the United States about the middle of the 19th century. Individual trees or small stands are found throughout the northeastern United States 02, 35, 50). Hunt 2 studied these larches in New England and New York in 1932 and reported that on well-drained loamy soils they made good growth and proved to be a valuable addition to the forests of the Northeast (35). At the age of 50 years on good soils (site X) the trees were 72 feet high, 10.8 inches d.b.h. and yielded about 5,100 cubic feet outside bark per acre. Cook reported that at Cooxrox Forest in northeastern New York 15-year-old dominant trees had an average height of 40 feet and were beginning to give merchantable products (15). The greatest interest in this species has centered in New York State where the state forests included about 11 million trees in 1958, and at present nursery inventory is over 1 million trees. In the Penn­ sylvania state forests about 1.6 million trees were planted between 1899 and 1958. Northeast, Elsewhere, in the Corn Belt, Lake States, and the there is a great deal of current interest in the species be­ cause of its rapid growth. The present output in this country is rela­ tively low and its suitability for various market requirements is not finally determined. It may be especially suited for small, privately owned woodlands because its wood is suited for general farm use and merchantable yields are produced at early ages (figure 1). European larch is a variable species. tion of seed sources, Because of improper selec­ as early as the middle of the nineteenth century some plantings in Germany failed (90). By now it is proven that trees of different geographic origins differ from each other in growth rate, stem form, resistance to disease and other characteristics that are important in forestry. 3 Figure 1.— European larch near the State Forest Nursery in Gerrich, New Hampshire, 42 years after planting. The average height was 75 feet and diameter at breast height, 13 inches. h Cieslar initiated the first provenance study of European larch at Mariabrunn, Austria, ral stands. the in 1879. He included two seed sources from natu- One of these was from Sudetenland, Czechoslovakia, other from the Austrian Tyrol. A few years another study with later he established 14 provenances from the same regions. with Tyrolean provenances, As compared trees from Sudetenland had more rapid growth, longer live crown, heavier wood, higher wood specific gravity, bark percentage, dle and lower earlier needle appearance in spring and earlier nee­ fall in autumn (9, ^LO, 11, 12). Following the first experiment by Cieslar, provenance studies were established in most European countries: Britain, Czechoslovakia, Switzerland, Sweden, Great Germany, Belgium, Finland, Denmark, None of these experiments was large or well designed. ly, they yielded little valuable new information. and France. Taken individual­ However, when summar­ ized as a group by Kalela (38) and Schober (73), they yielded some new information, such as the fact that susceptibility to larch canker (Dasyscypha willkomii (Hartig) Rehm.) was related to origin. There are a few small seed origin studies of European larch in the United States. These are of limited value because only a few prove­ nances are represented from , natural stands. Littlefield and Elia- son made a study of three seed origins in New York. Seventeen-year old trees originating from planted stands in Scotland and Silesia had significantly greater heights than those from the Italian Alps (48). Earlier study of 5-year-old seedlings in New York showed that seed origin from Scotland had greater height growth than that from Silesia (47). Stoeckeler reported that in northeastern Wisconsin origins from 5 Silesia, the German Alps, and Czechoslovakia ranked higher in 4-year height growth than origins from the Austrian and Italian Alps and Germany (81). However, the origin data were insufficient to tell whether or not the faster growing origins are from native stands. Cook observed that an origin from planted stands in Scotland was more attacked by squirrels than a provenance from Silesia (14). A summary of European larch studies, mainly on its racial varia­ tion, was reported by McComb (50). However, the applicability of European findings under American conditions remained untested. The first extensive study of geographic variation in European larch under American conditions was initiated by Dr. Henry I. Baldwin. He received 42 seed lots of European larch and 3 seed lots of other larches from the International Union of Forest Research Organizations (IUFRO) and shared these with other research agencies. were established in two locations, Plantations the Fox State Forest (Hillsboro, New Hampshire) and a State Forest of New York (Bakersville, New York). Baldwin reported 4-year heights of transplants in New Hampshire and made some studies on the period of height growth in different prove­ nances of European larch (3^, 4:). I performed the measuring and scoring of the 12-year-old American IUFRO larch plantings in the summer of 1958. The present study of geographic variation in European larch was based on these and earlier data obtained. 6 II. OBJECTIVES The objectives of the present study were as follows: 1. Determine the European larch variation pattern in seed weight and relate this to juvenile tree growth. 2. Determine the geographic variation of European larch with re­ spect to 12-year heights and diameters, late snow damage, susceptibility to the woolly larch aphid (Chermes strobilobius Kalt.) 3. stem form, basal sweep, lean, and other characteristics visible at this age. Determine the ultimate natural origin of the planted stands represented in the IUFRO tests. 4. Determine the relationship between the genetic pattern of vari­ ation and toxonomic variation in European larch. 5. Determine the consistency of height growth of the same prove­ nances tested at different locations in Europe and the United States. 6. Show the relationship of the results of the IUFRO European larch provenance test to larch improvement programs in general. 7. Determine the reliability of early measurements and possible improvements in conducting experiments. 7 III. 1. TAXONOMY Distribution The range of European larch occupies two main distribution areas and a few small, outlying areas (figures 2 .and 3)« One of the main portions of the range is in the Alps Mountains of northern Italy (figure ij.), southern Switzerland, Austria, parts of southern Germany and northeastern Yugoslavia. and adjacent Within this area European larch grows at elevations between 300 and 2,500 meters. Its southernmost point is in the French Alps at 43°55’N, just a few miles north of Nice (25). Lake Garda at 45°40’N (52). Its other southern limit in Italy reaches The westernmost point of European larch is at Val Jouffrey, France (25). This main portion of the range ex­ pands into lowlands of Eastern Austria where it reaches 48°40'N in the North and 16o20'E in the East (86). The second principal distribution area of European larch is in southern Poland and eastern Czechoslovakia * It grows here at elevations of 150 to 500 meters scattered in flat and rolling regions in Poland, and at 500 to 1,000 meters in the Tatra Mountains and other parts of Slovakia, eastern Czechoslovakia. 50°05'N, near the town Bronica, It reaches its northern limits at Poland, its western limits at 18°13'E and its eastern limits at 2 2 ° 4 3 !E, near the Polish-Ukrainian border. This principal area of the range of European larch expands throughout Slovakia (eastern Czechoslovakia) and reaches its southernmost point Figure 2.--Distribution of the genus Larix. European larch (Larix decidua Miller), which is isolated from other species, has one of the smallest ranges in the genus. 9 3 3 © & s cd 3 3 3 CD CD cfi 3 3 cd PO L A N D CD rQ X! © -P 3 a •p bJO -3 3 ■H rH © 3 3 3 •H J3 o 3 3 3 S3 35 3 P CQ m *H 3 i —( ,3 P 3 3 © 3 •H o 3 © a 3 35 w O 3 0 3 a o c t f P f*i 1 I CO O 0 P P. 0 3 a tH O •H 3 3 S3 X) •H O *H —( i —1 o 0 0 i O ft XJ Eb . .. A + > ■H •H U bA a t o *H > ■rl at 3 Xt ■H rQ •H 01 0 T3 U . O • 3 f t a tc d > > C C S■ i H 3 3 Xi ■H *H —1 O 0 0 xt bb • . f H 3 > • 3 3 > . 3 3 •H ft •H O 3 •H a • u 3 > •H •rH •rH •rH 3 3 3 o •rH •H *H •rH i —1 i —1 —1 i S h 0 0 0 3 E j S S rH bb bb bb • . L • 5 H 3 T O •H m 3 0 b i ) rH O • L 3 3 3 ■H 3 O *H 3 O S 3 . 3 • » “D 3 -rH 3 •H ft 0 rH O 4 ^ ft 0 iH • -3 • a f —1 3 -P 3 0 XJ •H O O O . 3 3 3 *H -H 3 •H P rH 0 i —IP 3 3 3 i —I & .. "H ■H 3 •H 3 3 -P o ft . iJW Figure 6.— Graphic representation of the phylogenetic relationship in the genus I^arix^-/ after Sukachev (37). Current names have been used. 23 IV. 1♦ MATERIAL AND METHODS Plantation Establishment and Care The International Union of Forest Research Organizations, called IUFRO, was formally organized in 1892 and initiated several cooperative studies. The Subcommittee on Forest Seeds and Forest Tree Races under the leadership of Dr. Werner Schmidt decided to start a larch prove­ nance study at its meeting in 1940. Seed collection for this study continued for three years— 1942, 1943, trees and their ages varied. and 1944. The number of mother The majority of collections was made from native stands of European larch; the rest included seed from planted stands of European larch and other larch taxa. The total num­ ber of 56 provenances included 51 seed lots of European larch, Siberian larch, one of Japanese larch, two of and one that originated from spontaneous European and Japanese larch hybrids. These seed lots were distributed to 11 different countries in Europe and North America. The experiment in Finland and Denmark failed. In nine other countries experiments were established in 17 different locations (table 4). The number of provenances in these experiments varied from 6 to 56 (88). United States participation in the IUFRO larch study was initiated by Dr. Henry I. Baldwin, who had been in contact with IUFRO and some members of the Subcommittee on Forest Tree Races since its 1929 Congress in Stockholm. In March 1946, Dr. Baldwin received 46 seed lots of dif­ ferent larch provenances from Prof. Sven Petrini, General. The seed, IUFRO Secretary averaging about 100 grams per provenance, was 2k Table 4.— Location of IUFRO larch provenance studies in North America and Europe (88). Expt. No. Country, District, Location Latitude Longitude Altitude meters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 U.S.A., New Hampshire, Hillsboro U.S.A., New York, Barkersville Canada, Alberta, Kanaskis For. Exp. Sta. England, Wiltshire, Savernake For. England, Herefordshire, Haugh For. England, Herefordshire, Mortimer For. England, Shropshire, Walcot For. England, Worchestershire, Wyre For. Scotland, Pertshire, Drummond Hill Germany, Niedersachsen, Bremervorde Germany, Fulda, Neuhof-West France, Loiret, Nogent s. Vern France, Meurthe et M . , Arb. d. 1. Sivr. Italy, Pistia, Aquerino For. Sweden, Halland, Hjuleberg Switzerland, Zuerich, Hoenggerberg Czechoslovakia, — , Podbansko For. Seed Sources number 43°09 *N 71°55 fW 240 46 43°07*N 7 4°02 fW 460 37 50°56fN 115°10fW 1350 10 51°24'N 1 ° 3 8 ,W 150 38 52°01'N 2°36 fW 150 14 5 2 ° 1 9 ?N 2 ° 5 3 ’W 230 13 52 °25'N 3°01'W 260 27 52°25'N 2 ° 2 2 ’W 80 11 56°34’N 4°0 6 ’W 300 20 53°28'N 9 ° 0 8 TE 50 25 5 0 °40TN 9°33 TE 46°50’N 2 ° 4 8 fE 150 9 4 8 °45’N 6 ° 0 9 1E 300 6 44°01 ’N 11°05’E 960 15 56°56 'N 12°44TE 150 21 47°25TN 8°29 fE 535 17 950 41 56 25 subdivided among four different research agencies. were established in two locations only: ation Commission of However, plantations one by the Forestry and Recre­ the State of New Hampshire and another by the York State Conservation Department. Details of sites and New methods used in establishment of these American IUFRO larch provenance studies are presented in tables 5 and 6. In New Hampshire a separate study was made with 47 provenances of European larch which were transplanted to the nursery as 2-0 and mea­ sured at the age of 4 years as 2-2 (4). The IUFRO larch plantings in New Hampshire were partially damaged by porcupines (Erethion dorsatum dorsatum Linnaeus) particularly in one corner of the area. It was also attacked a few times by woolly larch aphid (Chermes strobilobius Kalt.) and other minor pests. At the IUFRO larch plantation in New York a few trees had dead tops be­ cause of porcupine damage and some trees had abnormal upper crowns be­ cause of snow damage. 