UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Investigation into the effects of self-pollination on pseudotsuga menziesee (mirb.) franco Orr-Ewing, Alan Lindsay 1956

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1956_A1 O8 I6.pdf [ 11.17MB ]
Metadata
JSON: 831-1.0106274.json
JSON-LD: 831-1.0106274-ld.json
RDF/XML (Pretty): 831-1.0106274-rdf.xml
RDF/JSON: 831-1.0106274-rdf.json
Turtle: 831-1.0106274-turtle.txt
N-Triples: 831-1.0106274-rdf-ntriples.txt
Original Record: 831-1.0106274-source.json
Full Text
831-1.0106274-fulltext.txt
Citation
831-1.0106274.ris

Full Text

AN INVESTIGATION INTO THE EFFECTS OF SELF-POLLTNATION ON PSEUDOTSUGA MENZIESII (MIHB.) FRANCO by Alan Lindsay Orr-Ewing A thesis submitted i n pa r t i a l fulfilment of the requirements for the degree of Doctor of Philosophy i n the Faculty of Forestry and the Department of Biology and Botany We accept this thesis as conforming to the standard required from candidates for the degree of Doctor of Philosophy Members of the Faculty of Forestry Members of the Department of Biology and Botany In presenting this thesis i n pa r t i a l fulfilment of the requirements for an advanced degree at the University of Br i t i s h Columbia, I agree that the Library shall make i t freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my department or, i n his absence, by the University Librarian. It i s understood that any copying or publication of this thesis for financial gain shall not be allowed without my written permission.. Alan L. Orr-Ewing Faculty of Graduate Studies P R O G R A M M E O F T H E Jffitral (^ml^mwirmiim for tire Jlegree of A L A N LINDSAY ORR-EWING, M . C . B.S.F. (Edinburgh University) 1939 M.F. (University of California) 1951 MARCH 26th, 1956, at 2:00 p.m. IN THE BIOLOGICAL SCIENCES BUILDING Room 33 COMMITTEE IN CHARGE D E A N H . F. A N G U S , Chairman A. H. H U T C H I N S O N G. S. A L L E N V . C. B R I N K K . C O L E D. J. W O R T R. W. W E L L W O O D B. C. B I N N I N G K . G R A H A M External Examiner—DR. J. W. D U F F I E L D Industries Forest Nursery, Nisqually, Washington AN INVESTIGATION INTO THE EFFECTS OF SELF-POLLINATION ON PSEUDOTSUGA MENZIESII (MIRB.) FRANCO ABSTRACT The genetic systems which prevent self-fertilization in the Angiosperms are based on the incompatability of pollen and style. Past studies have also shown that many of the Gymnosperms, including Psuedotsuga, produce little or no viable seed after self-pollination. The causes of this seed failure were not known. t Controlled self- and cross-pollinations were accordingly made on eight Pseudolsuga meuziesii, sample cones were removed at various dates and the ovules embedded. The remaining cones were collected at maturity. The yield of viable seed from different trees varied considerably. The investigation was divided into two parts, a cytological study of development within the ovules of two trees which produced very little viable seed and the effects of self-pollination on the progeny. It was found that neither pollen germination nor its subsequent develop-ment up to the time of syngamy were inhibited. Proembryo and early embry-onic development were normal but nearly all the embryos collapsed before they had reached an advanced stage. It is concluded that embryo collapse is probably an inbreeding effect caused by the action of lethal and semi-lethal genes. It was found that inbreeding resulted in progeny which were usually smaller and less vigorous than those from both controlled cross- and wind-pollinations. PUBLICATIONS Direct-seeding experiments in the southern coastal region of British Columbia: E. H . Garman and A . L . Orr-Ewing. Technical publi-cation T.31, British Columbia Forest Service, 45 pp. 1949. Life history of the deer mouse: A . L . Orr-Ewing, Forestry Chronicle 26: 1-12. 1950. Rodent control in direct-seeding through use of poisons: A . L . Orr-Ewing, Research Note No. 17. British Columbia Forest Service, 8 pp. 1950.. Inbreeding experiments with the Douglas fir: A . L . Orr-Ewing, Fores-try Chronicle: 7-20. 1954. GRADUATE STUDIES Field of Study: Genetics Genetics A. H. Hutchinson, et al Histology and Cytology A. H. Hutchinson Silvics and Silviculture G. S. Allen Other Studies: Forest Ecology E. C. Stone and A. W. Sampson Forest Management P. M. Barr Biometrics E. Harris Forest Research R. N. Colwell and E. C. Stone i i ABSTRACT Extensive investigations have shown that the genetic systems which prevent self-fertilization in many of the Angiosperms are based upon the incompatibility of pollen and style. Past studies have also shown that many of the Gymnosperms, including Pseudotsuga, produce l i t t l e or no viable seed after self-pollination. The causes of this seed failure, however, had not been determined. It was considered that a cytological study of development within the ovules after the self-pollination of Pseudotsuga  menziesii should show whether seed failure was caused by comparable incompatible systems. Controlled self- and cross-pollinations were accordingly made on two trees at Vancouver in 1952 and on three trees each, at Victoria and Lake Cowichan in 1954. Cone samples from both pollinations were removed at various dates from each tree and the ovules embedded for later sectioning. The remaining cones were collected at maturity and the seed extracted. It was found that the yields of viable seed varied greatly in individual trees and the investigation was accordingly divided into two parts. The cytological study was confined to two of the trees which had produced very l i t t l e viable seed after selfing. The second part of the investigation was concerned with the effect of self-pollination on the progeny. The cytological study showed that neither germination of the pollen nor its subsequent development until the time of syngamy were in any way inhibited and both compared favourably with development i i i i n the ovules from cross-pollinated cones. Proembryo formation and early embryonic development were normal but the embryos i n both trees collapsed approximately ninety days after self-pollination. This collapse appeared to be caused by some failure i n the v i t a l relationship between the young embryos and their surrounding gametophytes. The la t t e r appeared healthy and comparable i n every respect with those i n which embryos from cross-pollination were growing vigorously. It i s considered that embryo collapse after self-pollination i s probably an inbreeding effect caused by the action of lethal and semi-lethal genes when brought together i n a homozygous state. This explanation could account for the very variable effects of s e l f -pollination on the trees studied. Some viable seed was also obtained from cones isolated and not pollinated. The seedlings are diploid and the possible causes of • agamospermy are discussed. A study of the inbred progeny shows that self-pollination usually results i n seedlings which are smaller and less vigorous than those from controlled cross-pollination. The seedlings from wind-pollination were intermediate and the small size of some suggests that self-pollination may have occurred. The practical aspects of self-pollination i n relation to both the natural and a r t i f i c i a l regeneration of forest land are b r i e f l y discussed. 