SEED DISPERSAL. SEED GERMINATION AND SEEDLING ESTABLISHMENT OF HIERACIUM AURANTIACUM m AN OLDFIELD COMMUNITY Thesis for the Degree of M. S. MICHIGAN STATE UNIVERSITY BASIL G. STERGIOS 1970 ~ ‘ car’ a. x \e. 1“. ‘W‘ivmmw I ‘2: ”W” LIBRARY Ill/I/ll/fi////II///I/////I////I///II/////I/I/////IZI///I//////I M‘°"‘3““5"““° 31293 1003 University - 12 “V A“; ”3‘ ’ “-218 b '-'\ e a e r g a I. l a Wt . 7 ., c}. 3’6; mo \ I“. r. r I" '1 f ’/ k PI“ 3 I‘ 2‘ It ; , J u:- Plate 1. Hieracium aurantiacum L. ! Louden Oldfield. June, 1969. ABSTRACT SEED DISPERSAL, SEED GERMINATION AND SEEDLING ESTABLISHMENT OF HIERACIUM AURANTIACUM IN AN OLDFIELD COMMUNITY BY Basil G. Stergios Seed dispersal, seed germination and seedling establishment in Hieracium aurantiacum were studied in an 18-year-old oldfield community in southwestern Michigan. Most of the seed produced was deposited within 2 meters of the source. Seeds deposited nearest the source were found to be significantly larger and more viable than those de- posited at a distance. Seedling survival appeared to be enhanced by establishment within the parent population rather than establishment in adjacent vegetation. Germina- tion of seeds produced early in the year was significantly greater than that of seeds produced later in the year; cold treatment greatly enhanced germination in the late seed, but not in seed produced early in the season. Exclosure experiments indicated that insect predation and seedling mortality increases through the summer. These results suggest that although seed production in this species is Basil G. Stergios high, successful reproduction by seed seems confined pri- marily to areas already occupied by the parent pOpulation and appears relatively unimportant in the colonization of adjacent vegetated areas. SEED DISPERSAL, SEED GERMINATION AND SEEDLING ESTABLISHMENT OF HIERACIUM AURANTIACUM IN AN OLDFIELD COMMUNITY BY Basil G; Stergios A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Botany and Plant Pathology 1970 ACKNOWLEDGMENTS I wish to thank my committee Chairman, Dr. Stephen N. Stephenson for his continued guidence and his conscien- tious review of this manuscript. I also wish to thank the other members of my committee: Dr. John H. Beaman and Dr. Brian Moss. I wish to express my appreciation to Dr. William B. Drew for his continued support.and encouragement throughout the develOpment of this study. I wish to thank Dr. George Lauff, Director of the Kellogg Biological Field Station for the use of facilities for this investigation. ii TABLE OF CONTENTS ACKNOWLEDGMENT S O O O O O O O O O O O O O 0 LIST OF FIGURES O O O O O O O I O O O C O 0 LIST OF PLATES O O O O O O O O I O O O O 0 Chapter INTRODUCTION 0 I O O O O I O O O O O O O 0 METHODS AND MATERIALS . . . . . . . . . . . Description of the Study Area . . . . . History of the Study Area . . . . . . . Description of the Plant Studied . . Seed Dispersal and Seed Size (Volume) . Germination Studies in the Laboratory . . 1 Seedling Establishment and Survival Stud'e MSULTS O O O O O O O O O I O O O O O O O I General Population Behavior and Phenology Seed Dispersal and Deposition . . . . . . seed Size 0 O O O I O O O O O O I I C Germination Studies in the Laboratory . Seedling Establishment and Survival . . DISCUSSION 0 O O O O O O O O O O O O O O O S UMMRY O I O O O O O O O O O O O O O O C 0 LITERATURE CITED . . . . . . . . . . . . . iii 5 Page ii vii l3 l3 19 22 29 38 44 46 48 Page APPENDICES O O O O O O O O O 0 0 O 0 O 0 0 O O O O O 52 Appendix 52 A . . . . . . . . . . . . . . . . . . . . Appendix B . . . . . . . . . . . . . . . . . . . . 53 Appendix C . . . . . . . . . . . . . . . . . . . . 55 Appendix D . . . . . . . . . . . . . . . . . . . . 58 Appendix E . . . . . . . . . . . . . . . . . . . . 74 iv LIST OF FIGURES Figure Page 1. Aerial photograph of the Louden oldfield (one inch = 100 feet) with accompanying overlays O O O O I O O O O O O O O I O O O O O 4 2. Mean length of individual scapes across four aggregations of H. aurantiacum . . . . . . . . l4 3. Seed, scape and flowering head production in H. aurantiacum aggregations from July 15 to November 1, 1968 . . . . . . . . . . . . . . . l7 4. Seed trap results . . . . . . . . . . . . . . . 20 5. Mean seed deposition (field seed pool) along transects of varying lengths . . . . . . . . . 23 6. Relationship between seed size and distance dispersed O O O O O O O O O O O O I O O O O O 25 7. Relationship between total seed germination and distance of dispersal . .,. . . . . . . . 27 8. Mean number of seeds germinated without a cold period at monthly intervals from June through September, 1969 . . . . . . . . . . . 30 9. Mean number of seeds germinated after cold treatment 0 O O O O I I O O I O O O I O O O O 33 lO. Cumulative percentage germination with and without a cold period of seed produced at four monthly intervals during summer, 1969, and of seeds stored for one year . . . . . . . 35 11. Mean log (T+l) number of surviving seedlings protected and unprotected from predators during summer, 1969 . . . . . . . . . . . . . 39 Figure Page 12. Mean number of new seedlings inside and outside aggregations of H. aurantiacum . . . . 41 13. Cumulative mean percentage germination of Hieracium aurantiacum, H. florentinum, and putative H. aurantiacum X H. florentinum seeds sown on moist blotting paper in petri plates. June, 1969 . . . . . . . . . . . . . 77 vi LIST OF PLATES Plate 1. 2. Hieracium aurantiacum L. Louden oldfield. June, 1969 O O O O O O O O O O O O C O O O A View of two aggregations of Hieracium aurantiacum in the Louden oldfield, also showing the exclosure design . . . . An over-all view of the Louden oldfield looking from north to south. June, 1969 . A panoramic view of the Louden oldfield. June, 1969 O O O O I I O O O O O O O O O I Demonstrating the use of a portable auto vacuum cleaner adapted for seed collection in the field 0 O O O O O O O O C I O O O I A comparison of two species of Hieracium and a putative hybrid, showing differences in the color of the rays and size of the inflorescence . . . . . . . . . . . . . . A comparison of two species of Hieracium and a putative hybrid, showing a close-up of the rays and their color difference . . A comparison of two species of Hieracium and a putative hybrid, showing differences in pubescence and texture of the basal leaves . . . . . . . . . . . . . . . . . . vii Page Front Piece 58 60 62 66 68 70 72 INTRODUCTION Many of the plant species of advanced successional communities possess, in addition to well developed and significant means of seed production and dispersal, some means of vegetative reproduction. Despite abundant seed production by these species there is rarely evidence of extensive propagation by seedling establishment. On the other hand, vegetative reproduction is obvious in oldfields and other successional habitats, being largely responsible for the highly patterned nature of