2. Range of Provenances Studied In the summer of 1958, when the major data were collected, the IUFRO larch plantation in New Hampshire consisted of 45 different provenances and the New York plantation included 52 provenances, shown in the following tabulation: as 26 Table 5,— Description of sites used in establishment of American IUFRO larch plantations. Data on the IUFRO Larch Plantation in New York in New Hampshire Location Fox State Forest, Hillsboro, Hillsborough County, New Hampshire N. Y. State Forest, Barkersville, Saratoga County, New York Latitude, longitude 43°09'N, 7 1 ° 5 5 TW 43°07fN, 74°02fW Altitude 800 feet (240 meters) 1500 feet (460 meters) Slope and aspect 0-5 percent, east 0-5 percent, Soil series Marlow & Whitman (fertile) Gloucester (submarginal for agriculture) Soil type Loam (stony phase) Fine sandy loam B horizon Platy fragipan at the depth of 20 inches Yellow brown loam Parent material Platy olive glacial till Glacial till from gran­ itic rock Internal drainage Well drained on Marlow; poorly drained on Whit­ man Very well drained Past use of site Old field, in use up to time of planting Old field, abandoned for several years Mean total precipi­ tation 45 inches 40 inches Snow fall 64 inches 69 inches Mean July tempera­ ture 68°F 78°F Mean January temperature 22°F 20°F southwest 27 Table 5.— (Cont.) Data on the IUFRO Larch Plantation in New Hampshire in New York Mean annual temperature 46°F 46°F Last frost average date May 12 May 10 First frost average date October 6 Species European larch European larch European and Japanese larches Spontaneous hybrids European larch Japanese larch Siberian larch Tamarack October 1 Seed Source Natural stands Planted stands Mixed species Planted stands Origin unknown Planted stands Planted stands Natural stands Total Number of Provenances Number of Provenances in New Hampshire New York 31 26 10 10 1 — 1 1 7 1 1 1 — 1 1 45 52 The most important data were obtained from the 31 European larch provenances from indigenous stands. seed lots is presented in table 7. Description of the origin of these These provenances represented all three major subdivisions in the natural range of European larch: Alpine, the Polish-Slovak!an and the Sudeten (figure X ) . the The other IUFRO provenances from planted stands of European larch and other larch taxa are described in table 8. Data from these larch provenances and other provenances of unknown origin in the New York plantation were of limited value. 28 Table 6.— Description of methods used in establishment of the American IUFRO larch plantations. Nursery stock pro­ duced at Fox State Forest, Hillsboro, New Hampshire Saratoga Nursery, Saratoga Springs, New York Seed sown at the nursery May 9 to 23, 1946 June 4, 1946 Stock outplanted in the field Early April, 1948 April 22 and 23, 1948 Stock age 2-0 to i o Data on the IUFRO Larch Plantation in New Hampshire in New York Ground preparation Sod removed in 1-foot square No sod removal Spacing 5 x 5 6 x6 Arrangement of provenances Rectangular plots Rows Plantation cleaning Sprouts removed several times One partial cleaning Pruning At age of 11 years, to 6-9 feet above ground None Present state (1958) Clean, well marked, roads moved (figure 6) Brushy, well marked and 4 x 6 Figure 7.— The IUFRO larch plantation at the Fox State Forest, Hill boro, New Hampshire, in 1958, 10 years after planting. marked and well maintained. It was well 30 Table 7.— Origin data, IUFRO larch provenances from native stands of European larch tested in the United States. Provenance IUFRO No. North Latitude o t District, Location East Longitude t o Altitude meters 1 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Austria u Salzburg, Bluhnbach Karnten, Hollenburg Tyrol, Innsbruck Lower Austria, Krumbach Lower Austria, Lamerau T y r o l , Landeck Steiermark, Murau Karnten, Obervellach Tyrol, Pitz Thai Tyrol, Ried-Tossens Lower Austria, Wartenstein Tyrol, Steinach-Muhlbach Steiermark, Waldstein Salzburg, St. Michael i.L. Steiermark, Murau-Paal Tyrol, Ried-Pflunds Tyrol, Steinach-Gries 49 46 47 47 48 47 47 46 47 47 47 47 47 47 47 46 47 29 33 14 31 05 08 08 55 05 00 40 06 14 05 04 55 02 13 14 11 16 16 10 14 13 10 10 15 11 15 13 14 10 11 10 18 23 12 10 37 10 13 50 37 55 28 15 39 06 35 28 600 900 900 600 700 750 950 1100 1100 1050 800 900 550 1700 950 1900 1900 21 23 24 25 Italy Belluno, Monte Col Trento, Lago (V. Fiemme) Trento, Feugo (V. Fiemme) Bolzapo, Val Venosta 46 46 46 46 32 17 20 35 12 11 11 10 26 23 27 40 1350 925 1400 1100 26 27 Switzerland Wallis, Lotschenthal 46 23 46 57 Graubungen, Untervaz 7 47 9 32 1500 550 38 39 42 43 Poland Kielze, Pinczow Kielce, Zagnansk Krakau, Nowy Sacz- Toki Kielze, Starachowice 50 50 49 51 30 58 38 08 20 20 20 21 30 42 42 03 270 350 450 200 46 47 51 52 Czechoslovakia — , Hubertkirch 50 50 — , Hubertkirch 49 Zilinsky, Liptowski C. V. 49 Zilinsky, Liptowski M. 04 04 02 02 17 17 19 19 18 18 40 40 700 700 825 1000 — ■ 31 Table 8.— Origin data, IUFRO larch provenances from planted stands tested in the United States. Provenance IUFRO No. Species of Larch Country, District, Location North Latitude O 28 29 30 Japanese European European 31 European 32 European 34 35 36 44 45 European European European Mixed: Eur. & Jap. European 49 European 53 European 54 Siberian 55 European & Hy­ brids Denmark, Djursland, Meilgaard Germany, Hannover, Harbke Germany, Wurtt. Baden, Neckargemund Germany, Sch l . Holst., Neu Munster Poland, SIask Dabrowsky, Proskau Poland, Mazury, Schlobitten Poland, Gdansk, Sobbowitz Poland, Gdansk, Stolp Poland, Warszawa, West Mala Wies Czechoslovakia, Brenensky, Hrottowitz Czechoslovakia, Ostravsky, Parchowitz Scotland, Morayshire, A1droughty Estate Sweden, Jonkopings lan, Visingso Sweden, Jonkopings lan, Visingso 1 East Longitude o i 56 31 52 10 49 23 10 37 11 00 8 49 54 05 10 00 50 34 17 48 54 54 54 52 19 18 17 21 08 09 28 10 47 37 06 00 49 16 16 07 49 30 17 42 57 39 3 23 58 02 14 20 58 02 14 20 32 3. Plantation Arrangement and Replication The IUFRO larch plantation in New Hampshire was arranged in 20 x 20-, 25 x 32.5-, 25 x 30-, and 25 x 45-foot square plots. the following plots located at random: 2 for each of 13 provenances, one for each of 31 provenances, and 3 for one provenance. was divided into three compartments by two roads. division had no statistical meaning. It contained The plantation However, this sub­ For analysis of the replicated provenances, two completely randomized blocks were recognized. One of these occupied the western third and the other the eastern two-thirds of the plantation. As of the summer of 1958, the plantation included 38 provenances represented by over 20 trees and 7 provenances repre­ sented by 10 to 20 trees, border trees not counted. The IUFRO larch plantation in New York was arranged in 29 rows, each consisting of 50 plants. with tamarack. The rows on either side were planted Each IUFRO larch seed lot was represented by a single plot and the provenances were planted in numerical order. Each plot consisted of a part of a row or parts of two adjacent rows, depending on the amount of planting stock available. In the summer of 1958, this plantation included 11 IUFRO larch provenances represented by over 20 trees, eight by 10 to 20 trees, and 19 by less than 10 trees. This representation did not include trees with broken tops and those growing on the plantation edges. 33 4. Measuring and Scoring Seed testing was done at the Saratoga Nursery, New York, by E. J. Eliason. He determined 1,000-seed weights and germination percents at 8 , 15, 20, and 45 days after seeding in Petri dishes under room condi­ tions and in a Jacobsen germinator. The types of measurements and scorings in the field are described in table 9. Heights were measured with simple graded poles at earlier ages; at the plantation age of 12 years a sliding pole was used with a rope and pudley. ameter tape. Diameters at breast height were measured with a di­ Each tree was measured for height and diameter. Other scorings were made on the interior trees, that is, those that were not in the rows exposed to the outside of the plantation or to an open space of 15 or more feet. Early measurements, before 1958, were made by H. I. Baldwin in New Hampshire, Eliason in New York. and under the supervision of E. J. The 1958 measurements and scorings were the author*s responsibility. 5. Statistical Analysis Plot means and provenance means were used in the analyses of variance. tion. Within-plot variation was largely due to natural competi­ For this reason separation of genetical from environmental components within plots was hot attanpted* When analyzing genetic differences, each provenance mean was based on all live trees in the plot except trees which were facing 3k Table 9.— Description of measurements and scorings made in the American IUFRO larch plantations. Character Studied Total height Plantation Measured or Scored Date Age From Seed years Unit of Measuring or Scoring New Hampshire (2-2 transpl ant s ) New Hampshire New York New Hampshire New York New York Sept O c t ., July, July, Oct. , 1952 1954 1958 1958 1954 7 9 12 12 9 0.5 0.5 0.5 0.5 0.1 New Hampshire July, 1958 12 0.1 inch New York July, 1958 12 0.1 inch New Hampshire and New York July, 1958 12 Stem form New Hampshire and New York July, 1958 12 Basal sweep New Hampshire July, 1958 12 Lean New Hampshire July, 1958 12 Late snow damage New Hampshire July, 1958 12 Damage by the woolly larch aphid New Hampshire July, 1958 12 Damage by por­ cupines New Hampshire July, 1958 12 Total height Total height Total height Total height Diameter at breast height Diameter at breast height Diameter at breast height Stem form O c t ., 1949 4 0.1 foot foot foot foot foot inch Number of crooks per tree in the lower 3/4 of stem Number of trees that exhibited slight crooks in the upper 1/4 of stem Number of treeswith basal sweep and ex­ hibiting J-form Number of trees di­ verging more than 5° from vertical Number of trees with­ out foliage on the majority of interior branches Number of trees with more than 3 percent of needles attacked by aphids in the lower portion of crown Number of trees with large sharp crooks 35 open spaces of at least 12 feet and trees with broken tops. Plots with more than 20 percent porucpine-damaged trees were entirely ex­ cluded from calculations of height and diameter means. Justification for these exclusions is discussed later. Provenance mean heights based on the four largest trees per prove­ nance and on the one largest tree per plot were also calculated. These were used to determine the value of quick measurement methods. Percentage data used in the analyses of variance were transformed to angles corresponding to percentages, Angle = Arc sin /Percentage. The IUFRO larch plantations in Hillsboro, New Hampshire, and Barkersville, New York, included 25 identical provenances of European larch from natural stands. These were subject to analysis of vari­ ance as follows: Source Total Plantations Provenances Plantations x Provenances (Error) Degrees of Freedom 49 1 24 24 Data of ten replicated provenances in two blocks in New Hampshire and in one block in New York were analyzed as follows: Source Total Blocks Provenances Blocks x Provenances (Error) Degrees of Freedom 29 2 9 18 Data based on the measurements and scorings in the New Hampshire plantation alone were analyzed after grouping the provenances into 10 geographic or altitudinal regions (table 10). This analysis of variance 36 Table 10.