1 i v TABLE OF CONTENTS LIST OF TABLES vi:. LIST OF PLATES v i i i INTRODUCTION 1 PART 1. I CONTROLLED BREEDING TECHNIQUES 8 Isolation of the Female Cone 8 Pollination 11 Pollen Germination Tests 13 Pollen Vi a b i l i t y After Storage 14 II CYTOLOGICAL INVESTIGATIONS WITH TREE 1 20 General . 20 Microtechnique 21 Pollen Germination and Subsequent Development 22 Fert i l i z a t i o n and Pro embryo Development • 25 Embryo Development 28 III CYTOLOGICAL INVESTIGATIONS WITH TREE A. 40 Comparative Ovule Abortion After Self-Pollination 40 Fe r t i l i z a t i o n and Pro embryo Development 43 Embryo Development 45 Comparison of Embryo Collapse i n Trees 1 and A 57 IV THE POSSIBLE CAUSE OF EMBRYO COLLAPSE AFTER SELF-POLLINATION 60 Self-Incompatibility 60 Inbreeding Effect 65 V PARTHENOCARPY AND AGAMOSPERMY 69 • PART 2. I RESULTS OF THE 1953 PROGENY TEST.... 81 Comparative Height and Diameter Development 81 Variation i n Form 84 Variation between Progenies 86 Seedlings from Unpollinated Cones 86 II RESULTS OF THE 1955 PROGENY TEST 88 Seed Extraction and Cleaning . 88 Sowing .. 89 Germination . . • • 91 Seedling Mortality 93 Comparative Height Development......... 93 III SOME PRACTICAL ASPECTS OF SELF-POLLINATION. 101 Natural Regeneration •• 101 A r t i f i c i a l Regeneration • 102 BIBLIOGRAPHY 105 V I TABLES Table 1. Dates of Isolation and Pollination of Cones i n 1952-and 1954 13 Table 2. Results of Pollination with Two-year-old Pollen....... 16 Table 3. Results of Pollination with Fresh and One-year-old Pollen 17 Table 4. Pollen Development After Self- and Cross-Pollination.. 24 Table 5. Development i n Archegonia After Self- and Cross-Pollination. Tree 1.... 26 Table 6. Development of Embryos After Self-Pollination. June 9. Tree 1 30 Table 7. Development of Embryos After Self-Pollination. June 16. Tree.l 31 Table 8. Development of Embryos After Self-Pollination. . . , June 21. Tree 1 32 Table 9. Development of Embryos After Self-Pollination. June 28. Tree.l... 34 Table 10. Development of Embryos After Self-Pollination. July 7. Tree 1 35 Table 11. Development of Embryos After Self-Pollination. . July 13. Tree 1........ 36 Table 12. The Number of Embryos After Self-Pollination. Tree 1.. 37 Table 13. Comparative Ovule Abortion i n Pollinated and Unpollinated Cones...... 43 Table 14. Development i n Archegonia After Self- and Cross-Pollination. Tree A.... 44 Table 15. Development of Embryos After Self-Pollination. June 16. Tree A 45 Table 16. Development of Embryos After Self- and Cross-Pollination. June 20. Tree A.. 46 Table 17. Development of Embryos After Self- and Cross-Pollination. June 23. Tree A.. 47 Table 18. Development of Embryos After Self- and Cross-Pollination. June 28. Tree A.. 49 v i i Table 19. Development of Embryos After Self- and Cross-Pollination. July 2. Tree A... 51 Table 20. Development of Embryos After Self- and Cross-Pollination. July 8. Tree A... 52 Table 21. Number of Embryos After Self- and Cross-Pollination. Tree A 54 Table 22. Germination Test on Seed from Unpollinated Cones 74 Table 23. Frequency Distributions and Coefficients of Variation. by Height Classes 83 Table 24. Frequency Distributions and Coefficients of Variation by Diameter Classes 83 Table 25. Results of Seed Extraction and Cleaning After Different Pollinations. 1955... 89 Table 26. Summary of Germination Results. 1955 92 Table 27. Summary of Seedling Mortality. 1955 93 Table 28. Frequency Distributions and Coefficients of Variation by Height Classes..;..;. 96 V X 1 1 PLATES To face page Plate I. Development of the Male and Female Cones of Pseudotsuga 13 Plate II. Pollination Technique and Pollen Germination 19 Plate III. Development from Pollen Germination to Proembryo After Self-Pollination. Tree 1... 27 Plate IV. Early Embryo Development After Self-Pollination. Tree 1 38 Plate V. Embryo Collapse After Self-Pollination Tree 1 39 Plate VI. Early Embryo Development After Self- and Cross-Pollination. Tree A 55 Plate VII. Comparative Embryo Development After Self- and Cross-Pollination. Tree A........ 56 Plate VIII. Comparative Embryo Growth Following Self- and Cross-Pollination. Tree 1 and A.. 59 Plate IX. Ovules and Embryos from Isolated and Unpollinated Cones 80 Plate X. Inbred Seedlings from Tree 2 97 Plate XI. "Apomictic" Seedlings and Comparative Development of Seedlings from Wind-Pollination. Tree 2. 98 Plate XII. Inbred Seedlings from Trees C and 2 99 Plate XIII. Comparative Heights and Diameters of the Progeny of Tree 2, from 1953 to 1955 100 i x ACKNOWLEDGEMENTS , The writer would l i k e to acknowledge his indebtedness to Dr. J. W. Duffield, formerly of the Institute of Forest Genetics, California, who suggested the topic for this dissertation. The work for this dissertation was begun while the writer was at the University of British Columbia and completed while he was i n the employment of the British Columbia Forest Service. Grateful acknowledgements are expressed to Mr. R. H. Spilsbury, Forester-in charge of the Research Division for his continued support over the latter period of the work. The valuable criticisms and encouragement of Dean G. S. Allen, Faculty of Forestry, and Dr. A. H. Hutchinson, Professor Emeritus, Department of Biology and Botany, University of Bri t i s h Columbia, are gratefully acknowledged. Helpful advice was also given the writer by Dr. V. C. Brink, Professor of Agronomy, University of British Columbia. The writer would also l i k e to express his appreciation of the help and advice on s t a t i s t i c a l analysis and experimental design which was given him by Mr. A. R. Fraser, Assistant Forester, Research Division, British Columbia Forest Service. The main part of the cytological work was completed i n the Victoria Laboratory of the Forest Biology Division, Science Service, Canada Department of Agriculture and the ready cooperation of Mr. R. Lejeune and Dr. R. E. Foster, Officers-in-Charge of the Forest Zoology and Forest Pathology Units respectively, i s gratefully X acknowledged. Photographic assistance was kindly provided by Mr. R. Banyard, Photographer for the Forest Biology Division, who prepared the plates and also photographed plate I, figures 1 to 5, plate II, figures 1 to 4, plates VIII and XIII. Miss B. Davies, Photographer, Public Relations Division, B r i t i s h Columbia Forest Service, photographed plates X, XI and XII, and her help was much appreciated. The graphs i n plates VIII and XIII and the plate numbers were drawn by Mr. J. Firth, Draughtsman, Surveys Division, B r i t i s h Columbia Forest Service, whose assistance i s gratefully