these communities. The purpose of this investigation was to determine the importance of achene (and thus seed) reproduction in the population dynamics of a typical oldfield plant species, orange hawkweed (Hieracium aurantiacum L.). Three aspects of achene biolOgy were considered: 1. seed production and dispersal from selected populations (aggregations); 2. seed size and germination behavior relative to dispersal grad- ients; and 3. seedling establishment and survival in the field. Putwain, et al. (1968) found essentially no seed reproduction by Rumex acetoseZZa in English pasture-lands, the pOpulation having been maintained primarily by vegetative reproduction. Harper (1967) points out that reproduction by seed and by vegetative means each represent differing components of a population's "r" value (intrinsic rate of increase) or biotic (reproductive) potential. The greater biotic potential inherent in seed reproduction is largely offset by high seedling mortality in the field (Salisbury, 1942; Harper, 1967). However, the biotic potential inherent in vegetative reproduction is smaller, but involves a lower mortality risk than in seedlings (Harper, 1967). Thus, the investments made by the plant for vegetative reproduction are more secure (Harper, 1967). Salisbury (1942) points out that seed plants reproducing vegetatively have a decided advantage over non-vegetatively reproduced plants, especially under adverse environmental and competitive conditions, and are assured successful propagation even when seeds are ineffectively or sparsely produced. METHODS AND MATERIALS Description of the Study Area. The study area is located at the W. K. Kellogg Biological Station in the northwest corner of Ross township in northern Kalamazoo County (Tls, R9W, Section 5) on the former William Louden farm (now part of the W. K. Kellogg Biological Station). The Louden oldfield is approximately 332 meters by 201 meters, with the long axis orientated north and south (Appendix D, plates 3 & 4). The study area is well isolated from other oldfield weed seed sources by cultivated areas on three sides and woodland on the other. It is bounded on the north by the Zeal Langshaw farm, on the west by the former Bailey farm (now part of the W. K. Kellogg Biological Station), and on the south by the con- tinuation of the Louden farm. The area where the field is located is part of the Gull Lake out-wash plain, consisting primarily of sand with very little gravel (Leverett & Taylor, 1915). The local relief of the area is dominated by ice-melt pit depressions, some of which are now lakes. The topography of the area is varied (Fig. 1 and overlays). It consists of an upper, generally level area Fig. 1. Aerial photograph of the Louden oldfield (one inch = 100 ft.) with accompanying overlays:~ A. Distribution of H. aurantiacum aggregations in the study area. B. Distribution of soil types in the study area determined from soil profiles. C. SIOpe description of the study area; soil sample sites and soil profile sites. .u A L H Y a 0 I m U I O l 4 one nuscass ‘nud r080. ~'—w .‘l' AJ ”THO“ “fl“! '3. X 0 >wn>§m bmmfihr mcfi~jOZ $2220 a. §.nr__,..e>2 6 to the west-north-west that drops off abruptly to a depres- sion in the northwest corner, and a more extensive lower area to the east and southeast. The field is cut diagonally from the northeast to the southwest corner by a 4° to 7° lepe that has been subject in places to rather severe erosion. There are areas of definite truncation of the soil profile (Appendix B) as well as areas of marked slope- wash accumulation. The latter condition is most pronounced in the southeast corner of the field (Profile site C, Fig. 1). Soil analyses of the study area (see Fig. l for locations) were made on six samples. All analyses were done by the soil testing laboratory at Michigan State University. The results of these analyses are given in Appendix A. From eight soil profiles taken in the field (Appen- dix B), four possible soil types were identified: Oshtemo sandy loam, Kalamazoo sandy loam, Ockley sandy loam, and Washtenaw loam--an alluvial soil (personal communication with Ivan F. Schneider, Department of Soil Science, Michigan State University). Based on relationships of these soil types to topographical features in the field, the soils were mapped as shown in Fig. l. The floristic composition between the upper and lower areas of the Louden oldfield is similar to that encountered by Caruso (1963) in the Ft. Custer Military Reservation. Poa compressa appears to be of greater im- portance in the upper area while in the lower area Agropyron repens is more commonly encountered. Interspaced among the matrix of Poa and AgrOpyron are aggregations of other species, which form a pattern of patchiness similar to that described by Caruso (1963), and Evans and Cain (1952). Also, numerous shrubs and small trees are invading the field. Some of the most important species which occur in the study area are Antennaria plantaginifolia, Chrysanthemum Zeucanthemum, Hieracium aurantiacum, Hieracium florentinum, Hieracium pratense, Rubus fZageZZaris, Agrostis hymalis var. hymaZis, Solidago canadensis, Solidago juncea, Solidago nemoralis, Solidago graminifolia, and Solidago missouriensis. A list of the 118 species found in the Louden oldfield be- tween May and September, 1969, is given in Appendix C. Botanical nomenclature follows Gleason (1963), Gleason and Cronquist (1965), and Gould (1968). Voucher specimens from the study area have been placed in the Beal-Darlington Herbarium, Michigan State University. History of the Study Area The study area had been utilized for crOp farming before 1911 but lay fallow until 1915. Due to unfavorable t0pography, which was subject to frequent erosion, the field was not cropped but was fenced off and used for pasturing sheep intermittently from 1915 until about 1942. During the seasons when no sheep were present, the field lay idle. Cropping began in 1942 and continued until 1948. During this period, corn and other grain crOps were planted, giving very poor yields (personal communication with Margaret Louden, Hickory Corners, Michigan). The crops were gen- erally rotated; oats were planted in the spring and wheat in the autumn of one year followed by corn the next. The field is believed never to have been marled or manured to any extent. Apparently herbicides were never applied to the field. Hay and alfalfa were also grown, but only in the very southern most portion of the field (from examina- tion of the series of actual photographs made since 1938 it appears that this portion was periodically under culti- vation as part of an adjacent field to the south). The field lay fallow until 1951, at which time it was planted with wheat. Description of the Plant Studied Orange hawkweed is native to EurOpe but is now widely distributed from Newfoundland to New Jersey, west to Minnesota and northern Indiana (Fernald, 1950). It has been found as far south as West Virginia to the tip of the northern panhandle, and further south in the Allegheny Mountains to the East (Myers, 1950). In the Tyrol and Austria it has been cultivated for some time as an ornamental (Pugsely, 1921). Orange hawkweed has been reported from an elevation of 5500 feet on Mt. Washington (Harris, 1930), and from higher elevations in Europe (Zahn, 1923). The time when orange hawkweed enters the succes- sional sequence of an oldfield has not been recorded. This entry undoubtedly varies considerably with particular com- binations of soil types, other plant competitors, and sur- rounding environmental conditions. Adjacent to the study area, individuals of H. aurantiacum were found in the third year stage in three independent oldfield study plots (un- published data, Dr. John Cantlon, Department of Botany & Plant Pathology, Michigan State University). Seed Dispersal and Seed Size (Volume) Seed dispersal of Hieracium aurantiacum was deter- mined from six aggregations in 1968 (five in 1969) selected for their similar size and their isolation from other aggregations. Three line transects (extending radially 25 dm in 1968; 35 dm in 1969, from the center of each aggregation) were randomly selected. Three observations were made at each half-meter along these dispersal grad- ients by trapping seeds on plastic container lids coated with 'Tanglefoot' brand bird repellant (distributed by John Dill, Inc., Kalamazoo, Michigan). The lids were supported 10 cm above the ground by plastic stakes. The traps were 10 installed just prior to the initial seed production period of mid-June. They were collected about 20 days later after most of the seed crOp had been dispersed. Seed deposition (field seed pool) at varying dis- tances between aggregations of H. aurantiacum in the study area was examined (June and July, 1969). Six pairs of aggregations were selected with distances between each of them of 2, 6, 8, l4, l6, and 20 meters respectively. Three seed traps were placed at each meter interval along each transect. The traps were collected and the seeds counted at the end of the initial seed production period. Seed volume was calculated (to hundredths of a mm3) by measuring the lengths and diameters of seventy seeds trapped from each half-meter increment along the dispersal gradient. The equation v = nr20.8L + 1/3nr20.2L represents total volume of each seed; four-fifths of the total seed volume taken as the volume of a cylinder and one-fifth as a cone (where r is the radius and L is the length). Germination Studies in the Laboratory In order to examine the relationship between germi-. nation and distance of dispersal, seeds were trapped for. the germination studies along transects from six ll aggregations. Seventy-five seeds were sown in similar trays on moist soil from the study area. Two germination repli- cates were employed for each increment along the gradient. The trays were covered with cheese cloth and maintained at room temperature. All the seeds had previously been cold treated for 90 days at 0°C. The number of germinated seeds was recorded every two days until all germination had ceased (16 days). Volume rather than weight was used as an indi- cation of seed size to avoid possible error in weight caused by the "tanglefoot" compound. In order to determine the relationship between ger- mination and time of seed production in H. aurantiacum, seeds were collected from the Louden oldfield at monthly intervals from June through September, 1969. The seeds were germinated on moist soil in petri plates. Some of the plates were stored at 0°C for 30 days before germination tests were begun. The remainder of the seeds were germi- nated immediately upon collection from the field. Percent- age germination was recorded every two days. Seedling Establishment and Survival Studies To determine the effect of predation on seedling mortality, 19 sets of insect exclosures were erected. The exclosures were constructed of plastic screening and placed over 4-dm2 plots at the edge of H. aurantiacum aggregations 12 in the study area (Plate 2). The experiment was begun in March, 1969. All 38 plots were inoculated with 1000 seeds collected from the same aggregations during June, 1968. The control screens were elevated 15 cm above the ground and the sides were slit to allow the entry of flying in- sects. The tops of the control exclosures were permanently sealed at the bottom and kept temporarily closed at the top by Clothespins. The experimental plots were sprayed with insecticide twice weekly and surviving seedlings were counted once per week from May through September. Seedling establishment and survival was measured in 1968 from seeds deposited inside aggregations. Eighty-eight 4-dm2 plots all of similar size were selected at random from six aggregations. Surviving seedlings, having been marked with plastic toothpicks, were those that had remained since the previous sampling. The following April (1969) the plots were again sampled to determine the number of over-wintered seedlings. Establishment and survival was measured in vegetated areas adjacent to the aggregations in 1969. Two hundred and forty 4-dm2 plots were selected along 3—meter transects radiating from the aggregations. Seedlings in these plots were sampled every seven days as previously described. RESULTS General Population Behavior and Phenology Most of the stoloniferous growth in Hieracium aurantiacum occurs early in the season (mid-May to the end of June). The young rosettes become established during the remainder of the season. In the Spring of the following year bolting occurs. In 1969 anthesis occurred between May 21 and June 7. By late June the initial flowering had ceased, and seeds had matured in about one-fourth of the Louden oldfield population. Seedlings which became estab- lished did not mature during the same season. After an initial phase of vigorous growth the seedlings became dormant for the rest of the season. Scapes produced at the.outer limits of H. aurantiacum aggregations were significantly taller than those at the center, and the number of flowering heads per scape was significantly greater at the periphery than toward the center (Fig. 2). Differences in rosette density across an aggregation were not measured. Since the aggregations sampled were probably 10 or. more years old, one might expect some differences in the various parts of the population. Koyama and Kira (1956) 13 14 Fig. 2. Mean length of individual scapes across four aggregations of H. aurantiacum. Mean number of flow- ering heads per scape across the same aggregations. The points represent tZSY. MEAN NO. HEADS /SCAPE CI. .4 :L n... of. cu $33 moqu: 23;: 33.0. I 202.40me04 .10 mmhzwo 20m... .20 n. o. o o n o. a. 0“ E0 to U0. tn. ION Ina vOn r I) n '0' to? on (W0) 1H9|3H 3dVOS NVBW l6 and Stern (1965) showed that in time an increase in