— Provenances of European larch from the native stands classi­ fied by geographic regions and elevation. Origin of Seed Sources Region Provenances of European Larch Elevation IUFRO numbers meters Pol and Sudetenland, Chechoslovakia Slovakia, Chechoslovakia Eastern Austria Central Austria Central Austria Western Austria Western Austria Italy Switzerland Low Low Low Low Low High Low High High L. & H. 200- 450 700 825-1000 550- 800 600- 950 1100-1700 750- 900 1050-1900 925-1400 550-1500 38 46 51 14 1 9 7 11 23 27 39 47 52 5 3 15 4 10 25 26 42 — — 6 8 — 13 17 21 *" 43 — — 12 16 — — 18 24 “ **• between provenances from different regions was made by the statistical method as follows: Source Total Between Regions Within Regions (Error) Degrees of Freedom 30 9 21 Standard method was used in correlation analyses of provenance means or characteristics of individual trees. Significance of r (correlation coefficient) was determined by Snedecor's method (79). 37 V. 1. STUDIES RELATED TO SEED Geographic and Altitudinal Variation in Seed Weight The variation in average seed weight in three species of larch represented in the American IUFRO larch tests is shown below. These values were based on 1000-seed averages which included an unidentified proportion of empties which were difficult to separate from the sample. The seed of European and Japanese larch were of about the same weight but both these species had lighter seed than Siberian larch, as shown in the following tabulation. Species Provenances Studied number Seed Weight milligrams European larch Siberian larch Japanese larch 33 2 1 4.3 to 8.6 11.3 to 14.2 6.2 The geographic pattern of variation in seed weight within European larch is shown in table 11. Analysis of variance based upon prove­ nances classified by 10 different regions showed no significant differ­ ences among regions at the 5-percent level. The possible presence of an altitudinal trend was tested by a correlation analysis in which altitude of provenance was the independent variable and seed weight was the dependent variable. Thirty-one pairs of these values were analyzed and showed a strong correlation (signifi­ cant at the 5-percent level, with r = .371 for 29 d.f.). This 38 Table 11.— The geographic pattern of variation in European larch with respect to seed weight. Origin of Seed Source Region Elevation Seed Weight By Provenance meters Eastern Austria Central Austria Pol and Slovakia Sudetenland Central Austria Western Austria Western Austria Switzerland Italy milligrams 550- 800 600- 950 200- 450 800-1000 700 1100-1700 1050-1900 600- 950 550-1500 900-1400 6 .0, 5.0, 6.3, 5.4, 5.9, 6 .2, 6 .2, 6 .6, 4.4, 6.7, 4.6, 5.0, 4.7, 5.6 5.0 5.0 5.7, 5.7, 8 .6 7.1, Average milligrams 4.7, 4.3 5.0, 5.2 5.9, 4.6 5.9, 5.6 5.9 6.7, 5.8 4.9 5.0 5.4 5.5 5.5 5.6 5.9 6.1 6 .5 6.6 indicated that provenances from higher altitudes had heavier seed. The general trend of this altitudinal variation is illustrated in the following tabulation: Elevation meters Provenances Studied number Average Seed Weight milligrams Below 750 750 - 1000 Over 1000 13 8 10 5.2 5.6 6.3 Bouvarel studied 37 non-IUFRO seed lots of European larch and found differences in seed weights as follows (8): 39 Seed I*ots Studied number Average Seed Weight milligrams 6 3.5 Austria (300-1200 meters) 13 5.5 Western Alps (Brianconais) (1200-1700 meters) 18 9.3 Origin and Elevation Sudetenland-Poland (300-800 meters) These findings support the conclusion that high-altitude prove­ nances of European larch mainly in the western parts of its range have heavier seeds than those from low altitudes in the eastern part of the r a nge. At present, this variation must be regarded as phenotypic. As far as known, there have been no critical experiments showing that high elevation sources produce larger seeds if grown at low elevations. 2. Relationship Between Weight and Tree Height Seed weight is one of the factors which may affect the performance of young seedlings because a higher content of stored food naturally favors early growth. example, This has been observed in many species. For the height of white pine (Pinus strobus L.) seedlings at the age of 1 and 2 years depends very strongly on seed weight (57). This was also evident in European larch from Engler's data when provenances came from relatively the same region (21, 38). U) In this study a test was made to determine the relation between seed weight and average height of 4-year-old seedlings (table 12). Analysis of 27 pairs of these values showed a significant negative correlation (r = -.392, d.f.). significant at the 5-percent level with 25 This negative correlation indicated that provenances with heavier seed (those from higher elevations) exhibited relatively slow growth. Similar results were obtained from an analysis of Cieslar’s data (38). Analysis of his data on seed weights and 1-year heights for five provenances showed a negative correlation (r = -.385, not significant at the 5-percent level with 3 d.f.). These results apparently disagree with earlier conclusions that seed weight and juvenile growth are positively correlated. The dis­ agreement is more apparent than real because present data were based on seed sources from a broad geographical and altitudinal range in which case hereditary factors acting with respect to height growth overruled the effect of seed weight on seedling growth. It is concluded that seed weight in European larch is not a reli­ able factor in indicating future height growth. 3. Rate of Germination The IUFRO European larch tested in the United States showed an average germination of 26 percent. The Japanese and Siberian larch IUFRO seed lots had average germination of 27 and 55 percent, respectively. 10. Table 12.— Seed weights and 4-year heights in 27 corresponding prove­ nances tested in Hillsboro, New Hampshire. The seed weighttree height correlation was negative (r = -.392, significant at the 5-percent level with 25 d.f.). Provenance IUFRO No. Average Seed Weight feet 26 25 21 7 38 8.6 7.1 6.7 6.6 6.3 1.7 1.7 1.5 2.7 2.5 11 9 14 46 42 6.2 6.1 6.0 5.9 5.9 2.0 2.0 2.2 2.2 2.9 13 17 24 52 10 5.9 5.9 5.8 5.7 5.7 2.4 2.3 1.8 2.0 1.6 4 18 51 16 1 5.7 5.6 5.4 5.2 5.0 2.6 2.1 2.5 1.9 2.2 3 8 15 6 5 5.0 5.0 5.0 4.7 4.6 2.7 2.5 2.0 2.2 2.1 43 12 4.6 4.3 to * 00 milligrams Average 4-Year Height 2.4 U2 European larch showed a broad variation in germination percentage. The range for individual seed lots was from 1 to 75 percent. However, no attempt was made to relate this variation to the genetics of the species because germination may have depended on many environmental factors such as storage and transportation conditions. A study to determine the rate of germination was conducted for 45 days in 31 different provenances of European larch and 3 provenances of other species. There were no statistically significant differences among provenances from different regions in European larch. bined data for European, Siberian, The com­ and Japanese larches showed germina­ tion rates as follows: Species of Larch European Siberian J apanese Germination in Different Periods After Seeding 0-7 8-10 11-15 21-45 16-20 Days Days Days Days Days - percent 58 26 7 26 51 10 14 23 31 1 0 30 I 0 22 Average Germination Time days 11 9 18 European larch completed more than half of the total germination in 7 days and 99 percent in 20 days. These results agreed with ger­ mination data in Germany, where European larch completed 75 percent of its germination in 10 days (60). These results indicate that in this species germination involves few problems to obtain uniform progeny the first year. In Siberian larch germination did not start quite as soon as in European larch, but once initiated the average germination rate was slightly faster. Japanese larch germination was slower than either U3 of the other two species. This slowness agrees with the results of Langner, who found that Japanese larch may germinate as late as 65 days after sowing (40). The more rapid germination in European larch appears unrelated to the thickness of seed coat, because the coat is thicker in this species (.2 mm) than in Japanese larch (.1 mm). It may be related to the radicle or the higher rate of respiration in European larch seed than in that of Japanese larch (64). hk VI. EFFECT OF SPACING I studied the effect of spacing on height and diameter growth in order to determine whether to include border trees in the plot means for the comparative analyses and to determine whether differences in spacing of trees within plots affected height growth sufficiently to make an adjustment for spacing necessary. At the IUFRO larch plantation in New Hampshire, the majority of plots were either inside the plantings or were bordered by one or two rows of trees especially planted to eliminate border effect. However, along the roads 24 plots were exposed on one or two sides to open spaces of 12 feet or more. On the basis of height and diameter mea­ surements which were made for each tree in the plantation, means were calculated separately for interior trees and border trees (table 13). The interior trees averaged 2.5 feet taller and 0.4 inches greater in diameter than the border trees. Analyses of variance based upon plot means showed that these differences were significant at the 1-percent level. Following these findings, the border trees were excluded from the plot means used in genetic analyses. The IUFRO larch plantation in New Hampshire included 38 plots of European larch from natural stands in which average space per tree was calculated by dividing plot size by number of trees. portions of the plots were used in calculations. Only interior The space per tree in different plots varied from 23 to 63 square feet because of initial spacing and survival. The correlation between space per tree and Table 13.— Average 12-year heights and diameters at breast height of interior trees compared with those of border trees, based on the New Hampshire IUFRO larch data. The differences between border trees and interior trees are significant at the 1-percent level. IUFRO N o . and Plot Symbol Mean Heights Border Interior Trees Trees — —--- f e e t ------- Mean Diameters Interior Border Trees Trees ----- inches ------- b b a b a b a b a a a b a a a a b a a b a a b b 18.1 18.1 21.0 16.8 16.8 14.3 14.7 12.1 11.8 15.5 8.0 14.8 11.0 13.7 13.9 19.8 16.3 18.3 13.3 23.0 19.6 20.9 14.3 16.0 20.1 19.2 21.0 17.7 17.7 16.5 17.3 11.8 16.9 17.6 15.2 18.0 ■ 13.7 14.6 19.0 22.2 17.9 21.1 17 .1 23.6 21.0 25.9 17 .9 18.0 2.2 2.5 2.9 1.7 2.5 1.7 1.9 1.0 1.5 2.6 .9 2.0 1.6 1.2 1.8 2.7 1.8 2.7 1.8 2.4 1.6 2.9 1.6 2.2 2.8 2.6 2.9 2.4 2.5 2.2 2.3 1.4 2.1 2.3 2.3 2.7 2.0 2.1 2.4 2.7 2.4 2.7 2.0 2.6 2.2 2.9 2.2 2.4 Average 15.9 18.4 2.0 2.4 1 3 6 6 7 8 9 12 13 16 21 23 26 27 30 31 34 35 36 45 46 47 52 55 average height per plot was weak (r = -.158, not significant at the 5-percent level for 36 d.f.). Accordingly, there was no reason to adjust plot mean heights for differences in spacing. i|6 VII. GENETIC DIFFERENCES IN GROWTH CHARACTERS 1. Twelve-Year Heights and Diameters The geographic pattern of variation in 12-year heights among origins of European larch was determined on the basis of New Hampshire data. Analysis of variance showed very significant differences among provenances from different regions (table 14). Sudetenland, and Slovakia were significantly (at the 5-percent level) faster growing than those from other regions. origins were from Switzerland, Austria. Provenances from Poland, Italy, The slowest growing and the higher elevations in Among provenances from the Alps, those from lower elevations in Central Austria showed the best height growth and were significantly superior (at the 5-percent level) to provenances in Switzerland. Differences among provenances from the same region were studied from analysis of variance of the plot mean heights of the 25 prove­ nances which were represented once in New Hampshire and once in New York, and of the 10 provenances which were represented, twice in New Hampshire and once in New York (tables 15 and 16). There were no instances in which provenances from the same region differed signifi­ cantly. Similar analyses were made to determine the geographic variation pattern in 12-year diameter breast high (tables 14 and 15). The Slo­ vakian provenances were not significantly larger in diameter than the slowest growing origins although they had been in height. Otherwise, hi Table 14.