acknowledged. The writer would l i k e to express his appreciation to Mrs. I Guthrie, Librarian, Public Relations Division, and Mr. J. Long, Nursery Superintendent, Reforestation Division, B r i t i s h Columbia Forest Service for assistance i n various aspects of the study. The financial assistance of the National Research Council, Ottawa, for the"period of 1952 to 1953 and their willingness to extend this assistance i s very gratefully acknowledged. The writer would f i n a l l y l i k e to acknowledge his appreciation of the unfailing support of his wife throughout the entire study. INTRODUCTION The mating systems of plants and animals have interested breeders for many years, particularly i n respect of the relative merits of inbreeding and outbreeding, the one favouring reproduction by closely related parents and the other by parents not closely related. An even closer form of inbreeding i s possible i n hermaphroditic plants as self-pollination results i n fusion of the male and female gametes from the same plant. It was soon appreciated, however, that certain plants produced no viable seed at a l l after self-pollination and, according to East (1929), Kolreuter i n 1746, was the f i r s t person to make this very significant observation. Darwin (1900, page 439) later carried out extensive selfing and crossing experiments with many plants and was able to demonstrate that selfing not only resulted i n greatly reduced seed y i e l d but also reduced vigour i n the progeny. He concluded that, generally, cross-fertilization was beneficial and s e l f - f e r t i l i z a t i o n injurious, at least with the plants with which he had experimented. It was not u n t i l comparatively recently, however, that the f u l l significance of these earlier findings was appreciated. According to Dobzhansky (1951, page 300), the appearance of sexual reproduction was perhaps the most important advance i n the evolution of l i f e and the simplest form of retrogression of sexuality i s s e l f - f e r t i l i z a t i o n . He explains that the danger of s e l f - f e r t i l i z a t i o n i s inherent i n hermaphroditism and that many hermaphrodites have developed various devices to prevent selfing, although some s t i l l reproduce facultatively or obligatorily by this means. Many plant species have developed f l o r a l mechanisms, such as protandry and protogyny, to prevent s e l f - f e r t i l i z a t i o n but i t has only been realized comparatively recently that several very efficient genetic systems have also been evolved for the same purpose. These systems have been extensively investigated i n many of the Angiosperms and, as explained by Bateman (1952), their primary function i s to prevent s e l f - f e r t i l i z a t i o n . He states that they d i f f e r from the many devices that interpose a mechanical barrier to self-pollination i n that they act at a later stage by interposing a physiological  barrier between self-pollination and s e l f - f e r t i l i z a t i o n . The effects of inbreeding on the progeny have also been extensively investigated and Darwin's conclusions that selfed progeny had less vigour than those that had been crossed, have been confirmed with many different species. Later investigators, i n addition, have been able to appreciate the genetic interpretation of inbreeding and have used this to great advantage. There have, however, been remarkably few detailed studies on either self-incompatibility or the effects of inbreeding i n the Gymnosperms. It i s true that there have been some minor studies which have involved the selfing of a few trees and a comparison of the y i e l d of viable seed with that from either wind or controlled cross-pollination. In a few of these studies, the seeds have been sown and the development of the inbreds compared with those of the controls. Such investigations have been almost exclusively confined to the Pinaceae and the most interesting w i l l be b r i e f l y summarized. One of the earliest studies was that of Sylven (1910), who - 3 -selfed five Picea abies, and found that only four of the trees produced any viable seed, ana that the yield was less than that obtained from wind-pollination. These seeds were sown and i t was found that the mortality rate of the inbreds was much the higher for the first few-years after germination. Langlet (1940) later reported that at twenty-seven years of age, the average height and diameter of these trees were less than half of those resulting from wind-pollination, but that the effects of inbreeding varied from tree to tree. There had also been considerable mortality among the inbreds. Kolessnikoff (1929) selfed a number of Pinus sylvestris and obtained no viable seed at a l l . Dengler (1932) carried out extensive selfing studies both with this species and with Pinus montana and found that either no viable seed was produced or that the yield was much below that obtained from seed from controlled cross-pollination. He suggested that some repeated inhibition mechanism must occur, either at syngamy or during early growth of the embryo, and that i t must check normal development. Dengler (1939) later reported that development of the inbred trees varied greatly at the age of seven years; some were comparable to the crossbreds whereas others were poorly shaped and stunted in growth. These trees were examined again by Scamoni (1950), who reported that the mortality of the inbreds had been heavy and that the majority were greatly inferior in growth and form to the crossbreds. In 1938, Langner (1951) selfed seven Larix leptolepis and Larix europaea and found that only three of these trees produced any viable seed. Langner (1952) later found that there had been - i n -considerable mortality among the inbreds and that growth i n both height and diameter was markedly inferi o r to that of the individuals from controlled cross-pollination. He also observed that there was much variation i n the effects of inbreeding i n the case of individual trees, the progeny of one tree showing much more evidence of inbreeding depression than did the others. Langner concluded that this parent tree must have contained a large number of recessive genes which were responsible for the crippled growth of the seedlings and that these genes were exposed by selfing. Von Gothe (1952) selfed other trees of the same species and obtained no viable seed from Larix europaea and only a small amount from Larix leptolepis. Von Gothe (1953) later reported that the inbreds lacked vigour and were a l l k i l l e d by frost the f i r s t winter. These studies a l l indicate that selfing results i n the production of l i t t l e or no viable seed and of progeny that i s usually inferior i n both growth and vigour. The effects of selfing, however, appear to be variable since other studies have shown that i t does not always result i n reduced seed yields or i n weak progeny. Piatnitsky (1936), for example, selfed five species of Larix. and reported that there was no evidence of " s e l f - s t e r i l i t y " i n four of these trees, and germination of the inbreds was only lower i n one tree compared to that from wind-pollination. Syrach Larsen (1937, page 72) reported that there was no evidence of inbreeding depression i n the four-year-old seedlings from s e l f - and controlled cross-pollination i n Thu.ja plicata; the y i e l d of viable seed from selfing i n this study, however, was substantially below that obtained after cross-pollination. Forshell (1953) found that - 5 -selfing i n Pinus sylvestris usually produced less viable seed than did controlled cross-pollination; i n the case of one particular tree, however, selfing resulted i n more viable seed than did crossing. The evidence from past studies definitely suggests that there i s some impediment to normal seed development following selfing, and that i t usually results i n progeny which i s neither vigorous nor well formed. The question naturally arises as to whether the effects of selfing on the production of viable seed could be caused by incompatibility systems similar to those found i n the Angiosperms. It i s interesting, therefore, to observe that there are two opinions as to whether such systems are possible i n the Gymnosperms. Whitehouse (1950), for example, stated that the two main outbreeding devices i n the Gymnosperms are dichogamy and dioecism and that, although these both occur i n the Angiosperms, other outbreeding devices are found which are based on the incompatibility of pollen and style. He wrote: "These are worthy of special attention since no comparable mechanisms are known or indeed appear to be possible i n Gymnosperms". Whitehouse Later considered the p o s s i b i l i t y of a physiological antagonism developing between the cells of the carpel and the pollen tubes of similar genetic constitutions. He added, however; Nevertheless, i t appears unlikely that such an antagonism could attain sufficient selective value i n a Gymnosperm, since the thickness of nucellar tissues which the pollen-tubes must penetrate to reach the embryo-sac i s hardly sufficient to be capable of interposing a physiological barrier. - 6 -Bateman (1952) did not agree with these conclusions and stated: Reference to Coulter and Chamberlain (1910), shows that i t is typical of Gymnosperms for pollen tubes to be compelled to grow through a well-developed nucellus before fertilization i s possible. Since in Capsella ... and Parthenium ... a single layer of cells in the stigma is sufficient to inhibit self-pollen, the massive nucellus of most Gymnosperms should be quite capable of acting as a physiological sieve for pollen tubes. The absence of a style in the Gymnosperms i s not therefore a sufficient explanation of the absence of self-incompatibility from this phylum (if i t i s indeed absent). Bateman also added this footnote: "In a private communication Dr. A. Gustafsson claims to have found self-incompatibility in Pinus sylvestris. thus proving that the absence of a style does not prevent Gymnosperms from evolving self-incompatibility". It can be appreciated, therefore, that there was considerable scope for a cytological study to determine whether or not such an incompatibility system exists. According to Sears (1937), incompatibility systems in the Angiosperms can occur in three distinct stages. (1) Pollen germination may be decreased, in which, case the inhibitory reaction occurs only in the stigma. (2) Pollen germination i s normal but pollen tube growth i s inhibited in the style. (3) The pollen tubes grow normally, reach and fertilize the ovules but no seed develops. A cytological study, therefore, should determine whether or not any of these stages were present in a Gymnosperm. Past studies by Allen (1942), Duffield (1950), and Meyer (1951), had a l l indicated that the selfing of Pseudotsuga menziesii had invariably resulted in l i t t l e or no viable seed being produced. The species was, therefore, considered " s e l f - s t e r i l e " , so that i t was a very suitable choice for an investigation of this nature. It should be noted that the term self-incompatible has been used throughout the following pages i n place of that of s e l f - s t e r i l e as i t i s the more correct. Lewis (1954) has recently pointed out that: The term s e l f - s t e r i l i t y was originally applied to a mating system i n hermaphrodite plants, a l l individuals of which produce normally functioning male and female gametophytes and which f a i l to produce zygotes after s e l f - f e r t i l i z a t i o n . But the discovery that certain cross-pollinations also f a i l e d to produce zygotes, and the association of the term s t e r i l i t y with the lack or i n v i a b i l i t y of one or other of the two gametophytes has lead to the term incompatibility. Several authors, quoted i n this dissertation, have used the older term " s e l f - s t e r i l i t y " , and i n such cases i t has been retained. Furthermore, since no studies on the development of any inbred seedlings had previously been conducted, this represented another new f i e l d of investigation. This dissertation, therefore, has been divided into two parts; the f i r s t deals with the cytological study and the second part with f i e l d studies of the progeny from the trees investigated. In addition, reference i s made i n both parts to the possible occurrence of Agamospermy. PART 1. I. CONTROLLED BREEDING TECHNIQUES Isolation of the Female Cone It should be noted that throughout this investigation, the microsporangiate and megasporangiate s t r o b i l i have been referred to as male and female cones respectively. McLean and Cook (1951* page 673) have stated that although these reproductive organs are sporophytic i n origin and, therefore, i n no way sexual i n character, i t i s customary to refer to them i n this manner. The f i r s t requirement for controlled breeding i s that the female cone must be effectively isolated from any contamination by unknown sources of pollen. The development of the cones, therefore, has to be carefully checked i n order that they are isolated at the right time. Pseudotsuga i s monoecious and the male and female cones are often found on the same branchlet, although there i s a general tendency for the female cones to be more numerous i n the upper half of the crown. The appearance of the male and female cone buds i n late February i s shown i n figure 1, plate I, the female buds are both larger and more pointed than the male buds and occur l a t e r a l to the terminal vegetative bud. A more advanced stage i s shown i n figure 2; the male buds are now considerably swollen and the female buds should be isolated at this time for later controlled pollination. In figure 3, the male cones have emerged from the bud scales and w i l l soon mature, while the female