density within older portions of a plant pOpulation forces a sharing of limited resources with a compensating reduction in devel- opment of individuals. Less vigorous plants may be compet— ing for sites long occupied by H. aurantiacum in the center of the aggregation, while more vigorous colonizers occupy the outer edge. Reproduction within populations of H. aurantiacum declined rapidly after the initial bolting early in the season. Seed production (as represented by seed deposition) was closely correlated with flowering head and scape pro- duction (Fig. 3). The continued appearance of seeds after most scape production had stopped apparently resulted from the presence of plants in adjacent areas not sampled and previously dispersed seeds that had fallen or been blown onto the traps. A decrease in the number of seeds per head further indicated a decline in reproductive vigor as the season progressed. Seeds from the flowering heads of indi- viduals in eight aggregations were counted in June, 1969 and again in September. The mean number of seeds per head produced was 74 in June and 58 in September. The difference in the number of seeds per head between the two periods was significant (t = 7.8; p = <0.05; N = 14). l7 Fig. 3. Seed, scape and flowering head production in H. aurantiacum aggregations from July 15 to Nov. 1, 1968. Variance estimates are 28?; .29 8qu 62.539: dz 8. 7 5 5 . 7. 6 5 4 3 I. 8 a e 8 Mb m. 0 O O O a a M o o o L D D b b P b P D D b P b b a m 5 Al Ml m m m: > > 085:3“. awhh80¢ b.50( — 8:88.... :3 cucuté.3:umo.. 5:9. 62.56.... mi: ..... same tar—.323 cum» 842 bud - 3.00.. 30.33.8th .au uhgg 80¢; mgr—.ga . u 08300: $9200 02.3: .1333"... +0 L70“ LBS" I501 :1 W w 333.20» .oz 94 398225 8%.». dz 0.51 N04 005‘ O- 0 90 41 DAYS [:J scnss E] noweame HEADS I sue TRAPPED l9 Seed Dispersal and Deposition Seed trap results showed that deposition was re— stricted primarily to within 25 dm of the source aggrega- tions (Fig. 4). An analysis of variance of the data from the 1969 transects revealed no significant differences between total seed output of the aggregations sampled (F14, 345 = 0.84; p = >0.05). This also indicates that wind had very little influence on the direction of dispersal since sample transects were randomly orientated. Data from both 1968 and 1969 indicated that the pattern of dispersal of seeds with respect to distance from the aggregation re- sembled a negatively sloped exponential curve similar to that discussed by Wolfenbarger (1946) and Faulwetter (1917) for dispersal of propagules from a source. In keeping with this pattern, the majority of H. aurantiacum seeds dispersed were deposited before they had left the aggregation (Fig. 4). Similar seed dispersal patterns have been reported by Isaac (1930) for douglas-fir, and also for black spruce (Lake States Exper. Sta., 1939; (cited by Wolfenbarger, 1946)). Hafmann (1911) suggested that the exponential reduction in numbers of gymnosperm seedlings from a seed source reflected the seed dispersal pattern. The dispersal pattern of seeds of H. aurantiacum suggested that deposition never reaches zero, but stabilizes at about 0.5 seeds per cm2 in the area 2 to 2.5 meters 20 Fig. 4. Seed trap results. Points indicate mean seed numbers per cm2. Estimates of dispersion are 252} MEAN N0. SEED TRAPPED / CM' 7.0-! 5.01 1‘ O n 9' O n 2.01 LOfi .4 AGGREGATION - ——--——----——-—-—-W-—-- --—- PERIPHERY ----- SEED DINII‘I’ID M Jun. I... (I - Io Tu") SEED 01'0“le II JUNE, I... ( I I 4. TRAP.) O- 0.»-..— I'O I; 2'0 25 3'0 35 DISTANCE (DM) FROM CENTER OF AGGREGATION 22 beyond the seed source (Fig. 5). This is to be expected (Wolfenbarger, 1946) in dispersal distributions like those already described. Seed Size In the vicinity of H. aurantiacum aggregations in the Louden oldfield, seed volume and distance dispersed were closely related (Fig. 6). The seeds from any given point on the gradients did not differ significantly in their volumes. The smaller seeds were dispersed farther and, over the distance sampled, the relationship was nearly . linear. Variability in seed size has been shown to play an important role in the propagative potential of certain plants (Black, 1956 & 1957; Cavers & Harper, 1966; Janzen, 1969; Salisbury, 1942). Salisbury (1942) points out that smaller seeds are more widely and efficiently dispersed. He indicates that wider diSpersal enables a more probable dissemination to all suitable habitats. The relationship between seed germination of H. aurantiacum and distance of dispersal is shown in Fig. 7. The daily percentage germination for each increment of the gradient is also given. The germinability (to 16 days) declined significantly as the distance of dispersal in- creased. Smaller seeds that were dispersed greater dis- tances showed lower germinability and a tendency for increased germination after an initial delay (Fig. 7). 23 Fig. 5. Mean seed deposition (field seed pool) along transects of varying lengths. Variance estimates are :23)? ENOILVQIIOOV 3H1 NIHUA HOIUBOGIIO 0338 I “an: E -l sou uomooaao 033: l 103. a Inc-'1‘] mm” 0' ——-— ' I I ' A . l o ., _ / t 1' 4/ l _ I / I 3 <3 5. <3 <3 5. a 3 5 .. O " o 0 Q #5 .j -' d ,m/oaas JO uaaunu man DBTAWE ON METERS BETWEEN AGGREGATIONSGEEO 8W) 25 Fig. 6. Relationship betweed seed size and distance dispersed. 95% confidence limits are indicated. MEAN SEED VOLUME (MN’) 0.35 0.30 'I 0.201 n . -o.coa up 4.31: as, . 0.0001 punsou'a a - a...“ n‘ - 0.» Y I 0.202 '0.047(X) --—-——----d 0.80 0.!5-1 0.l0- 0.00II 9-! #- - - - - - PEIMRY 0F AGGREGATIOI - - - '- 5 Co Co {.0 1.5 DISTANC E IN METERS 27 Fig. 7. Relationship between total seed germination and distance of dispersal. Bars indicate per diem percentage germination. pmmomZADOm ommpzzpjoz .u“..‘ .o J' x» . .. l. 4 {r camawmmwn... ... .4.-...» .m . ' I'.‘ ‘ 9. I 1.! T. n...u.1 ... § 9km- a 63..., .2 1 v .4.|l I. I .. dove... r.r<.4.. H I I: 7...... .. ink... I Ihhwfinwwi ...... Iflriaé - .. I Wumvufiw [II 60d 50" 0 30 20 owhdzzzmuo momwm .OZ W O O O O O O O O 0 l 9 8 7 6 5 4 3 2 I b p I L p E p r E b .0... «manna. H..,. I - I \I . ‘qlfl’o>p.fior.mucfl.r9mofl’4 .4. .4, \«s I; v 5 X .........3.....«... ,. ..;..... .. 1 3 4 I. 5. 9 . 3 l 2 m 7 2 2 3 . 3. _ .. 0 w..." a... .».n»w.p«?. “€644... -. - ......I. O 6 = ”.56....‘aowohuo. ‘. ‘mu. 5%. aHHIII I. hav‘rtoa ' u = I v. .. fl (Na-Huwbsflc 4.4.04... .i; tape... 3 .. p s s r... . ....?.Wa..ris. ._ f Y S S S R .S N 0 S .4..... ,I. . .-...........-,.I R w... .I. flan»... .... A "mh'IIIIIII. II ..N.~..