— Geographic variation in European larch with respect to 12year height and diameter breast high, New Hampshire data. Origin of Seed Sources Region Elevation meters Average Height feet Difference Levels* at Percent 5 1 Average Diameter Breast High inches Difference Levels* at Percent 5 1 Pol and 200- 450 23.8 S s 3.07 s s Sudetenland 700 23.5 S s 3.00 S . S Slovakia 800-1000 21.6 s . s 2.50 x • Central Austria 600- 950 18.6 I s . . s 2.45 X • • Western Austria 750- 900 18.4 I . I . . 2.37 I I . Eastern Austria 550- 800 17.4 I . I I . 2.30 I II Western Austria 1050-1900 17.2 I . I I . 2.25 I II Central Austria 1100-1700 17.1 I . I I . 2.25 I II Italy 900-1400 16.0 I . I I . 2.15 I II Switzerland 550-1500 14.2 I I Ill 2.05 I II * • Provenances from the regions marked f,SM had superior growth to those from the regions in the same column marked "l". the ranking of the provenances and the significance relationships were almost the same for diameter as for height. 2. Stem Straightness Stem straightness may depend on hereditary factors. Burger ob­ served that offspring from poorly formed mother trees exhibited more 18 Table 15.— Geographic variation in European larch with respect to 12year heights and diameters based on the data on the IUFRO larch plantations in Hillsboro, New Hampshire, and Barkersville, New York. Average Origin of Seed Source Diameter IUFRO ElevaBreast No ._____ Region________ tion_____ High meters inches 42 47 38 43 51 4 1 6 14 7 17 23 52 12 16 24 8 13 18 5 21 27 15 26 25 * Sudetenland Pol and Pol and Poland Slovakia Western Austria Central Austria Eastern Austria Eastern Austria Western Austria Western Austria Italy Slovaki a Eastern Austria Central Austria Italy Central Austria Western Austria Western Austria Eastern Austria Italy Switzerland Central Austria Switzerland Italy 700 450 350 200 825 900 600 700 550 750 1900 925 1000 800 950 1400 950 900 1900 600 1350 550 1700 1500 1100 2.95 3.05 2.90 2.60 2.40 2.85 2.70 2.60 2.65 2.80 2.35 2.80 2.20 2.15 2.15 2.25 2.35 2.10 2.25 2.15 2.05 1.90 1.60 1.65 1.35 Average Height feet 21.5 21.5 20.0 19.4 18.8 18.0 17.9 17.8 17.6 17.2 17.0 16.9 16.7 16.6 16.2 15.1 15.0 15.0 14.5 14.1 13.4 13.0 12.2 11.4 10.6 Levels of Significance in Differences ______ of Heights*_____ 5-percent 1-percent s s S S s S s s s s s s s s I I I I I I I I I I I s s s s s s s s s s • • • I I I I I I I • I I I I I I • I • I I I * I I • I I I I I I I I I I I I I I I I I I I I Provenances marked "s" had significantly greater heights than those in the same column marked "i". The significance relationships were almost the same for the diameter data. U9 Table 16.— Differences in 12-year height among provenances of larch, based on the data of two randomized blocks planted in New Hampshire and one block in New York. IUFRO No. 45* 1 6 34* 23 52 3 55* 8 21 * Origin of Seed Sources Region Elevation meters Czechoslovak! a Central Austria Eastern Austria Poland Italy Slovakia Central Austria Sweden Central Austria Italy --- 600 700 --925 1000 900 --950 1350 Average Height feet Differences in Provenances at the 5-Percent Level** 22.0 18.5 18.3 S 17.7 17.3 17.0 17.0 16.0 15.8 14.2 I I I s s s I I I I I I . . . . . I Provenances from planted stands. ** Provenances marked HS" had significantly greater heights than those in the same column marked ,fIM . crooks than those from straight stems, but included no statistical proof of his finding (6). Other factors may affect stem straightness, including soil fertility (24, 28). In the present study two methods were used to score stem straight­ ness. First, 1 scored the upper stem for crooks of any size. crooks were due almost entirely to crooked leader growth. Those Second, I counted the number of large crooks (more than 1.5 inches offset) per 100 feet of stem in the lower three-quarters of stem of each tree. The New Hampshire data were used to establish the geographic pattern of variation in stem straightness (table 17). Analysis of variance showed that provenances from Sudetenland, Poland, and low 50 Table 17.— Geographic variation in European larch with respect to stem form, New Hampshire data. Origin of Seed Source Region Elevation meters Trees With Crooks in Upper Stem percent 5-Percent Difference Level* Crooks per 100 Feet of Stem Length number 5-Percent Difference Level* Sudetenland 700 42.5 H 5.1 L H Poland 200- 450 41.5 H 4.4 L . Western Austria 750- 900 38.7 . H 8.8 H . Eastern Austria 550- 950 25.5 . . H 3.2 L . Central Austria 600- 950 24.8 . . H 3.3 L . Slovakia 800-1000 24.0 L . H 2.3 L . Italy 900-1400 22.3 L L H 2.8 L . Switzerland 550-1500 19.5 L L . 3.6 L . Western Austria 1050-1900 16.6 L L . 4.3 L . Central Austria 1100-1700 8.5 L L L 1.7 L L * Provenances from region marked ,fH" exhibited significantly higher values than those from regions in the same column marked ML M . elevation Western Austria had significantly higher percentages of trees with small crooks in the upper stem than most provenances from other regions. Austrian provenances showed altitudinal differences; those from lower elevations (below 1000 meters) had significantly more small crooks in the upper stem than those from higher elevations. Provenances that had larger number of trees with small crooks in the upper stem also had large number of crooks in the lower stem. Analysis of variance showed that crooks per 100 feet of stem length were most numerous among provenances from low elevations in Western Austria. Provenances from Sudetenland ranked second and had signifi­ cantly more crooks than the lowest ranking provenances from high ele­ vations in Central Austria. In order to determine the variation among provenances within regions, the combined data from New Hampshire and New York were sub­ jected to the analysis of variance. Significant within— region differences were found among provenances from low-elevation Western Austria as shown in the following tabulation: Origin Crooks per 100 Feet of Stem number IUFRO- 7 from Bandek, Tyrol IUFRO-13 from Steinach, Tyrol IUFRO- 4 from Innsbruck, Tyrol 15 12 6 Number of crooks in IUFRO-7 was significantly higher (at the 5percent level) than in IUFRO-4. No within-region variation was determined among provenances from other regions. 3. Basal Sweep and Bean "Basal sweep" is a term used to indicate a bend at the base of a J-shaped tree with an otherwise vertical bole. of the entire stem from the vertical. "Bean" is a deviation Both characteristics resulted 52 from bending of already-formed stems rather than crooked growth of the leader. A great deal of the bending occurred during the first two years after planting and resulted in a basal sweep which was confined to the lower 3 or 4 feet of the bole. Also, a large proportion of the trees appeared to have been bent in 1956 or 1957, just prior to the measurements. stems. Those were the leaners with straight but off-vertical Intermediate types in which there was a bend part way up the bole were uncommon. Apparently the bending was caused by the weight of snow and ice which laid the recently planted and inadequately rooted trees on the ground. When they subsequently became firmly rooted, they grew vertically. This type of snow damage is not noticeable in firmly rooted larch seedlings grown from seed in the nursery. The explanation for the lean is similar except that the wind acted more strongly and the damage occurred later in the life, involving the partial upheaval of a large root system. Plant characteristics that are associated with the mechanics of basal sweep or lean, such as root size in rela­ tion to the tree height or shape and density of crown, and may react differently on the basis of origin. are hereditary Fischer studied eight mother trees and their open pollinated progenies with respect to stem form and found that basal sweep was under genetic control (20). In extreme cases, such as reported by Burger in Adlisburg, mother trees of very poor stem form with crooks and with basal sweep exhibited in their open-pollinated offspring various distinct crooks on the stem including basal sweep ((5). 53 In order to determine the relation between height and lean, I classed the 109 trees of provenance IUFRO No. 45, that were growing in three plots, as either leaning or vertical. Distribution of 12- year heights in these two classes was as follows: Average Twelve-Year Heights, Feet Provenance Mean Plot A Plot B Plot C Leaning trees Vertical trees 23.2 22.0 24.0 23.8 25.3 24.3 24.2 23.4 Analysis of variance between the heights of leaning trees and those of vertical trees showed no significant differences at the 5-percent level. This indicated that heights within a provenances had little or no effect on lean. Different results were obtained in the analysis of the correlation between average 12-year heights and corresponding average percentages of leaning trees in 38 different provenances. The correlation coefficient, r = .523, was significant at the 1-percent level, which indicated that provenances with greater heights had higher percentages of leaning trees. Analysis of variance of the transformed percentages of leaning trees showed distinct differences between provenances from different regions (table 18). Provenances from Poland, Sudetenland and lower elevations in Western Austria exhibited more leaning trees than those from other regions. The fewest leaning trees were found in Italian and high-elevation Austrian provenances. The analysis also showed that there were altitudinal differences among the Austrian provenances. The provenances from higher elevations probably owed their superiority to higher root-stem ratios and finner anchoring in the ground. $h Table 18.— Geographic variation in European larch with respect to per­ centage of trees with 5° or more lean, and percentage of trees exhibiting basal sweep, New Hampshire data. Region Elevation meters Trees With 5° or More Lean!/ percent Pol and Sudetenland, Czech. Western Austria Eastern Austria Central Austria Switzerland Tatra Mtns., Czech. Italy Western Austria Central Austria 150- 300 700 750- 900 550- 800 600- 950 550-1500 800-1000 900-1400 1100-1900 1100-1700 56.5 49.5 48.7 22.3 21.3 20.5 11.0 7.5 7.3 4.0 5-Percent Difference Leveli^ H H H L L L L L L L Trees Exhibiting Basal Sweep^/ percent H H H . L L L 59.2 37.2 71.9 27.7 41.7 7.4 24.7 36.7 22.5 15.9 Provenances from regions marked " h " had significantly more leaning trees than those from regions in the same column marked ,lL " . It.S.D, 5% 23.0; L.S.D. 1% 37 .0. □f trees ex­ Analysis of variance of the transformed percentage < hibiting basal sweep also showed significant differences in provenances from different regions (table 18). As in the case of lean, provenances from lower elevations in Western Austria and Poland ex­ hibited the highest percentages of trees with basal sweep. Provenances from higher elevations in Austria exhibited the fewest trees with basal sweep. Italian provenances have been more subjected to bending during juvenile growth than at more advanced ages and consequently exhibited relatively higher percentage of trees with basal sweep than those with lean. 55 Among the faster growing provenances, those from the Tatra Moun­ tains (Czechoslovakia) and from Eastern Austria exhibited relatively low numbers of trees with lean or basal sweep. However, the fastest growing provenances from Poland were highly affected. 