buds are much enlarged and the cones are beginning to emerge. Exposure of female - 9 -buds as late as this stage would involve some risk of contamination since pollen from more precocious trees could now settle on the female cone. There are no specific dates for isolation since development of both male and female cones varies greatly from tree to tree and even on the same tree. Duffield (1950) found that the opening of Pseudotsuga cone buds was apparently more sudden and synchronized over a single tree than i n Pinus. No such synchronization was observed i n the trees i n this investigation, however, and i t was found that the female buds at the top of the tree and on southerly aspects invariably had to be isolated f i r s t . The dates of both isolation and pollination are given i n table I, which shows that neither of the operations could be completed i n one day, owing to the varied state of development of the female cones i n different parts of the tree. Constant inspection of the female cones i s the only safe method of insuring that isolation and pollination are completed at the correct time. There was also some variation i n development between l o c a l i t i e s , the trees A , C, and D, located near Victoria,, had to be isolated earlier than those at Cowichan Lake. Viscose sausage casing was used for isolating the female buds and the bags were constructed i n the manner recommended by Duffield (1950)j figure 2, plate II, il l u s t r a t e s a cone isolated by this means. These bags are cheap, l i g h t , and transparent^and development can be readily observed; they are also permeable to moisture and can withstand the variable spring weather. There i s no chance of pollen contamination, provided that the cones are isolated at the right - 10 -time and that the bags are not removed too soon after pollination; these conclusions are supported by those of Nienstaedt and Kriebel (1955)• The one disadvantage of the viscose bags is that they readily absorb heat which results in some cone mortality. Duffield (1950) noted that Pseudotsuga was more susceptible to heat injury within the bags than were most species of Pinus; Nienstaedt and Kriebel (1955) found that the internal temperatures, were high compared with those within other types of bags. This heat injury can be substantially reduced by restricting the selection of cones to the north side of the tree and by avoiding south-facing aspects. The mortality is shown in table I, and in some trees i t was quite serious. In the majority of cases, the female cones died prior to emergence from the bud scales; in other instances, they died soon after pollination. The cone i s at a critical stage of i t s development at the time of year when isolation and pollination takes place and some mortality must be expected. Each isolation bag was identified with a light aluminum tag which indicated i t s number and the number of cones contained, and whether they were to be self- or cross-pollinated, or to be left unpollinated. This information was recorded on the special form illustrated in figure 1, plate II, which was developed from the one described by Gumming and Righter (1948). Care was taken to remove a l l male buds which might otherwise have been isolated with those of the female. Male cones were left inside the bags only in the case of female cones which.were to.be self-pollinated. -11 -Pollination Pollen for controlled breeding must be available at the time when the female cones are most receptive; branchlets with male cones, therefore, were removed at the time that the female cones were isolated. These were brought indoors and placed i n water i n a warm room to accelerate development of the cones. Paper was spread beneath each container and the pollen was readily collected when the cones matured. In order to avoid any possible contamination, this operation was conducted i n a separate room for each pollen l o t . The pollen was f i n a l l y placed i n flasks which were l i g h t l y plugged with absorbent cotton and stored at 4° C. u n t i l required. In the meantime, the isolated female cones were periodically checked so that they could be pollinated when most receptive. It was considered that the stage i l l u s t r a t e d i n figure 4, plate I, was the most favourable for pollination as the ovuliferous scales were f u l l y extended. Pollinations were made by means of a hypodermic syringe f i t t e d with a rubber bulb and a pollen agitator as recommended by Duffield (1950). The operation i s i l l u s t r a t e d i n figure 3, plate II. The end of the needle was inserted through the viscose casing bag and a cloud of pollen blown directly on to the cones, then the small hole i n the bag was sealed with scotch tape. The pollen parent and date of pollination were recorded on the special form (figure 1, plate II) and a coloured chicken ring was slipped on the branchlet at the base of the bag. This showed that the enclosed cones had been pollinated - 12 -and the colour indicated the type of pollination. In this investigation, red rings indicated self-pollination and yellow cross-pollination; green rings indicated that the cones had been purposely l e f t unpollinated. The coloured rings greatly simplified later sampling and prevented mistakes. It can be seen from table I that pollination of the cones of any particular tree generally required several days to accomplish and was not completed u n t i l a l l the cones had reached the receptive stage; this varied i n different parts of the tree. When the scales on the cone had closed and i t had begun to invert, the isolation bag was removed and replaced by a protective kraft bag. Figure 5, plate I, i l l u s t r a t e s a cone whose scales have closed so that pollination i s no longer possible. The dates on which these bags were removed also varied for the different trees as indicated i n table I. The kraft bags were necessary for protection against insects and squirrels and had no adverse effect on the development of the cone. Ordinary kraft paper bags were used but the seams were a l l stapled and reinforced with waterproof glue; one of these bags i s i l l u s t r a t e d i n figure 4, plate II. Figure 6, plate I, shows a sixty-foot tree with the isolation bags i n place. - 13 -TABLE I. DATES OF ISOLATION AND POLLINATION OF CONES IN 1952 AND 1954. Tree No. of No. of Dates of Dates of Dates of No. of bags cones isolation pollina- bag cones isolated tion removal collected 1* 47 100 11-15/3 30/3-7/4 27-28/4 74 2* 30 105 15-23/3 27/4 76 A 73 310 22/3-13/4 16/4-3/5 1-9/5 , 224 C 57 202 25/3-6/4 13-16/4 24/4-3/5 104 D 71 271 26/3-10/4 15-26/4 4-8/5 126 F 47 280 29/3-20/4 29/4-5/5 14-15/5 212 H 58 278 31/3-20/4 21-29/4 14/5 179 J 48 268 1-2/4 21-29/4 13-14/5 220 * Trees 1 and 2 were located i n Vancouver and pollinated i n 1952. Pollen Germination Tests In a controlled breeding study, the pollen should always be tested prior to use i n