é-J \ I , .. Fe “II .HIW,J4III-IIP P DISTANCE IN METERS FROM CENTER OF AGGREGATION 16 DAYS 1s :1 'dl ' '(Ik (4‘ I.) a\' ‘1’ I 0 DAYS 8 DAYS Lvflf D 6 DAYS 35.99% AY (“WM 4 D .S 29 Seeds trapped with "tanglefoot" showed no significant dif- ferences in germinability from seeds not so contaminated (t = 2.4; p = >0.05; N = 6). A number of investigators have pointed out the apparent relationship between germinability and seed poly- morphism. Johnson (1935) observed that the percentage germination in Avena was closely correlated with the av- erage seed weight. Becker (1912, cited by Salisbury, 1942) showed that seeds of two different sizes of Rapistrum and CakiZe differed in their germinability. Cavers and Harper (1966) found that germination of Humex was significantly higher in seed from the upper part of the panicle than from the lower. They also found that, in general, the upper seeds were heavier than the lower. Germination Studies in the Laboratory Seeds of Hieracium aurantiacum produced at different times during the season showed a significant difference in germinability for the first 10 days when germinated without cold treatment (Fig. 8). The germinability of seeds pro- duced in June was far greater than that of seeds produced during the latter part of the season. There was no signif- icant difference in the germinability of seeds collected from different aggregations. 30 Fig. 8. Mean number of seeds germinated without a cold period at monthly inter- vals from June through September, 1969. Variance is given by iZSx. .L ‘I' I4 f0 T4 l3¢ ltd ll‘ IO-1 1 9 1 I1 q 4 0 7 0 5 ouh5fi~33t2°$¢ - fwu Ta a — —.—.,.—.— ‘- " 1w:- v..<‘" 4" T3” Wfiafi Llf‘fi_ Em?“ It *3 IO__-“ § - “‘5 _. _.__vh ..’__ _. N —I2 T I I T I V V I r I o o o o 0 ° 0 O o o 9 m 0 b o 0 v n N _ NOIlVNIWEBD 39V1N3983d BAIlV'InWI'IODV GERMINATED IMMEDIATELY ON MOIST SOIL AT 22' 0 (I969) SEED D A Y S SEED STORED FOR so DAYS 0N MOIST SOIL. AT ZERO°G (I969) ONE YEAR OLD SEED(JUNE.I968) J33; .’.‘- 71‘.” I." 4sz '1 i i. ’4 .53: 37 cold and dry spells, their seedlings are readily killed (Cavers & Harper, 1964). Delayed and intermittent germination caused by seed dormancy is probably genetic in nature (Cavers & Harper, 1966). Gustafsson (1935) looked at germination behavior in the genus Hieracium, in which embryos of apomictic species developed independently of the development of the achene. This resulted in the dispersal of mature achenes whose embryos were at different stages of development. He postu- lates that some seed embryos may continue nuclear division and eventually mature, resulting in delayed germination. Other embryos may never develop to maturity after dispersal, and the seeds become inviable. Since apomixis has been demonstrated in H. aurantiacum (Gustafsson, 1946 & 1947; Stebbins, 1941 & 1950), it is possible that delayed ger- mination in the smaller (less developed) seeds in the Gull Lake area populations resulted from a similar mechanism. The percentage germination of H. aurantiacum seeds collected in June, 1968 and aged for one year in a dry container was slightly below that of seeds germinated fresh in 1969 from the same aggregations. No fresh seeds were germinated in 1968. Experiments have indicated that storage of seeds tends to decrease germinability (Gustafsson, 1935; Gardner, 1921). However, Steinbauer and Grigsby (1960) obtained no germinability differences with Rumex. 38 Seedlin Establishment and Surviva Results from experiments in the field indicated that many seedlings of Hieracium aurantiacum died largely from predation by grasshoppers. From early spring until late July, 1969, the number of seedlings and resulting survival was not significantly different among the plots (Fig. 11). From the first week in August the survival of seedlings in the unprotected plots continued to be signif- icantly lower than in the protected plots. A general down-1 ward trend in over-all seedling survival toward the latter part of the season may have been due in part to a severe. dry spell. Desiccated seedlings not touched by predators were observed in dried-out surface soil in both the pro- tected and unprotected plots. Cavers and Harper (1967) pointed out the difficulty of determining the causes of sudden mid or late season mortality. In their studies with Rumex (1967), disappearances coincided with mid-summer drought and heavy rains. The precise causes of seedling death remained an unsolved problem. Seedling survival inside aggregations in 1968 was significantly greater than that occurring in adjacent veg- etated areas in 1969 (Fig. 12). The possible error intro- - duced by comparing data from two different seasons was not determined. However, differences in weather and also in H. aurantiacum seed production between the two years (Fig. 4) 39 Fig. 11. Mean log (T+l) number of surviving seedlings protected and unprotected from predators during summer, 1969. Variance is given by i 28x. Gabby—bent: I l I I I awhouhocm a 95:. r 8 v V v v n m u a m m m m m n 1 m I 1 8 l I 1 3 A A c. I m 9 1. 0 c. 0 R e. W D I b D D D \Ipl b b 0.0 «.0 .36 U.o° v0.0 .0.- ca; 0.. I 0.. htuaoucgt loathzauca h8(z.loo axuccozao¢¢0 _ htuo.>u NICO! aging Quanta... ouun >h.>.ho¢ haunt. 08(208. 2.23.81 SDNI'IOBBS 'ON (I41) 90") NVBN 41 Fig. 12. Mean number of new seedlings inside and outside aggregations of H. aurantiacum. Variance estimates are :zsi. MEAN NO. SEEDLINGS / 40M“ 4m u 0 4‘ N O l IO'I ICU DEEDLIIOS “MED m Lunar-Dun. I... “M ------ SUM!” can” MED m mum I... DEW """" "MVIVM SEEN.“ MED “ITS!“ mum I... “A”. """""" KW Guam “HID m “MIMTM I... DIAS“ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ + 3 1 DAYS FROM ANTI'ESIS 43 suggests that it may be significant. Cavers and Harper (1967) found that many seedlings of Rumex established during the summer died. Few H. aurantiacum seedlings survived the winter in the Louden oldfield aggregations (Fig. 12). Most seedlings that disappeared did so during the cotyledon stage. Vegetated areas immediately surrounding aggrega- tions of H. aurantiacum in the Louden oldfield resemble the 'closed communities' referred to by Cavers and Harper (1967). Seeds of Rumex readily germinated in these closed habitats, but resulting seedlings failed to survive. DISCUSSION The Louden oldfield is typical of an advanced suc- cessional community, where diversity of and competition among plant species is high (Hanson & Churchill, 1961; Bard, 1952; Clements, 1916; Keever, 1950). Both vegetative re- production and reproduction by seed occur among plants in the Louden oldfield community. The highly patterned nature of this community is indicative of relatively successful vegetative reproduction and fairly limited seed dispersal. Aggregations of Hieracium aurantiacum in the Louden oldfield community appear to become stabilized by vegetative reproduction, then tend to resist further colonization by seedlings. Vegetated areas adjacent to these populations are also 'closed' to colonization by invading seeds of H. aurantiacum, and establishment is even more strongly re- sisted in these areas. My results suggest that successful propagation of H. aurantiacum in a local area is achieved primarily by vegetative means. The majority of orange hawkweed seeds produced appear to be deposited within rather than beyond the source aggregation. Reduction of seed size and