4. Late Snow Damage In Southern New Hampshire the spring of 1958 was abnormal. The buds of European larch started growing in mid-April and leaves were fully developed by April 26. fell. On that date about one foot of wet snow This late snow caused severe defoliation of the upper portions of the crowns and less severe stripping of the needles on the lower branches. In the two months following defoliation new needles devel­ oped on the outer branches but not on the inner branches of damaged trees. This caused the needles to be clustered and trees to have ragged appearance. In July trees were scored that exhibited the bare branches in the inner portion of the crown. This characteristic was interpreted as related to the late snow damage. Percentages of trees with bare in­ side branches per provenance provided the data for analysis of vari­ ance (table 19). Provenances from Sudetenland and Slovakia showed the highest percentage of defoliated trees. Italian, Swiss and high- elevation provenances were affected the least. provenances showed altitudinal differences; Western Austrian low elevation provenances were very significantly more defoliated than the high-elevation provenances. 56 Table 19.— Geographic variation in European larch with respect to per­ centage of trees exhibiting bare branches in the inner por­ tion of crowns as a result of the late snow in the spring of 1958, New Hampshire data. Provenances From _____ Region Elevation meters 700 800-1000 550- 800 750- 900 150- 300 600- 900 1050-1900 550-1500 1100-1700 900-1400 Sudetenland, Czech. Slovakia, Czech. Eastern Austria Western Austria Pol and Central Austria Western Austria Switzerland Central Austria Italy * Levels of Significance Trees Heavily in Differences*_____ Defoliated____ at 5% Level at 1% Level percent 47.0 44.5 34.8 28.3 26.0 16.3 14.3 13.5 11.0 7.0 H H H • H H • H * • • • H • • • • • • • • • L L L L L • • L L L L L » • L L L L L L L L • • • L • • • L Provenances from regions marked MHn exhibited significantly higher percentages of bare branches inside the inner portions of crowns than those in the same column marked "l " . 5. Damage by Woolly Larch Aphid The woolly larch aphid (Chermes strobilobius Kalt.) was first observed attacking the IUFRO larch plantation in Hillsboro, New Hampshire, in 1957, seven years after it was established. sucking insect belonging to the Chermides family. black spruces as well as European larch; masses on the needles, This is a It attacks red and it appears as white woolly as dark individuals on the underside of twigs and as clusters of individuals at the base of the leaves (17). In the spring of 1958 following the snow damage on the foliage at the end of April and subsequent damp weather conditions, the IUFRO 57 larch plantation was again heavily attacked by the woolly larch aphid. In July this attack resulted in yellow or brown stripes on the needles of the short shoots. Injured leaves bent sharply at the point of injury and infested short shoots appeared as dense clusters rather than as open rosettes (figure 8). By August needles that were attacked started to fall, leaving the short shoots of heavily infested trees nearly devoid of foliage. This IUFRO larch plantation in Hillsboro, New Hampshire, was scored in July with respect to number of trees that showed more than 5 percent of needles injured at the lower parts of the crown. Per­ centages of such trees per provenance were transformed to Arc sin /Percentage and subjected to the analysis of variance, which revealed significant differences at the 5-percent level between provenances from different regions (table 20). Provenances from the Sudeten Slovakia and Eastern Austria were more attacked by the woolly larch aphid than those from Poland, Central Austria and in high elevations in Western Austria. Italian, Swiss and low-elevation Western Austrian provenances had intermediate damage. Provenance IUFRO-53, from planted stands, exhibited only two per­ cent of trees heavily infested by the woolly larch aphid, while European larch from natural stands showed 4 to 65 percent. 58 Figure 8.— Sudeten larch was heavily attacked by the woolly larch aphid (Chernies strobilobius Kalt.). The clumps of foliage arose in response to the aphid damage. 59 Table 20.— Geographic variation in European larch with respect to percentage of trees heavily infested by woolly larch aphid, based on the data of IUFRO larch plantation in Hillsboro, New Hampshire, obtained in July 1958. Origin Region Elevation meters Heavily Injured Trees percent Sudeten Mtns., Czech. Tatra Mtns., Czech. Eastern Austria Italy Switzerland Western Austria Poland Western Austria Central Austria Central Austria 700 800-1000 550- 800 900-1400 550-1400 750- 900 170- 270 1050-1900 1100-1700 600- 950 55.5 42.0 41.0 21.5 20.5 18.3 15.8 14.5 12.0 10.7 * 5-Percent Significance Level* H 0 0 L L L L L L L H H • 0 0 L L L L Provenances from the regions marked " h " were significantly more attacked by the woolly larch aphid than those in regions marked t ! r tf it • 6. Damage by Other Agencies a. Porcupine Damage Porcupines (Erethion dorsaturn dorsatum Linnaeus) may cause con­ siderable damage in European larch plantations (35, 80). In winter when food is scarce, larch and other conifers are often girdled. Vigorous trees usually recover from partial damage. However, complete girdling or invasion of injuries by disease results in die—back of tops, development of crooks, and delayed height growth. 60 At the IUFRO larch plantation in New Hampshire, the author tallied the crooked trees whose damage was apparently caused by porcupines. Because of inadequate replication and because attack by porcupines were not random, genetic variation in susceptibility to porcupine injury was not established. In one corner of the plantation, within a radius of about 150 feet of porcupine dens, 40 percent of the trees were damaged. Plots in this area were unsuited for comparative height studies. The remainder of the plantation showed a relatively small number of trees attacked (an average of about 4 percent per plot). Also, the author scored the porcupine damage in a 21-year-old larch plantation at the Fox State Forest, Hillsboro, New Hampshire. Under similar vulnerability of porcupine attack, 43 percent of European larch trees were girdled while Japanese larch had only 13 percent of the trees damaged. b. Larch Canker in Europe Larch canker is caused by the fungus Dasyscypha willkommii (Hartig) Rehm. Its fructification covers the infected areas with red to dirty- white or yellow dots of less than one-eighth inch in size. causes a depression in the bark of the stem. thickness, The fungus As the tree grows in the infected areas on the flanks of the canker remain in bowl-like cavities. At later stages the wood is weakened to the extent that young trees may be killed outright and the older trees are pre­ disposed to wind or snowbreak (29, 35). 61 The disease is native in the Alps. Thence it spread to planted stands in other parts of Europe where it became the most important destructive agent of European larch (65). The fungus was probably introduced into the United States on the stock received from Scotland in 1904 or 1907. It was first found in this country in 1927 at the Harvard Forest *s plantation at Hamilton, Massachusetts, and was later reported from Ipswich, Massachussetts. The diseased trees were removed from these plantations in 1930. It has not been reported in this country since that time (29, 35). Larch canker attacks in various degree European as well as Japanese and Siberian larches. Kiellander reported the following data on susceptibility to larch canker (39): Provenances Studied number Species European larch Japanese larch Siberian larch Trees Attacked by Canker Dead Tops % per provenance % per provenance 19 2 1 29 to 92 3 to 6 43 3 to 86 1 3 I ran a MtTf test on his published data and found that provenance IUFRO-51 from Slovakia with 29 percent of attacked trees was signifi­ cantly less susceptible to larch canker than provenances the most heavily attacked. from Poland, European investigators believe that provenances Slovakia and Sudetenland are more resistant to canker than those from other regions (65). 62 7. Summary of the European Larch Racial Test Data Summary of the data on eight characteristics of European larch studied in the New Hampshire IlIERO larch plantation is represented as percents of the plantation means in table 21. Graphic illustration of characteristics of provenances from different regions is shown in figure 9. A key to the origins based on statistically proven differ­ ences is presented in table 22. Within-region variation was studied on the basis of combined New Hampshire and New York data. There were significant differences between provenances from low elevations in Western Austria in stem straightness. The provenance IUFRO-7 had significantly more crooks per 100 feet of stem than the IUFR0-4. No within-region genetic variations were detected in height performance. 63 Table 21.— Summary of the data of the European larch racial test (IUFRO) at the Fox State Forest, Hillsboro, New Hampshire. Origin Region, Elevation Crooks in Over Bare Lower Upper 5° Basal Aphid Inside Height DBH Stem Stem Lean Sweep Damage Branches - - - - - percent of the plantation mean - Switzerland 550-1500 meters 76 35 90 73 Italy 900-1400 meters 86 89 70 84 Western Austria 1050-1900 meters 93 93 108 Western Austria 750-900 meters 99 98 Central Austria 1100-1700 meters 92 Central Austria 600-900 meters 20 86 57 29 100 90 30 62 29 61 61 61 220 145 198 196 76 120 93 43 32 16 43 50 47 100 101 83 93 83 114 48 69 Eastern Austria 550-900 meters 94 95 80 96 87 76 171 148 Slovakia 800-1000 meters 116 103 58 90 43 78 176 189 Sudetenland 700 meters 126 124 128 159 193 101 232 199 Pol and 200-450 meters 128 127 110 156 220 161 66 110 Plantation mean * feet inches 18.6 2.42 n o .* 4.0 — — 60 _ - percent i of trees - - - - 26.7 25.6 Number of crooks per 100 feet of stem length. 36.6 23.9 23.6 6k CHARACTERISTICS OF E U R O P E A N VAR, PO LON IC A VARIETIES POLAND VAR. D ECI DU A C Z EC H OS LO V AK I A S UD ET EN LA ND ELE VATIONS LOW LOW IT A L Y SLOVAKIA E A S T E R N LOW LOW SMALL C R OOK S IN UPPE R ST E M LARGE C R OOK S IN LOWER STEfc LOW BA SAL SWEEP HI GH LOW END BAR / f L£,N ; | : SWI TZERLAND C LEG I I L ± LARCH *|* WOOL L Y LARCH "J* DEFOLIATED I NSI DE BRANCHES H EIG HT* TOTAL A P HI D I A VE RA GE SUPERIOR H EIG HT DIAMETER Figure 9.— Graphic summary of the data by characters and regions, vakian origins were the best with respect to production of quality timber. Slo- 65 Table 22. 1. Rapid height growth (1.8 to 2.0 feet per year) 2. 2. 1. Classification of origins by statistically proven (at the 5-percent level) differences. Slow Large diameters (3 inches in 12 years) j many trees leaning (50 to 56 percent) and many trees with small crooks in upper stem (41 to 53 percent) 3. Few trees attacked by woolly larch aphid (16 percent) POLISH ORIGINS 3. Many trees attacked by woolly larch aphid (46 percent) SUDETEN ORIGINS Moderate diameters (2.5 inches in 12 years), few leaning trees (11 percent) and few trees with small crooks in upper ..................... SLOVAKIAN ORIGINS stem (24 percent) to moderate height growth (1.2 to 1,6 feet per year) 3. Many crooks in lower stem (9 per 100 feet of stem), many trees with basal sweep (72 percent), many trees leaning (49 percent) . . . . LOW-ELEVATION WESTERN AUSTRIAN ORIGINS 3. Few crooks in lower stem (less than 5 per 100 feet of stem), few trees with basal sweep (less than 42 percent), few trees leaning (less than 23 percent) 4. Few trees lean ( 5. 5. 