order to ensure that i t i s viable, since s t e r i l e pollen i s not uncommon. No previous germination studies had been conducted with Pseudotsuga pollen and different types of media had to be t r i e d before a satisfactory one was found. The method recommended by Righter (1939) for germinating Pinus pollen was not satisfactory since Pseudotsuga pollen germinates poorly i n d i s t i l l e d water. Various concentrations of both sucrose and agar were t r i e d and the most satisfactory medium was f i n a l l y found to be two per cent bacto-agar. This was heated and then poured into the petri dish to be used for the germination test. When the agar had cooled, pollen from one particular l ot was l i g h t l y scattered on i t i n five different places and a few drops of water were placed on the inside of the l i d i n order to maintain a high humidity. The petri dish - 14 -and i t s contents were then incubated for twenty-four hours at 26° C. and at the end of this period, the five areas of pollen were examined under a low-power microscope. A f i e l d of f i f t y pollen grains was selected at random i n each area and the amount of germination was checked. Pollen was considered to have germinated when i t had swollen and the outer exLne layer had been ruptured; there was usually some protrusion of the pollen tube as well. The vigour of different pollen lots varied considerably as clearly i l l u s t r a t e d i n figures 7 and 8 , plate II. These two pollen lots from trees D and F respectively were collected i n 1954 and the germination test was conducted one year later. Only one pollen grain has germinated i n the pollen v i s i b l e i n figure 7; the pollen from this tree was never particularly vigorous and i t s v i a b i l i t y declined far more rapidly i n storage than did any of the others collected at the same time. It i s quite logical; however, to expect that there w i l l be a considerable variation i n pollen v i a b i l i t y from one tree to another and the need for germination tests when a r t i f i c i a l crosses are to be made i s indicated. Figure 5 , plate II, i l l u s t r a t e s typical Pseudotsuga pollen grains; these are wingless, bright yellow, about eight microns i n width and roughly e l i p t i c a l i n shape. Their general appearance i s very similar to that of a coffee bean, and the characteristic depression on the one side i s probably an adaptation for wind dissemination. Pollen V i a b i l i t y After Storage No previous work had been carried out on the storage requirements - 15 -of Pseudotsuga pollen although storage may be often necessary, since controlled crosses can not always be made the same year that pollen i s collected. Some exploratory work was, therefore, carried out between 1952 and 1955. Periodic germination tests were conducted during this period on several different pollen lots which were stored i n l i g h t l y stoppered flasks at 4° C. No attempt was made to control either temperature or humidity as the purpose of the test was to find out how long Pseudotsuga pollen would remain f u l l y viable i f stored i n an ordinary refrigerator. The results showed a gradual decline i n germination capacity of pollen over a three year period. This i s ill u s t r a t e d i n figure 6, plate II; there i s s t i l l some v i a b i l i t y i n t h i s two-year-old pollen but i t i s very low and considerable desiccation i s evident. In 1954> pollen from three two-year-old lots was used for a r t i f i c i a l pollination. The results are shown i n table 2 and i t can be seen that although a l l the pollen lots showed some germination, only one viable seed was produced. Some cones were also removed from the parent trees a few weeks after pollination i n order to check on pollen development. A few ovules were sectioned and i n one, ungerminated pollen was found twenty-three days after pollination, while i n another, an ungerminated pollen grain was found thirty-two days after pollination. This clearly indicated that the results from the germination tests used do not necessarily indicate the extent to which the pollen i s capable of f e r t i l i z a t i o n . - 16 -TABLE 2. RESULTS OF POLLINATION WITH TWO-YEAR-OLD POLLEN. Tree Pollen Pollen No. of No. of Germin.. Cutting tests parent germin.* cones cleaned of ungermin. collected seeds seed A 3 120/250 8 1 Seeds empty-4 128/250 22 6 Seeds empty D 2 54/250 1 _ — 5 191/250 5 8 1 1 with embryo F 4 128/250 2 4 - Seeds empty * This test was made April 1954 and was based on five samples In 1955> a similar test was conducted with pollen which had been stored for one year and, in this case, a direct comparison could be made since the same lots were used on the same trees in 1954. The results are shown in table 3 and indicate, that although neither of these pollen lots showed any significant decrease in germination, there was a highly significant decrease in the amount of viable seed produced. There is l i t t l e doubt that Pseudotsuga pollen loses i ts fertilization potential rather quickly and that improved storage techniques should be sought. These results show that the a r t i f i c i a l germination tests used do not necessarily indicate whether or not the pollen is capable of fertilization. Duffield (1954) has recently pointed out the l i t t l e that i s actually known about the physiology of gymnosperm pollen and the unreliability of such tests. He adds: However, i t ... seems reasonable to assume provisionally that pollen which cannot be made to germinate by usual means or which, on germination, shows unusually poor tube.growth, is likely to be ineffective in causing* - 17 -fertilization. Herein lies the present practical value of pollen testing as a routine procedure in breeding operations. TABLE 3. RESULTS OF POLLINATION WITH FRESH AND ONE-YEAR-OLD POLLEN Tree Pollen parent Year of pollin. Pollen germin.* Gone no. No. cleaned seeds Seeds germin. Cutting test of ungermin. seed F H 1954 240/250 33 420 400 mm 1955 244/250 4 67 2 9 with embryos J C 1954 230/250 41 1051 397/420 _ 1955 240/250 13 555 39 96 with embryos * This test was based on five samples - 18 -PLATE I Figure 1* Appearance of the male and female cone buds of Pseudotsuga, Victoria. February 23, 1954. Natural size. Figure 2. Later development, the male buds are considerably swollen and the cones w i l l soon emerge through the bud scales. Natural size. Figure 3. The female buds should be isolated before this stage i s reached, as the cones are emerging and the male cones are almost mature. Natural size. Figure 4* Pollination should take place at this stage. Natural size. Figure 5. The cone scales are closed and pollination can no longer take place. Natural size. Figure 6. A sixty-foot tree with the isolation bags i n position. March, 1954. PLATE I PLATE II Figure 1. A pollination record form. Figure 2. An isolated female cone; ample space has been l e f t i n the bag for future development, and a small strip of absorbent cotton has been stapled into the end of the bag to prevent pollen contamination. Approx. l/6 natural size. Figure 3. A young cone being a r t i f i c i a l l y pollinated with the modified hypodermic syringe. Figure 4. The kraft bag used for protection against squirrels. Approx. 