germina- bility appear to occur as a function of distance of disper- Sal. Seedlings that do manage to become established, either 44 45 within the parent population or in an adjacent 'closed community,‘ soon succumb to excessive dry spells, predation, and winter kill. Harper, in 1967, issued the challenge: "What is the relative energy expended in producing a veg— etative propagule and a seed? Can this expenditure be related to the relative risks of establishment by the two means of the relative ecological importance of local and long distance spread?" Salisbury (1942), and Harper et al. (1966; 1967), have suggested that the main function of seeds is to locate suitable habitats for colonization via long distance dispersal, to locate receptive sites for immediate germination at any distance from the seed source, and to saturate so-called closed communities and remain dormant until successional changes or other disturbances occur which create sites favorable for germination and survival. A relatively large amount of seed would be needed to fulfill all of these roles simultaneously. SUMMARY Scape height and number of flowering heads per plant are significantly greater at the periphery than toward the center of an aggregation of Hieracium aurantiacum. Most of the seed is produced during a short period in early summer. Seed production drOps off sharply as the season continues. The number of seeds per flowering head produced in June is significantly greater than the number pro- duced in September. Seed dispersal and deposition is restricted pri- marily to an area 20 to 25 dm beyond the source pOpulation. Beyond 25 dm, seed deposition stabil- izes at about 0.5 seeds per cm2. There is a close negative correlation between seed volume and the distance of its dispersal to 25 dm beyond the source aggregation. The germinability of seeds is negatively correlated with distance of dispersal to 35 dm beyond the aggregation. 46 10. 47 Seeds collected during August and September do not germinate readily without a cold treatment, whereas seeds collected in June show a high proportion of germination with or without a cold treatment. The germination of seeds produced early in the season is significantly greater than that of seeds produced later in the season. Established seedlings protected by exclosures sur- vive significantly better than those exposed to predation by insects. Survival of seedlings established within parent aggregations appears to be significantly greater than that of those established in adjacent veg- etated areas. Mortality among over-wintered seed- lings is high. LITERATURE CITED LITERATURE CITED Babcock, E. B. and G. L. Stebbins. 1938. The American species of Crepis; their relationship and distribu- tion as affected by polyploidy and apomixis. Car- negie Inst. Wash. Publ. No. 504. Black, J. N. 1956. The influence of seed size and depth of sowing on pre-emergence and eary vegetative growth of subterranean clover (Trifolium subter- raneum L.). Aust. Jour. Bot. Res. 7: 98-109. . 1957. The early vegetative growth of three strains of subterranean clover (Trifolium subter- raneum L.) in relation to size of seed. Aust. Jour. Bot. Res. 8: 1-14. Caruso, J. 1963. Interspecific pattern in an old-field community in southwestern Michigan. Unpub. M.S. Thesis. Michigan State University. Cavers, P. B. and J. L. Harper. 1964. Biological flora of the British Isles: Fumes crispus and Rumex obtusifolius. Jour. Ecol. 52: 737-766. Cavers, P. B. and J. L. Harper. 1966. Germination poly- morphism in Rumex crispus and Rumex obtusifolius. Jour. Ecol. 54: 367-382. Cavers, P. B. and J. L. Harper. 1967. Studies in the dynamics of plant populations: the fate of seed and transplants introduced into various habitats. Jour. Ecol. 55: 59-71. Clements, F. E. 1916. Plant succession. Carnegie Inst. Wash. Pub. 242. Dahl, G. E. 1952. Secondary succession on the Piedmont of New Jersey. Ecol. Mon. 22: 195-215. Darlington, C. D. 1932. Recent Advances in Cytology. P. Blakiston's Son & Co., Inc., Phila. 671 pp. 48 49 Darlington, C. D. and A. P. Wylie. 1955. Chromosome Atlas of Flowering Plants. Hafner Pub. Co., N.Y. 2nd. Ed. page 250. Davis, R. M. 1968. Effect of size area Open to coloniza- tion on species composition in early old-field succession. M.S. Thesis, Michigan State University. Evans, F. C. and S. A. Cain. 1952. Preliminary studies on the vegetation of an oldfield community in south- western Michigan. Contr. Lab. Vert. Biol. Univ. of Michigan. 51: 1-17. Faulwetter, R. C. 1917. Wind-blown rain, a factor in dis- ease dissemination. Jour. Agr. Res. 10: 639-648. Fernald, M. K. 1950. Gray's Manual of Botany. Amer. Book. Co., N.Y. 8th Ed. 1632 pp. Gardner, W. A. 1921. Effect of light on germination of light-sensitive seeds. Bot. Gaz. 71: 249-288. Gleason, H. A. 1963. The New Britton and Brown Illus- trated Flora of the Northeastern United States and Adjacent Canada. Hafner, N.Y. 3 vols. and A. Cronquist. 1965. Manual of Vascular Plants of Northeastern United States and Adjacent Canada. D. Van Nostrand Co., Inc., N.Y. 810 pp. Gould, F. W. 1968. Grass Systematics. McGraw-Hill Co., N.Y. 382pp. Gustafsson, A. 1935. Studies on the mechanism of parth- enogenesis. Hereditas 21: 1-112. . 1946. Apomixis in the higher plants. I. The mechanism of apomixis. Lunds Univ. Arsskr. N.F. Aud. 2, 42 (3): 1-66. . 1947. Apomixis in higher plants. II. The causual aspect of apomixis. Lunds Univ. Arsskr. N.F. Aud. 2, 43 (2): 71-178. Hanson, H. C. and E. D. Churchill. 1961. The Plant Commu- n'ty. Reinhold Pub. Corp., N.Y. 218 pp. Harper, J. L. 1967. A Darwinian approach to plant ecology. Jour. Ecol. 55: 247-270. Harris, S. R. 1930. An alpine station for Hieracium aurantiacum. Rhodora 32: 113. 50 Hofmann, J. V. 1911. Natural reproduction from seed stored in the forest floor. Jour. Agr. Res. 11: 1-26. Isacc, L. A. 1930. Seed flight in the Douglas-fir region. Jour. Forestry 28: 492-499. Janzen, D. H. 1968. Seed-eaters versus seed size, numbers, toxicity and dispersal. Evolution 23: 1-27. Johnson, L. P. 1935. General preliminary studies on the physiology of delayed germination in Avena fatua. Can. Jour. Res. C. 13: 282-300. Keever, C. 1950. Causes of succession on oldfields of the Piedmont, North Carolina. Ecol. Mon. 20: 229-250. Koyama, H. and T. Kira. 1956. Interspecific competition among higher plants. VIII. Frequency distribution of individual plant weight as affected by interac- tion between plants. Jour. Inst. Polytech. Osaka Cy Univ. Ser. D, 7, 73-94. Leverett, F. and F. B. Taylor. 1915. The pleistocene of Indiana and Michigan and the history of the Great Lakes. Mono. of the U. S. Geol. Survey. 53: 177-180. Myres, J. C. 1950. The genus Hieracium in West Virginia. Castanea 15: 92-101. Ostenfeld, C. H. 1910. Further studies on the apogamy and hybridization of the Hieracia. Zschr. ind. Abst. u. Vererb, lehre 3: 241-285. . 1912. Experiments on the origin of species in the genus Hieracium (apogamy and hybridism). New Phytol. 