4. (21 to 23 percent) Few trees with basal sweep (28 to 42 percent) 6. Many trees attacked by woolly larch aphid (41 percent). . . EASTERN AUSTRIAN ORIGINS 6. Few trees attacked by woolly larch aphid . . LOW-ELEVATION CENTRAL AUSTRIAN ORIGINS Very few trees with basal sweep (8 percent) SWISS ORIGINS Very few trees lean (4 to 8 percent) ITALIAN, HIGH- ELEVATION CENTRAL AND WESTERN AUSTRIAN ORIGINS 66 VIII. RELATIONSHIP BETWEEN GENETICS, TAXONOMY AND DISTRIBUTION Relationship Between Genetics and Taxonomy European larch consists of two varieties, Larix decidua var. decidua and L. decidua var. polonica. Genetical characteristics of populations belonging to these varieties are illustrated in figure 9. Statistical analysis showed that provenances from Poland, Sudetenland and Slovakia, which* belong to the L. d. var. polonica, had significantly greater heights than those from other regions belonging to L. d. var. decidua. Other characteristics studied were distributed in both varieties. Provenances from Slovakia of one variety differed from their closest neighbors, the East Austrian provenances of another variety, only in two characteristics, height and lean. 2. The Genetic Variation Patterns Summarized by Regions The pattern of geographic variation in European larch was largely systematic. Characteristics of provenances originating from the same geographic or altitudinal regions were similar. Evidently most of the geographic variation has taken place because of selection pressure. A discontinuity in genetical characteristics was observed between PolishSudeten origins and those from Slovakia. Altitudinal differences were evident in the provenances from Central and Western Austria. Discontinuities in some characteristics between provenances from different regions were not evident because the data were not precise. 67 a. Polish Larch Polish larch grows in scattered stands in central and southern Poland. It was represented in this study by four provenances which grew very rapidly and had large diameters (tables 21 and 22). Many trees exhibited lean, basal sweep and small crooks in the upper stem, but had only a moderate number of crooks in the lower stem. suffered only slightly from the woolly larch aphid. point of cellulose production, They From the stand­ these Polish provenances were best of all origins studied. The rapid growth, resistance to late spring frosts, and resist­ ance to larch canker have been reported as characteristics of Polish larch by several European investigators (2 3 , 33, 36, 73). However, most of these observations were based on unreplicated tests. b. Sudeten Larch The Sudeten population of European larch has its isolated range in Sudetenland, Czechoslovakia. two provenances. It was represented in this study by They were similar to the Polish provenances in all characteristics except in their greater susceptibility to the woolly larch aphid. Provenances from Sudetenland have frequently been represented in European tests. 19, 24, 45, 73). All these agree regarding its rapid growth (9, 10, 18, It was also observed that Sudeten larch, in compari­ son to larches from the Alps, was less resistant to drought but more 68 Figure 10.— Polish larch (IUFRO-38) in New Hampshire. Twelve-year height averaged 24 feet and diameter at breast height 3 inches. 69 resistant to larch canker; it had wider crowns, smaller branch diameters and paler, more pendulous twigs (18, 23, 45, 65). c. Slovakian Larch Slovakian larch, which grows in the Tatra and other mountains in the eastern part of Slovakia, was represented in this study by two prove­ nances. They grew rapidly in height but had smaller diameters than Polish or Sudeten larches. in the lower stem, or lean. woolly larch aphid. Only a few trees had basal sweep, crooks However, they were heavily attacked by the Considering both volume and quality.production, they seem the best origins to recommend for the Northeast. Siman suggested that Slovakian population of European larch consists of three ecotypes (77, 78). this conclusion. I have not been able to find the basis for Earlier investigators observed that Slovakian larch is slightly slower growing than either Polish or Sudeten larches, but more rapid than those from the Alps (67, 6 9 , 7 2 , 73). Slovakian larch in Sweden and elsewhere showed a moderate resistance to larch canker (64, 78). d. Low-Elevation (550-950 meters) Larch From Austria The low-elevation larches from Austria represented in this study originated from the northern and eastern foothills of the Alps. They grew significantly slower than provenances from either Poland, Sudeten­ land or Slovakia. 70 Three provenances were represented from low elevations in Western Austria. On the average they had moderate growth, but exhibited many undesirable characteristics, such as very high number of crooks in the lower stem, many leaning trees with basal sweep and many trees with small crooks in the upper stem. Provenances from low elevations in Eastern and Central Austria showed a similar growth rate to those from low elevations in western Austria, but had less crooked lower stems, and less basal sweep or lean. Eastern Austrian provenances were more susceptible to the woolly larch aphid than those from Central Austria. Rubner considered the low-elevation larch in Austria (and other parts in the Alps) to be a relict of the glacial period and called it low-alpine ecotype (66). Observations were also made in Europe that this larch flushes late, had a long height growth period (30-90 days), but is not exposed to late frosts (6). In general, provenances from Eastern Austria have demonstrated better growth rate than those from the western parts of the Alps (73). A great variability was observed in low-elevation provenances with respect to stem straightness (20). e. High Elevation Larch From Austria and Italy (925-1900 Meters), And Larch From Switzerland Provenances from high elevations, mainly in the western and southern Alps, teristics, grew slow. They showed, however, desirable stem charac­ such as few trees with lean, crooks or basal sweep. Only a few trees were attacked by the woolly larch aphid or defoliated by late 71 snow. Exceptions from the above characteristics were as follows: Italian provenances had a moderate number of trees with basal sweep and Swiss provenances had a moderate Rubner reported number of leaning trees. that origins of medium elevations in the Alps, called "high-elevation" provenances in this study, were highly suscep­ tible to larch canker and had better stem forms and better height growth than provenances from still higher altitudes (above 2000 meters) (66). These origins are found to be more resistant to drought than provenances from low elevations (24), they flush early and are exposed to late frosts (6, 44), especially those from the French Alps (_7) . The high-elevation origins have a 50- to 60-day period of height growth which is much shorter than that in low-elevation provenances (6, 44). It was also observed that these provenances exhibited heavier branches than the low-elevation origins (44). 72 IX. LARCH PLOTS OF MISCELLANEOUS ORIGIN American IUFRO larch plantations in New Hampshire and New York included 11 miscellaneous seed lots originating from planted stands. None of these larches were superior in height growth to the most rapidly growing provenances of European larch (table 24). There is no evidence that European larch planted outside of its natural range may have developed to an independent race. Some investigators inter­ preted that "Scottish larch", that was introduced to British Isles about 300 years ago, is a separate race (65, 75). However, these plantings were established with different genetical stock on differ­ ent dates (51), and consequently, one stand of European larch in Scotland may differ from another stand. Edwards reported that "Scot­ tish larch" varies in height growth as well as in cone characteristics (18, 19). IUFRO-53 (Scotland) .— This provenance from a planted stand in Aldrauthy, Scotland, grew slowly (table 24). However, it had some desirable characteristics such as few trees with crooks, lean, or basal sweep and resistance to the woolly larch aphid (2 percent of trees attacked as compared with 4 to 65 percent attacked in the 31 provenances of European larch from natural stands) (figure 11). These characteristics indicate that the parent stand of the IUFRO-53 was probably imported from a high elevation in Austria or Switzerland. IUFRO- 45 and IUFRO-49 (Moravia).— These origins from planted stands in Moravia, Czechoslovakia, grew rapidly (table 24) and 73 Table 24.— Twelve-year heights and diameters of provenances from planted stands compared with those of rapidly growing provenances of European larch from its natural range, combined New Hampshire and New York data. Provenance IUERO No. 47 42 ______ Height_______ Origin of Seed Sources Difference Natural Planted Levels at St and s ________ Stands_____ Average 5 Percent* feet Sudetenland Poland 45 49 38 43 51 Moravia Moravia Poland Pol and Slovakia Pol and 35 4 1 6 * 21.5 21.5 Western Austria Central Austria Eastern Austria S S Average d.b.h. inches 2.95 3.05 S • s 2.90 2.95 20.0 19.4 18.8 s S s 2.90 2.60 2.40 18.2 s 2.50 18.0 17.9 17.8 S S s 2.85 2.70 2.60 21.2 20.3 28 L. leptolepis 17.7 s 2.65 31 36 34 30 Germany Poland Poland Germany 17.6 16.9 16.7 16.5 s I I I • 2.30 2.20 2.15 2.35 55 L. eurolepis 15.8 I I • 2.30 29 53 Germany Scotland 15.8 12.9 I I I I I . . • 2.25 1.95 Provenances marked "s” are significantly superior in height growth than those in the same column marked "i” . 7k Figure 11.— European larch from planted stands in Scotland (IUFRO-53). It was relatively slow growing, but had straight stems and was highly resistant to the woolly larch aphid. 75 exhibited characteristics most similar to provenances of Polish-Sudeten origin. This finding agrees with Svoboda’s report in which he stated that the majority of the larch planted in Moravia are of Silesian (Poland) origin (84). IUFRO-28 (Denmark) .— This study included one seed source of Japanese larch (Larix leptolepis Sieb. & Luce (Pilger)) from a planted stand in Denmark. It was represented in pure plots in both American IUFRO larch plantations and in one plot at the New Hampshire planta­ tion it grew mixed at random with European larch (IUFRO-44). Japanese larch was 100 percent resistant to the woolly larch aphid (Chermes strobilobius Kalt.). When planted in pure plots, it had good height and diameter growth but was not superior to the best European larch (table 24). In 1958 I identified the same origin of Japanese larch (15 trees) being mixed within a plot of European larch, IUFRO-44 (23 trees). Analysis of variance showed that Japanese larch in this random mixture had significantly (at the 5-percent level) greater 12-year heights (averaged 25.1 feet) than European larch (averaged 18.9 feet). IUFRO-55 (Sweden).— This seed source originates from an old stand of European larch in Vinsingso, Sweden. Swedish records indicate that this stand was probably raised from seed bought in Scotland (76). Schmidt further states that the parental stand in Sweden was subject to pollination by Japanese larch (72). Accordingly, this seed lot may have included some of Fx hybrids (Larix x eurolepis Henry) between European and Japanese larches. were resistant to Meria. Kiellander determined that some trees I found some resistant to Chermes. 76 X. APPLICABILITY OF THE PRESENT DATA Applicability of Present Data in Other Regions A study was made to determine the reaction of the same provenances under different environmental conditions. questions: This was done to answer two (1) how applicable are the height data from New Hampshire and New York in other parts of the United States, and (2) how applica­ ble is the European data to our conditions? In order to reach these objective^, correlation analyses were made between heights of the same provenances planted in Germany, Great Britain, Italy, Sweden or Switzerland, and the United States. One of the best designed experiments in Europe was performed in Scotland (experiment TX in table 4). The 9-year heights reported from Scotland (93) and 12-year heights from the American IUFRO larch plantings showed very strong correlation (r = .833, significant at the 1-percent level with 13 d.f., table 23). Another well replicated experiment was established in northeastern Germany (experiment X in table 4) (72). Ten-year heights from this experiment and 12-year heights from the American plantings showed very strong correlation on the basis of eight provenances analyzed (r = .925, significant at the 1—percent level with 6 d.f., table 23). Similar results were obtained with 11-year heights from Italy (experiment -XF-in table 4) and 12-year heights from America. Correla­ tion analysis showed a very strong correlation (r = .874, significant at the 1-percent level with 12 d.f., table 23). 