1/6 natural size. Figure 5. Pollen grains of Pseudotsuga. X 92. Figure 6. Poor germination of two-year-old pollen from tree 1. Nov. 1954. X 92. Figure 7. Poor germination of one-year-old pollen from tree D. March, 1955. X 92. Figure 8. Vigorous germination of one-year-old pollen from tree F. March, 1955. X 92. P L A T E II - 20 -II CYTOLOGICAL INVESTIGATIONS WITH TREE 1 General The effects of self-pollination on seed v i a b i l i t y cannot be determined u n t i l controlled pollinations have been conducted and a germination test made on the seed. In 1952, Orr-Ewing (1954) s e l f -pollinated two Pseudotsuga near Vancouver and found that one of these trees (tree l ) produced very l i t t l e viable seed, only two seeds germinating from th i r t y - f i v e cones. The germination of seeds resulting from cross-pollination, however, was considerably higher. It was evident that a cytological examination of the ovules from this tree would show whether a true self-incompatibility system was responsible for the low yi e l d of seed. Such a system, however, could be expressed at several different stages of development, as has previously been demonstrated by Sears (193.7) i n the Angiosperms. It was possible, therefore, that the pollen grains of Pseudotsuga might not germinate after self-pollination or, again, that the development of the pollen tube might be inhibited. Bateman (1952) pointed out that the massive nucellus of most gymnosperms could act as a physiological sieve for pollen tubes. This could also occur i n Pseudotsuga since the pollen tubes have to grow through the apex of the nucellus to reach the archegonium. The cytological investigation was, therefore, divided into the following distinct studies. F i r s t l y , a study of pollen germination and development to the time that the pollen tubes reached the archegonium. Secondly, a study of subsequent development from the time the two sperm nuclei - 21 -were discharged into the archegonium and up to formation of the proembryo. Finally, should there be found no self-incompatibility system operative i n these f i r s t two stages, a study of the development of the embryos would be indicated. Microtechnique These studies involved a considerable amount of cone sampling at different dates from the time of pollination. The usual procedure was to remove two cones at each date, one of these had been self-pollinated and the other cross-pollinated. The ovules were quickly dislodged from the ovuliferous scales, and the integuments also were removed i n the case of collections made towards the end of May when they had become hard and made sectioning d i f f i c u l t . The ovules were then placed i n either of two killing-and-fixing solutions and a i r was removed with a suction pump. Carney's f l u i d with chloroform (Johansen 1940, page 40), and Allen-Bouin III (Sass 1940, page 19), were both used and found equally satisfactory. The former had been used by Allen (1945)> i n an earlier investigation with Pseudotsuga. and the l a t t e r was recommended by Johansen (1954), who also advised the substitution of propionic for gl a c i a l acetic acid. The ovules were l e f t i n Carnoy's f l u i d for 12 hours, washed quickly i n 95 per cent ethyl alcohol and f i n a l l y stored i n 70 per cent alcohol. When the Allen-Bouin III solution was used, the ovules were placed i n the solution for 24 hours and then washed i n tap water and brought up i n stages to 70 per cent alcohol. They were stored for a short time in a refrigerator and then dehydrated and embedded by Johansen's tertiary butyl alcohol method (1940, page 130). Anhydrous isopropyl alcohol, however, was used in the place of absolute ethyl alcohol, again on Johansen's recommendations (1954)* Parawax was used for preliminary infiltration, with tissuemat replacing i t for the final embedding. The embedded ovules were finally labelled with the date, tree number, the type of pollination, and the killing-and-fixing solution used. Sections were cut at twelve microns on a rotary microtome, affixed with Haupt's adhesive (Johansen 1940, page 20) and finally stained with either Heidenhain's iron haematoxylin or Feulgen's stain. Both stains gave consistently good results. The best schedule for haematoxylin involved mordanting for thirty minutes in a 4 per cent solution of iron alum, followed by staining in 1 per cent haematoxylin for ten minutes, and then destaining in 2 per cent iron alum. Coleman's (1938) recommendations were followed for the preparation of Feulgen's stain. The material was hydrolized for ten minutes in 1 N. hydrochloric acid, stained in Feulgen for ten minutes and finally washed in three changes of sulphur dioxide water. A very dilute solution of fast green dissolved in absolute ethyl alcohol and clove o i l was used for the counterstain. After dehydration in xylene, the sections were finally mounted in Permount. Pollen Germination and Subsequent Development Sections were made from self-pollinated ovules which had been collected at different dates and these were compared with sections from cross-pollinated ovules collected at approximately the same time after pollination. Particular care was taken to obtain clear sections of the micropyle so that the pollen grains could be counted and checked for germination. If germination had occurred, their development was recorded under three headings. (1) The generative c e l l had not divided. (2) The generative c e l l had divided to form the body and stalk c e l l s . (3) The body c e l l was dividing to form the two sperm nuclei. A summary of the results i s shown i n table 4 and i t was obvious that neither pollen germination nor i t s subsequent development were inhibited after self-pollination since only twelve pollen grains were found not germinated. The remaining one hundred and ninety-four pollen grains had a l l germinated and were i n various stages of development. One of these i s i l l u s t r a t e d i n figure 1, plate III, twelve days after self-pollination; the generative c e l l has divided to form the body and stalk c e l l s and the vestigial prothallial c e l l s can be seen. Another germinating pollen grain, seventeen days after self-pollination, i s ill u s t r a t e d i n figure 2. The pollen tube i s prominent and the tube, body, stalk, and prothallial c e l l s are a l l v i s i b l e . It was also evident that the nucellus did not act as a physiological sieve for pollen tubes since numerous tubes were observed growing towards the archegonia. One of these i s il l u s t r a t e d i n figure 3, fifty-four days after self-pollination. A