11: 347-354. Pugsley, H. W. 1921. On Hieracium aurantiacum. Jour. Bot. 59: 60-69. Putwain, P. D., D. Machin and J. L. Harper. 1968. Studies in the dynamics of plant populations: components and regulation of a natural population of Rumex acetoseZZa L. Jour. Ecol. 56: 421-443. Salisbury, E. J. 1942. The Reproductive Capacity of Plants. G. Bell & Sons, Ltd., London. 244 pp. Stebbins, G. L. 1941. Apomixis in the angiosperms. Bot. Rev. 7: 507-542. 51 Stebbins, G. L. 1950. Variation and Evolution in Plants. Columbia Univ. Press, N.Y. 643 pp. Steinbauer, G. P. and B. Grigsby. 1956. Some correlations between germination and dormancy of weed seeds in the laboratory and in the field. Proc. N. Cent. Weed Control Conf. 13: 35-36. . 1960. Dormancy and germination of the docks (Rumex sp.). Proc. Ass. Off. Analysts N. Amer. 50: 112-117. Stern, W. R. 1965. The effect of density on the perfor- mance of individual plants in subterranean clover swards. Aust. Jour. Agr. Res. 16: 542-555. U.S.D.A. August, 1960. Soil classification, a comprehen- sive system (7th approximation). Soil Survey Staff, Soil Conservation Service. Wolfenbarger, D. O. 1946. Dispersion of small organisms, distance of dispersion rates of bacteria, spores, seeds, pollen, and insects; incidence rates of diseases and injuries. Amer. Mid. Nat. 35: 1-152. Zahn, K. H. 1923. Das Pflanzenreich. 82. Heft. (IV.280). Abteilung II. Sect. XL. APPENDIX A Soil analyses at six sites from two east - west transects (aerial photograph, Fig. l in the text) APPENDIX A Soil analyses at six sites taken from two east - west transects in the Louden oldfield (Fig. 1 in the text). Data prepared by the soil teSting laboratory at Michigan State University from methods developed by personnel in the Department of Soil Science, Michigan State University. Test - Sites A B C D E F Soil pH 5.1 5.3 4.9 5.1 5.1 5.2 Buffer pH 6.6 6.4 6.2 6.4 6.7 6.6 *P (Kg. per hectare) 17.8 22.2 146.7 12.2 25.6 85.6 K (Kg. per hectare) 30 90 614.4 97.8 97.8 67.8 Calcium (Kg. per hectare) 355.6 1141 1141 355.6 717.8 355.6 Mg (Kg. per hectare) 55.6 202 202 97.8 217.8 55.6 Carbon (%) 0.89 0.42 2.47 0.90 0.62 0.83 Organic matter (%) 1.53 0.72 4.26 1.55 1.07 1.43 *P (% total exchangeable) 3.3 3.1 17.6 8.9 4.4 7.2 Calcium (% total exchangeable) 76.8 74.9 63.7 62.5 64.5 73.7 Mg (% total exchangeable) 20.0 22.2 18.8 28.8 31.3 19.2 *(method of anlaysis was the Bray T - 1, as appears in Pro. Soil Sci. of Amer. Vol. 29: 677-678. 1965.) 52 APPENDIX B Soil profiles taken from eight sites in the Louden oldfield (aerial photograph, Fig. 1 in the text) APPENDIX B Soil profiles taken from eight sites in the Louden oldfield. July, 1969. Classification based on U.S.D.A. (August, 1960). SITE ONE: Oshtemo loamy sand Ap: 0" - 9" Loamy sand Bt: 9" - 15" Sandy clay loam IIC: 15" - 48" Sand SITE TWO: Ockley loamy sand Ap: 0" - 5" Loamy sand (truncated) AB: 5" - 9" Sandy clay loam B21t: 9" — 27" Sandy clay loam - clay loam B22t: 27" - 48" Clay - clay loam (till) IIC: to 54" Sand SITE THREE: Ockley sandy loam Ap: 0" - 9" Sandy loam B21t: 9" - 18" Sandy clay loam - clay loam B22t: 18" - 48" Clay - clay loam (till) IIC: to 58" Sand 53 54 SITE FOUR: Kalamazoo loam Ap: 0" - 9" Loam AB: 9“ - 14" Sandy loam Bt: l4" - 28" Sandy clay loam IIC: 28" - 48" Sand SITE FIVE: Oshtemo sandy loam Ap: 0" - 9" Sandy loam AB: 9" - 14" Sandy loam Bt: 14" - 20" Sandy clay loam - clay loam IIC: 20" - 48" Sand SITE SIX: Kalamazoo sandy loam Ap: 0" — 9" Sandy loam A2: 4" - 12" Loamy sand AB: 18" - 22" Sandy loam Bt: 22" - 38" Sandy clay loam IIC: 38" - 48" Sand SITE SEVEN: Kalamazoo sandy loam Ap: 0" - 4" Sandy loam (truncated) A2: 4" - 12" Loamy sand Bt: 12" - 24" Sandy clay loam IIC: 24" - 48" Sand SITE EIGHT: Washtenaw loam (alluvial soil) Ap: 0" - 9" Loam 9" - 48" Loam (no mottles, no horizons) APPENDIX C List of species occurring in the Louden oldfield. May - October, 1969. APPENDIX C List of species occurring in the Louden oldfield from May through October, 1969. Herbs Achillea millefolium L. Agrostis hymalis (Walt) BSP. var. hymalis Agropyron repens (L.) Beauv. Allium vineale L. Ambrosia artemisiifolia L. *Andropogon scoparius Michx. Antennaria plantaginifolia (L.) Richards Anthemis arvensis L. Apocynum sibiricum Jacq. Asclepias syriaca L. Asclepias tuberosa L. Aster ericoides L. Barbarea vulgaris R. Br. Berteroa incana (L.) DC. Bromus inermis Leyes Capsella bursa-pastoris *Schizachyrium scoparium (Mich.) Nash. Carex normalis Mackenzie Carex pensylvanica Lam. Centaurea maculosa Lam. Cerastium nutans Raf. Cerastium vulgatum L. Chondrilla juncea L. Chrysanthemum leucanthemum L. Cirsium vulgare (Savi) Tenore Convolvulus spithamaeus L. Dactylis glomerata L. Danthonia spicata (L.) Beauv. Daucus carota L. Desmodium rotundifolium DC. Dianthus armeria L. Eragrostis spectabilis (Pursh) Steud. Erigeron annuus (L.) Pers. Erigeron strigosus Muhl. (Gould, 1968). Euphorbia corollata L. Euphorbia esula L. Gnaphalium obtusifolium L. Hieracium aurantiacum L. Hieracium aurantiacum X florentinum Hieracium florentinum A11. Hieracium floribundum Wimmer & Grab. .— Hieracium gronovii L. Hieracium longipilum Torr. Hieracium‘pratense Tausch. Hieracium pratense x florentinum Hypericum perforatum L. Juncus tenuis Wild. Lactuca canadensis L. (Yellow fls.) Lactuca canadensis L. (White fls.) Lepidium campestre (L.) R. Br. Lepidum virginicum L. Leptoloma cognatum (Schult) Chase Lychnis alba Mill Medicago lupulina L. Medicago sativa L. Melilotus alba Desr. 56 Melilotus officinalis (L.) Desr. Monarda fistulosa L. Nothoscordum bivalve (L.) Britt. Oenothera biennis L. Oxalis dillenii Jacq. Oxalis stricta L. Panicum sp. Panicum depauperatum Muhl Phleum pratense L. Physalis heterOphylla Noes. Plantago aristata Michx. Plantago lanceolata L. Plantago major L. Poa compressa L. Poa pratensis L. Potentilla argentea L. Potentilla canadensis L. Potentilla recta L. Prunella vulgaris L. Rudbeckia hirta L. Rumex acetosella L. Rumex crispus L. Solidago canadensis L. Solidago graminifolia (L.) Salisb. Solidago juncea Ait. Solidago Solidago nemoralis Ait. Solidago speciosa Nutt. Spiranthes tuberosa Raf. Taraxacum officinale Weber Tragopogon dubius Scop. Trifolium hybridum L. Trifolium pratense L. Trifolium repens L. Verbascum blattaria L. Verbascum thapsus L. Vernonia missurica Raf. Veronica arvensis L. Vines Celastrus scandens L. Parthenocissus quinquefolia (L.) Planch. Toxicodendron radicans L. Vitis riparia Michx. Shrubs Apocynum androsaemifolium L. Cornus alternifolia L.f. Cornus racemosa Lam. Crataegus sp. Lonicera tatarica L. missouriensis Nutt. 57 Rhus typhina L. Rosa multiflora Thumb. Rubus flagellaris L. Trees & Saplings Acer negundo L. Acer saccharinum L. Acer saccharum Marsh. Carya glabra (Mill) Sweet Carya ovata (Mill) K. Koch Malus sp. Pinus sylvestris L. Prunus serotina Ehrh. Quercus velutina Lam. Rhamnus frangula L. Sassafras albidum (Nutt.) Nees. Ulmus rubra Muhl. Non-Vascular Cladonia cristatella Cladonia rangiferina Caldonia chlorophaea APPENDIX D Colored Plates 58 Plate 2. A view of two aggregations of Hieracium aurantiacum L. in the Louden oldfield, also showing the exclosure design. -‘ 4 '..L: », A 7"..." rb' h;.' .. -}‘.:- '-"L. '."» Fr. " . "1'4. ‘0‘ \r f a . .3 . j: ; 9.3 ‘_‘ ‘K. | .