77 Table 23.— Correlation between average heights of IUFRO larch prove­ nances tested in northeastern United States (New Hampshire and New York) and the same provenances tested in Europe. Provenance IUFRO No. 47 45 49 51 35 4 1 6 31 14 7 23 36 34 12 30 55 29 24 8 18 5 21 27 53 26 at 12 Years U.S.A. 21.5 21.2 20.3 18.8 18.2 18.0 17.9 17.8 17.6 17.6 17.2 16.9 16.9 16.7 16.6 16.5 15.8 15.8 15.1 15.0 14.5 14.1 13.4 13.0 12.9 11.4 Correlation with the American height data, Degrees of freedom Average Heights at 9 Years at 10 Years S. Scotland NE . Germany ____ — — — 15.4 — — 14.4 — — 14.1 13.5 12.5 — 14.1 13.5 12.5 13.8 13.8 13.1 — 11.5 12.8 — — 12.8 — 9.5 =■ .833** 13 — at 11 Years N . Italy 13.1 13.3 — — 23.3 23.6 — — — 24.9 — — — — — — 22.6 22.9 — — — — 20.3 12.3 — — 12.2 11.1 — — — 11.3 — 12.3 11.8 11.9 — — 11.7 — — 18.4 14.1 11.1 11.5 10.0 — 9.0 .925** .87 4** — — 6 ** Very significant correlation (at the 1-percent level). 12 78 Different results were obtained in correlating the American data with those from Sweden and Switzerland. The IUFRO larch plantation in southern Sweden (experiment XV in table 4) was very heavily attacked by larch canker, Dasyscypha Willkommii (Hartig) Rehm. (78). A high percentage of trees had dead tops and height data were of limited value. Correlation between 12-year heights in American and Swedish plantations was not strong (r = .349, not significant at the 5-percent level with 15 d.f., table 4). Another non-significant correlation was obtained between 8-year heights in Switzerland (experiment XVI in table 4) (54) and 12-year heights in the United States (r = .483, not significant at the 5-percent level with 4 d.f.). This poor correlation was probably because the Swiss data were based on an unreplicated data. The IUFRO larch plantations in Hillsboro, New Hampshire, and Barkersville, New York, Included 25 provenances of European larch from native stands which originated from the same seed sources. Provenances which grew faster in New Hampshire also grew faster in New York. conclusion was based on two types of analyses of height data. This Correla­ tion between 12-year heights from New Hampshire and New York planta­ tions was very strong (r — .595, significant at the 1—percent level with 23 d.f.). Analyses of variance showed that between-provenance differences were very significantly greater than the provenance x plantation interactions (F = 4,02, significant at the 1—percent level with 24 d.f. for provenances and 24 d.f. for error). The results of the above studies showed that consistency of height performance was not dependent on the distance between locations. The height data on racial variation in European larch, obtained in properly 79 conducted experiments in Europe, can be trusted in the United States. Also, data obtained in the northeastern United States will probably be applicable in Michigan and other northern parts of this country. This information on consistency of height performance in differ­ ent environments is important not only in genetics research but also in nutrition experiments. Results obtained by testing one variety are probably applicable to other varieties of European larch. 2* Applicabi1ity to Future Production of Hybrids The crossability pattern of larches is not finally determined. According to the present classification, the genus consists of 15 different varieties or species (tables 2 and 3). There are 105 possible intervariatal and interspecific crosses (excluding reciprocals)* At the present, however, nations. there are only 10 known interspecific combi­ In five of them one parent species was European larch. Crosses between European larch and other larches have been reported as follows (94): Larix decidua Crossed With Larix Larix Larix Larix Larix Larix leptolepis (Japanese larch) gmelini (Dahurian larch) 1 aricina (Tamarack) occidentalis (Western larch) sibirica (Siberian larch) mastersiana (Master's larch) Scientific Name of Hybrid Larix x eurolepis Henry — Larix x pendula (Toland)Salisbury — 80 The most publicized hybrids are those between European and Japanese larch. These hybrids occurred naturally at Dunkeld, Scotland, and other locations in Great Britain and are easy to produce artifically (42, 51). Presently Hybrid Larch or "Dunkeld Larch" occupies an area of 6,000 acres in Great Britain (51). Larsen tested L. x eurolepis hybrids in Denmark and found that they grew faster than Japanese larch (42). In Great Britain nearly all "Hybrid Larch" plots grew better than European larch and were equiva­ lent or better in height growth and volume increment to Quality Class I Japanese larch. They were practically resistant to larch canker and "die-back" (51). These hybrids were tested in many other European countries, and in all cases they grew better than the parent species (30, 41, 94). There are few American tests of "Dunkeld Larch", from Dunkeld, tion. a seed source Scotland, which included some hybrids of unknown genera­ Paton made a study in Wooster, Ohio, and found that "Dunkeld" larches grew slightly better than the parent species (56). Cook re­ ported on good performance of larches from Dunkeld seed in Renselaer County, New York (16). Littlefield found also that 5-year-old seed­ lings of supposed hybrids grew slightly better than either European or Japanese larches in Montgomery County, New York (49). The present crosses between European and Japanese larches were made by using parents of unknown origin. Selection of parents from populations with most desirable characteristics may improve the qualities of hybrids considerably. This study showed a great 81 variability of European larch in height growth, stem straightness, resistance to the woolly larch aphid and other characteristics (table 21, figure 9). Naturally, this variation should be taken into con­ sideration when selecting the parents in future production of hybrids. a. Japanese Larch as a Crossing Partner Japanese larch (Larix leptolepis (Sieb. & Zucc.) Pilger) is a native of the mountains of central Honshu, where it grows in pure of mixed stands at elevations of 4,000 to 9,000 feet (figure 2, tables 2 and 3). Its natural range is small, being encompassed within an area of about 140 miles square and composed of several distinct popu­ lations. Some of these populations are only a few acres in size and the largest, are only a few miles across. larch in: It is superior to European lower nutrient requirements and is more resistant to attack by Dacyscypha willcomii, Adelges virilis, Argysthia laevigitella, Theniotrips laricivorus, Meria laricis, Chermes strobilobius, and squirrels. However, European larch is less sensitive to drought; early and late frosts, and breakage by ice and snow. Possibly it is more resistant to Phomppsis pseudotsugae (1_, 6, 14, 39, 74, ai, 92). In hybridizing European and Japanese larches, it is important to select not only the best parent types in European larch but also in Japanese larch. Very little work has been done on racial variation studies in Japanese larch. Langner reported green weights of 1-year- old seedlings from 25 different provenances (40). depended largely on the differences in spacing. However, these Seven seed lots of 82 Japanese larch, from the native stands were studied at Michigan State University, East Lansing, M i c h i g a n . T w o - y e a r data showed differences between provenances in fall-color, branching, susceptibility to early frost and height growth. The provenance from Kumanshiro, Tsumakoi, Gumma (63° 3 8 1 30nN and 138° 29* 3 0'fE, altitude 1750 meters) combined two desirable characteristics: resistancy to early frosts. rapid height growth and moderate The pattern of variation in Japanese larch appeared to differ from thp/t found in wide-ranging species with large continuous populations. Large differences appeared among provenances but these were not related to latitude, longitude, or altitude of origin. In other words, the variation pattern in Japanese larch was not as systematic as in European larch. 1/ Wright, J. W. and Genys, J. B. Unpublished data. 83 XI. FINDINGS APPLICABLE IN FUTURE PROVENANCE STUDIES • Need of Replication Many past experiments attempting to determine the racial differ­ ences between provenances gave only limited results because of poor or no replication. They had a low value even as demonstration plots, because no certainty existed whether differences were due to environ­ ment or due to genetics. A great majority of present investigators realize that differences in growth may occur because of soil variation, yet a few fail to consider the remaining complex of environmental factors. The present study shows that statistical analyses of height and other data reveal large "least significant differences" (table 15). It was impossible to determine which particular provenance, was outstand­ ing in one or another characteristic or to find the within-region vari­ ation patterns. These and other difficulties arose mainly because provenances were not adequately replicated. Only a few recommendations are published regarding the number of replications in forest genetic studies. Recently Wright and Freeland reported that statistical efficiency can be increased by replicating the sources as many times as possible and by using as small plots as is practical (95). 8U Reliability of Early Height Measurements In the New Hampshire IUFRO larch test, heights were measured at ages 4 (on 2—2 transplants), 7 and 12 years. Analyses of variance of provenances from different regions showed that 12-year data gave the highest and 4-year height data gave the smallest F value. The 12-year data showed a higher number of origins that had significantly (at the 5-percent level) inferior height growth to the two best growing origins than those of earlier measurements as follows: Degrees of Freedom Age of Trees When Measured years Between Regions number Error number 4 7 12 9 9 9 17 20 19 Computed F value 4.70** 7.87** 11.70** Origins Significantly Inferior to Two Origins With the Greatest Heights number 4 5 7 ** Very significant differences (at the 1-percent level) found among origins Correlation analyses showed that 4-year heights and 7-year heights were very strongly correlated (r = .502, significant at the 1—percent level with 35 d.f.) and 4-year heights were very strongly correlated with 12-year heights (r = .745, significant at the 1-percent level). It is concluded that early measurements can be applied to fore­ cast the approximate variation pattern. However, the measurements in oldGr stands are more desirable and exhibit a better statistical efficiency. 85 ^ c^ency o:^ Statistical Analyses by Using Different Sample Sizes Analyses of variance of 12—year heights from New Hampshire and New York plantations were made by using different statistical units representing plots. Most differences between provenances were detected when means of all trees per plot were used instead of the four largest trees per plot. No differences between provenances were detected on the basis of one largest tree per plot used as presentatives. Results are shown in the following tabulation: Nature of Sample Representing a Plot Number of Provenances Studied Computed F Value One largest tree 25 1.22 Mean of four largest trees Mean of all trees 19 2.24 Not significant at 5 percent Significant at 5 percent 25 4.02 Significant at 1 percent Level of Significance of F From the practical point of view, the means of four largest trees per plot could be used if a loss of statistical efficiency and a gain of time by measuring smaller number of trees are properly balanced. If the test plots are large, naturally, the sample of trees should be selected at random. 86 LITERATURE CITED 1. AIRD, P. and STONE, E. L. 1955 Soil characteristics and growth of European larch in New York. Jour. For. 53:425-429. 2. BALDWIN, H. I. 1958 European larch in New Hampshire. 15-21. N. H. Cons. Mag. 59: 3. 1956 The period of height growth in different provenances of European larch. IUFRO, 12th Cong., Oxford, 6 pp. 1949 Four-year growth of European larch of different origins. Fox Forest Note 41, 1 p. 4. 5. 6. 7. 8. 9. BOE, K. N. 1958 Silvics of western larch. No. 16, 17 pp. U.S. Dept, of Agr. Misc. Pub. BURGER, H. 1935 Einfluss der Herkunft des Samens auf die Eigenschaften forstlicher Holzgewachse. Die Larche. Mitt. d. schweiz. Anst. f.d.forstl. Versuchsw. 19(1):104-136. BOUVAREL, P. 1957 Appel pour la selection de Melezes d ?elite. tiere Francaise 7:599-600. j ____ and LEMOINE, M. 1958 Races de m^l^ze et grosseur des granes. Francaise 10:348-350. j CIESLAR, A. 1914 Studien liber die Alpen-und Sudetenlarche. ges. Forstwesen 40:171-184. Revue Fores- Revue Forestiere Centralbl.f ,d. 10 1907 Die Bedeutung klimatischer Varietaten unserer Holzarten fUr den Waldbau. Centralbl.f.d. ges. Forstwesen 33:1-19. 1899 Neues aus dem Gebiete der forstlichen Zuchtwahl II. Larche. Centralbl.f.d.g. Forstwesen 25:99-111. 11. Die 87 12. CIESLAR, A. 1895 13. t3ber die Erblichkeit des Zuwachsvermogen bei den Waldbaumen, Centralbl.f.d. ges. Forstwesen 21: CCX)K, D. B. 1955 Improvement of larch by selection of species and geographic races. Soc. Amer. Foresters, N. Y. Sec., N. Y. Forester 12:11-13. 14. 1954 Susceptibility of European larch to red squirrel damage. Jour. For. 52:491-492. 1946 Further notes on a larch plantation. 1942 Characteristics of the Dunkeld larch and its parent species. Jour. For. 40:884-885. 15. Jour. For. 44(1):58. 16. 17. 18. CRAIGHEAD, F. C. 1950 Insect enemies of eastern forests. Dept. Agr., 679 pp. Misc. Pub. 657, U. S. EDWARDS, M.,V. 1954 Scottish studies of the provenance of European larch. Interl. Union Forest Res. Organ. Cong. Proc. 11(1953): 432-437. 19.__ _______ 1952 and PINCHIN, R. D. Provenance studies. 43-57. Forestry Comm. Annual Rep t . 1951/52: 20. ENGLER, A. 1954 Ergebnisse eines Versuches uber die Vererbarkeit der WuchsForm bei der Larche (Larix decidua Miller), mit einem Beitrag zur Methodik der Erfassung von Merkmalen. Internatl Union For. Res. Organ. Cong. Proc. 11(1953):437-445. 21. _____ 1905 22. Einfluss der Provenienz des Samens auf die Eigenschaften der forstlichen Holzgewachse. Mitt. d. schw. Centralanst. f.d. forstliche Versuchswesen 8:81-236. FENAROLI, L. 1936 II larice nelle Alpi Orientali Italiane. Tipografia Mariano Ricci-Firenze-Via S. Gallo, 31, 270 pp. 88? 23. FISCHER, F. 1952 Einige Ergebnisse aus dem Internationalen Larchenversuch 1944. Mitt. d. schweiz. Anst. f.d. fortl. Versuchsw. 28:355-407. 1950 Die Jugendentwicklung von Larchen verschiedener Herkunft auf verschiedenen Standorten. Mitt. d. schweiz. Anst. f.d. forstl. Versuchsw. 26(2):469-497. 24. 25. FOURCHY, P. 1952 Ecologie du meleze particulierement dans les Alpes Franchises. Ecole Nationale des Eaux et Forests, Nancy. 137 pp. 26. G&UMANN, E. 1948 Einfluss der Meereshohe auf die Dauerhaftigkeit des Larchenholzes. Mittl. d. Schweiz. Anstalt f.d. forstl. Versuchsw. 15:327-391. 27. GEYER, H. F. 1935 Zur Systematic der Gattung Larix. dendrol. Gessellschaft 35:1-7. 28. 29. 30. GttHRN, U. 1956 Proveniensforsog med laerk. I 23:1-124. Mitt. d. deutschen Det Forsogsvaesen in Danemark, HAHN, G. G. 1940 European larch canker. Important Tree Pests of Northeast, Massachusetts Forest and Park Assoc., Boston, 140-143. p 5 HEIDEBRAN^, v., B. *tnd PFLUG, W. 1949 Vergleichende Beurteilungen von L’ a rchenkreuzung auf der Versuchsfl’ ache des Forstamtes Gahrenberg nach einer Aufnahme im Herbst 1949. Seminar at the Institute for Forest Regulation, Hann. MUnden, Germany. 5 pp. 31. HEIKINHEIMO, 0. 1956 Ergebnisse von einigen Anbauversuchen mit fremdlandischen Holzarten in Finnland. Communicationes Instituti Forestalls Fenniae 43. 32. HERMANN, H. 1933 Die Sudetenl'arche. 33. HERMER 1941 Tharandt. forstl. Jahrb. 84:363-431. Der Wald im Weichsel-und Wartheraum. 216-256. Forstarchiv 12: 89 34. HOWARD, A. I.. 1920 35. A manual of the timbers of the world, their characteristics and uses. MacMillan and Co., London. 446 pp. HUNT, S. S. 1932 European larch in the northeastern United States. For. Bull. 16, 45 pp. Harvard 36. HUNTER, B. T. 1948 The Polish larch (Larix decidua var. polonica, Ostenfeld; Larix polonica Raciborski). Scot. For. 1:21-25. 37. JANKAUSKAS, M. 1954 Maumedziai Lietuvos miskuose ir parkuose ir jiems auginti perspektyvos. Valst. Polit. ir Moksl. Literat. Leidykla, Vilnius. 255 pp. 38. KALELA, A. 1937 Zur Synthese der experimentellen Untersuchungen uber Klima-Rassen der Holzarten. Reprint from MetsUtieteelisen Tutkimuslaitoksen Julkaisula, Helsinki. 434 pp. 33. KIELLANDER, C. L. 1955 Report on the development of the international provenance test of larch in Sweden. For. Tree Breed. Inst., Ekebo, K&llstorp, Sweden. 4 pp. 40. LANGNER, W. 1958 Planung und erste Ergebnisse eines Japanlarchen-Provenanzversuches mit zugleich zlichterischer Zielsetzung Centralbl. f.d. ges. Forstwesen 75:168-196. 41.__ _______ 1957 Gedanken zur Larchenzuchtung auf Grund der Ergebnisse des BUndiger L’ a rchenkreuzungsversuches. Rep. of 5th meeting of Assoc, f. For. Genetics and For. Tree Breeding. Silvae Genetica 6:160-161. 42. LARSEN, C. S. 1956 Genetics in silviculture. Lawn, N. J. 224 pp. Essential Books, Inc., Fair 43.__ ________ 1937 The employment of species, types, and individuals in forestry. C. A. Reitzer, Copenhagen. 154 pp. 44. LEIBUNDGUT, H. and KUNZ, R. 1952 Untersuchungen Uber europ’ a ische LUrchen verschiedener Herkunft. Mitt, schweiz. forstl. Versuchsw. 28:408-496. 90 45. LIPTAK, V. 1951 Skumanie rosovej a proveniencnej otazky cerv smereka na Medzinarodnej pokusnej ploche na Podbansku. Poliana 7:158-170. 46. LITTLE, A. L., Jr. 1953 Check list of native and naturalized trees of the United States (including Alaska). U. S. Dept. Agr. Handb. 41, 472 pp. 47. LITTLEFIELD, E. W. 1939 Comparative height growth of different larches five years after planting. New York State Cons. Dept. Note on For. Invest. No. 27. 2 pp. 48. _______ and ELIASON, E. J. 1956 Report on an experimental plantation of several species of larch in New York. Zeitschrift fur Forstgenetic und Forstpflanzenzllchtung 5:166-169. 49. and Observations on a plantation of Dunkeld hybrid larch in New York. Jour. For. 36:1188-1192. 1938 50. McCOMB, A. L. 1955 The European larch: its races, site requirements and char­ acteristics. For. Sc i . 1:298-318. 51. M c D o n a l d , f . et ai. 1957 Exotic forest trees in Great Britain. Bull. 30. 167 pp. 52. MORANDINI, R. 1956 II larice nella Venecia Tridentina. For. Commission Firenze. 270 pp. 53. 1956 Experiences internationales sur les races du meleze (1944). Notes sur les premiers resultats obtenus en Italie. IUFRO 12th Congress, Oxford. 7 pp. 54. OJI INSTITUTE FOR TREE IMPROVEMENT 1953 Forests in Japan. 32 pp. 55. OSTENFELD, C. H. and LARSEN, S. C. 1930 The species of the genus Larix and their geographic dis­ tribution. Det. H g l . Danske Videnskabernes Selskab. Biol. Meddelser. No. 9. 101 pp. 56. PATON, R. R. 1944 Dunkeld larch in Ohio. Jour. For. 42:452-453. 91 57. 58. 59. PAULEY, S. S. et al. 1955 Seed source trials of eastern white pine. 244-256. PINCHOT, G. 1908 Alpine larch. 2 pp. For. Sci. 1: U. S. For. Serv. Silvic. Leaf1. No. 35, POLUNIN, N. 1959 Circumpolar Arctic flora. 514 pp. The Clarendon Press, Oxford. 60. REHACKOWA, O. 1954 Untersuchungen uber die j'ahrlichen Ver'dnderungen des Keimprocentes und der Keimgeschwindigkeit von Fohren und Larchensamen. Inst. f. Waldbau, zUrich. 61. REHDER, A. 1949 Bibliography of cultivated trees and shrubs hardy in the cooler temperature regions of the Northern Hemisphere. Jamaica Plain, Mass., Arnold Arb. of Harvard Univ. 825 pp. 62. 1940 63. Manual of cultivated trees and shrubs hardy in North America. The MacMillan Co. , New York. 996 pp. ROE, E. I. 1957 Tamarack. Lake States For. E xpt. Sta. Sta. Paper 52. 22 pp. 64. ROHMEDER, E. 1953 Der unterschiedliche Keimverlauf von Samen der europ'dischen und japanischen L’ drche. Forstwiss. Centralbl. 72:210-220. 65. _______ 1959 66. 67. Verlag Paul Parey, RUBNER, K. 1954 Zur Frage der Enstehung der alpinen L'archenrassen. Forstgenet und PflanzenzUcht. 3:49-51. ________ 1951 Die Tatralarche. 6 8 . ______ 1943 69. and SCH&NBACH, H. Genetic und Zlichtung der Waldbaume. Hamburg. 338 pp. Zeitsch. Forstw. Centralbl. 70:129-161. Das Areal der Sudetenl'arche. Thar. For. Jahrb. 94:34-99. ________ 1938 Die Ergebnisse zweier L’ a rchenherkunf ts-Versuche im Tharandter Wald. Thar. For. Jahrb. 89:465-491. 92 70. RUBNER, K.and SVOBODA, P. 1944 Untersuchungen on L&rchen-Zapfen verschiedener Herkunft. Intersylva 4:121-146. 71. SCAMONI, A. 1951 Ein L&rchenherkunftsversuch in Eberswalde und seine weitere Entwicklung. zttchter 21:36-38. 72. SCHMIDT, W. 1955 L’ archenherklinf te und Enrolepis-Bastarde im northwestdeutschen KUstengebiet (VersuchsfI’ dche Bremervorde) . Allg. Forst-und Jagdztg. 126:24-28. 73. SCHOBER, R. 1958 Ergebnisse von Larchen-Art und Provenienzversuchen. Genetica 7.: 127-154. Silvae 74. 1953 Die japanische L*arche. a. M . , 212 pp. 1949 Die Larche. Sauerl’ a nder's Verlag, Frankfurt 75. 76. 77. Verlag Schaper, Hannover. 285 pp. SCHOTTE, G. 1917 L'arken och dess betydelse for Swensk skogshusholling. fran Statens Skogsforsoksanstalt., 13/14:529-840. SIMAK, M. 1958 Den slovakiska 1‘ dr ken. 1:123-136. Sartryck ur Swenska Skogsv. Medd. Tidskr. 78. SIMAN, K. . f t / 1943 Prirozene rozsireni a promenlivost modrinu obecnego Larix decidua Mill. Sbor. Ceske akadenie zem. XVIII. 79. SNEDECOR, G. W. 1950 Statistical methods. The Iowa State College Press. 585 pp. 80. SPURR, S. H. 1956 Plantation success in the Harvard forest as related to planting site and cleaning. Jour. For. 54(9):577-579. 81. STOECKELER, J. H. 1955 European larch seed sources compete successfully with tamarack during 5-year test in northeastern Wisconsin. U. S. For. Serv., Lake States For. Expt. Sta. Tech. Note 440. 2 pp. 93 82. STONE, E. L . , Jr. 1957 British yield tables for European and Japanese larches in New York. Agronomy Paper 397. 4 pp. 83. SUDWORTH, G. B. 1918 Miscellaneous conifers in the Rocky Mountain region. Dept. Agr. Bull. 680. 44 pp. 84. SVOBODA, P. 1947 Hrotowice modrid Zpr. vyskum. 36-110. U. S, U s t . lem. C. S. R. No. 1: 85. SZAFER, W. 1913 Przycznek do znajimsci modrzevi eurazyatyckich ze szezegolnem uwzglednieniem modrzewid w Polsze. Kosmos XXXVIII. 86. TSCHERMAK, I*. 1935 Die natUrliche Verbreitung der Larche in den Ostalpen. Verlag Springer, Wien. 361 pp. 87. VARMA, J. C. 1949 A study of the variations of morphological characters within tree species with special reference to European larch of different provenances (Larix decidua Mill.). Forestry 23:20-30. 88. VEEN, B. 1954 Report on the test areas of the international provenance tests with pine, spruce, and larch of 1938-39 and 1944-45, and suggestions for future treatment and assessments. IUFRO Cong. Proc. 11 (1953):536-551. 90. VINCENT, G. 1958 Larchentypen in Mitteleuropa. Forstwesen 109:506-519. Schweiz. Ztschr. f. 91. VITE, J. P. 1953 Die artbedingte Resistanz verschiedener Larchen gegen Theniothrips laricivories. IUFRO Congress, 1953, 736-745. 92. WOOD, R. F. and EDWARDS, M. V. 1954 Provenance studies, European larch. Comm. (1952/53):40-41. 93. Report For. Res. WOODBURN, D. A. and KENNEDY, J. M. 1956 Drummond Hill Forest. International Larix decidua prove­ nance experiment. IUFRO 12th Congress (1956). Central and Western Scotland Tour Guide. No. 2. 9k 94. WRIGHT, J. W. 1960 Genetics and forest tree improvement. 515 pp., typescript. Submitted to F. A. 0. for publication. 95. _______ 1959 96. and FREELAND, D. F . , Jr. Plot size in forest genetics research. The Michigan Academy of Science, Arts and Letters 44: 1 7 7-182 ZIMMERLE, H. 1941 Beitrage zur Biologie der europ&ischen Larche in Wlirttemberg. Mitt, der Wurttemberg Forstl. Versuchsanst. 206 pp.