- .. 1.9),: if," 1" H. ,. 60 Plate 3. An over-all view of the Louden oldfield looking from north to south. June, 1969. Fig. a. View toward the Southwest Fig. b. View toward the Southeast Plate 4. 62 A panoramic view of the Louden oldfield. June, 1969. Fig. Fig. Fig. Fig. Fig. Fig. a. b. C. View to the Southeast from the West end. View to the Northeast from the West end. View to the West from the West end. View to the North from the Southwest end. View to the Northeast from the Southwest end. View to the West from the Southwest end. 66 Plate 5. Demonstrating the use of a portable auto vacuum cleaner adapted for seed collection in the field. 68 Plate 6. A comparison of two species of Hieracium and a putative hybrid, showing differences in the color of the rays and size of the inflorescence. A - Hieracium aurantiacum B - Hieracium florentinum C - H. aurantiacum X H. florentinum (putative) 70 Plate 7. A comparison of two species of Hieracium and a putative hybrid, showing a close-up of the rays and their color differences. A - Hieracium aurantiacum B - Hieracium florentinum C - H. aurantiacum X H. florentinum (putative) 72 Plate 8. A comparison of two species of Hieracium and a putative hybrid, showing differences in pubescence and texture of the basal leaves. A - Hieracium aurantiacum B - Hieracium florentinum C - H. aurantiacum X H. florentinum (putative) APPENDIX E Additional remarks concerning H. aurantiacum and the putative hybrid H. aurantiacum X H. florentinum from the Louden old-field. Summer, 1969. APPENDIX E Most of the literature available on Hieracium aurantiacum concerns genetic studies related to apomictic behavior. Hieracium aurantiacum is part of an agamic com- plex in which there exists an intimate association between apomixis, hybridization, and polyploidy (Babcock & Stebbins, 1938; Stebbins, 1950). Hieracium subg. PiZoanZa and subg. Archieracium represent two such groups. Gustafsson (1935) and Stebbins (1941) have distinguished these groups in that the latter reproduces apomictically by diplospory or gen- erative apospory, while the former reproduces by somatic apospory followed by parthenogenesis. The same investiga- tors suggest that in the diplosporous groups, obligately apomictic species are dominant. In those complexes where somatic apospory is present (e.g. H. subg. PiloseZZa), the dominants are facultative apomicts which regularly produce some sexual offspring. The gradual rather than sudden evolution of the apomictic process seems most likely to occur in those complexes where somatic apospory is present (Stebbins, 1941). It is further suggested (Babcock & Stebbins, 1938; Stebbins, 1941) that in most or all agamic complexes, facultative apomicts precede obligate ones in 74 75 the development of the complex. The presence of only obligate apomicts in a complex may indicate that the group is relatively old and considerably removed from its sexual ancestors (e.g. H. subg. Archieracium, and Crepis). Hieracium aurantiacum is a facultative, tetraploid apomict, reproducing seed by somatic apospory followed by parthenogenesis (Gustafsson, 1947; Stebbins, 1941). In the anthers, fairly regular meiosis occurs and pollen formation is normal (Darlington, 1932; Stebbins, 1941). In pollen which I examined, no aborted grains were observed. The basic chromosome number of Hieracium aurantiacum is nine (Darlington & wylie, 1955), and the Zn number is 36. Hybridization and out-crossing have been observed to occur frequently in Hieracium aurantiacum (Ostenfeld, 1910; 1912). In the Louden oldfield (summer, 1969) I col- lected a putative hybrid of H. aurantiacum X H. florentinum. The plant occurred only occasionally and was found exclus- ively among individuals of the more widely distributed, non-vegetatively reproducing H. florentinum. No individuals were discovered in or near aggregations of Hieracium aurantiacum, and only one was found in an area completely free from all species of Hieracium. The putative hybrid grew either as scattered individuals or in small patches. The hybrid plants are about equal in height with the taller H. florentinum. The basal leaves, stems, and inflorescences possess features characteristic of both 76 parents (Appendix D, Plates 6 & 8). The outer corollas are orange along the margins with red bands down their centers, while in H. florentinum they are bright yellow and in H. aurantiacum deep red (Plates 6 & 7). The putative hybrid does not appear to reproduce vegetatively. Pollen grains from the suspected hybrid were exam- ined and all appeared to have nuclei, indicating probable viability. Existing small patches of the putative hybrid in the Louden oldfield indicated that not only did the plant have viable pollen, but that at least some of the seeds produced were also viable. Seed heads were collected from both solitary individuals and those in patches. Examination of the seeds revealed that only about 20% of all those collected appeared "normal" or in a viable condition, whereas about 98% of the parents' seeds appeared normal. The sterile seed is readily distinguished by its soft, light-brown coat and lack of an embryo. Normal seed ap- pears black, firm and plump. Embryos can be observed. Similar criteria for distinguishing between sterile and potentially viable seed were described by Gustafsson (1935) for several species of Hieracium. Seeds from the suspected hybrid H. aurantiacum X H. florentinum were germinated at room temperature on moist blotting paper in petri plates. Seeds from the parent plants were also germinated under similar conditions (Fig. 13). All seeds which could immediately be determined as 77 Fig. 13. Cumulative mean percentage germination of Hieracium aurantiacum, H. florentinum, and putative H. aurantiacum X H. florentinum seeds sown on moist blotting paper in petri plates. June, 1969. ‘-_ ——_.,M MEAN PERCENT GERMINATION (LOG SCALE) LOO‘I moo-I f OJo-I .6 0.05- 0.0:- o.ooo- f! ----- H. AURANTIAOUN -------- n. nonnrmw ’ "1"" PUTATIVE mum l .I .’ I .I 0.00: I : '1 I v I I v t v 2 4 c 0 no I: :4 no no DAYS 79 sterile were excluded from the experiment. Hieracium aurantiacum showed a higher percentage germination than did the suspected hybrid, and was significantly different in total germinability from both the hybrid and H. florentinum (t = 3.01; p = <0.05; N = 12). No explanation can be given for the relatively low germination of H. florentinum. The low percentage of over-all germinability in the three plants can be accounted for in part by fungal attack in the petri plates. "IIIIIIIIIIIIIIIIIIIIIIIS