comparison of pollen development after s e l f - and cross-pollination - 24 -showed that there were no differences between the two at this stage. Some sections were also made from the self- and cross-pollinated ovules of tree 2, the other Pseudotsuga investigated i n 1952; these were removed only fiv e days after self-pollination but i n some ovules pollen germination had already occurred. The higher temperatures within the bags undoubtedly must have accelerated pollen germination i n the micropyle. The earliest germination recorded by Lawson (1909) for Pseudotsuga i n California was seven to ten days, whereas Allen (1943) i n British Columbia found that pollen remained about three weeks i n the micropyle before germination. Both these earlier investigations, however, were made from ovules collected after wind pollination. TABLE 4 . POLLEN DEVELOPMENT AFTER SELF- AND CROSS-POLLINATION TREE 1. Type of po l l i n . Days after No. of ovules No. of pollen Germin. Stage of pollen development poll i n . sampled grains Gener- Body & ative & stalk tube cells c e l l s Divis. of body c e l l Self n n it 12 17 26 27 15 2 15 13 37 9 84 76 36 9 76 73 17 2 4 0 19 7 72 70 3 45 206 194 Cross n 9 21 10 13 63 30 46 28 19 8 27 20 mm 23 93 74 - 25 -F e r t i l i z a t i o n and Pro embryo Development Sections were made from fifteen self-pollinated ovules which had been removed fif t y - f o u r days after pollination; the stages of development i n each archegonia have been summarized i n table 5* Syngamy and the subsequent divisions of the zygote were observed i n eleven of the ovules; i n two others, sperm nuclei had been discharged into the archegonia and were moving towards the egg nuclei. In the remaining two ovules, pollen tubes were either penetrating the nucellus or had already reached the archegonia. Sections were made also from four ovules removed fifty-eight days after self-pollination and early proembryo development was observed i n three of these. Development after self-pollination again compared very favourably with those from the cross-pollinated ovules. Eleven of the l a t t e r were removed f i f t y - f i v e days after pollination and syngamy and later development were observed i n only three of them. In five other ovules, pollen tubes were either penetrating the nucellus or had reached the archegonium. Figure 4> plate III, shows the two sperm nuclei moving towards the egg nucleus i n one of these self-pollinated ovules and figure 5 shows the sperm nucleus fusing with the nucleus of the egg. The subsequent division of the zygote i s ill u s t r a t e d i n figure 6 and i n figure 7, the two free nuclei have divided and three of the four resulting daughter nuclei can be seen moving towards the basal part of the archegonium. The early development of the proembryo i s illustrated i n figure 8. It i s quite evident from these figures that no physiological barrier to s e l f - f e r t i l i z a t i o n and development - 26 -of the proembryo had occurred. It was again apparent that the bags accelerated development as syngamy was observed fi f t y - f o u r days after pollination, whereas Allen (1943) found that two months was the usual time i n the case of wind-pollination. TABLE 5. DEVELOPMENT IN ARCHEGONIA AFTER SELF- AND CROSS-POLLINATION TREE 1. Type of Days No. of No. of Archegonia with these stages p o l l i n . after ovules archeg- ______________________________ p o l l i n . sampled onia Sperm Syngamy 1st & Proembryo nuclei 2nd develop-present zygotic ment divis.  Self 54 15 57 6 13 8 — H 58 4 14 - 1 — 5 Cross 55 11 40 5 1 1 1 - 27 -PLATE III Figure 1. A germinating pollen grain twelve days after self-pollination. A p r i l 14, 195.2. X 480. Figure 2. Vigorous pollen tube growth seventeen days after self-pollination; note the tube, body, stalk,. . and prot h a l l i a l c e l l s . A p r i l 19, 1952. X 480. Figure 3. A pollen tube penetrating the nucellus f i f t y - f o u r days after self-pollination. May 26, 1952. X 480 Figure 4* The two sperm nuclei i n the archegonium; one of these w i l l unite with the egg nucleus. May 26, 1952. X 345. Figure 5. Fusion of the male and female nuclei, note the disparity i n size. May 26, 1952. X 345. Figure 6. F i r s t division of the zygote. May 26, 1952. X 345. Figure 7. Three of the four free nuclei moving towards the base of the archegonium after the second division of the zygote. May 26, 1952. X 345. Figure 8. Early development of the proembryo, f i f t y days after self-pollination. June 1, 1952. X 345. P L A T E ET - 28 -Embryo Development The results of the examination of the previous stages of development had indicated that the impediment to normal seed development must occur after s e l f - f e r t i l i z a t i o n and at some time during development of the embryo. Sections were made, therefore, from self-pollinated ovules which had been removed at seven different lengths of time after f e r t i l i z a t i o n . For each of these collections, the number of embryos i n each ovule and the number of nuclei i n the terminal embryo were recorded. The number of nuclei used was that obtained by counting those v i s i b l e i n the embryo i n one focal plane. The terminal embryo was considered to be the dominant one of the ovule throughout this investigation. It was usually the largest and Buchholz (1918) found that, i n FJnus, i t was the successful one i n the struggle for supremacy among the embryos. He added that only very exceptionally was the successful embryo found to be the second one instead of the terminal. The terminal embryo, of course, has the advantage of being at the apex of the corrosion cavity whereas the other embryos are forced to develop behind i t . Buchholz (1946) pointed out that these other embryos became ensnared i n the suspensors of the terminal embryo and were gradually pushed back towards the micropyle. For each collection, the condition of the terminal embryo was carefully checked for any signs of collapse. It was f u l l y realized that killing-and-fixing could produce artifacts and that embedding and sectioning could damage the young embryos to the - 29 -extent that the material might be misinterpreted. The procedure throughout this investigation, therefore, was to classify embryos as "collapsed" only when this condition was quite obvious. The results of the examinations for each collection date are summarised as follows: (l) June 9. sixty-eight days after self-pollination. Fertilization was observed in a l l of the ten ovules sectioned; eight contained vigorous young embryos which were perfectly normal in appearance. One such embryo i s illustrated in figure 1, plate IV, which shows a female garnetophyte of normal appearance with sharply stained nuclei and active mitosis in evidence. It was interesting to find syngamy in the archegonia of four ovules which contained young embryos, particularly since only the proembryo stage was present in two other ovules. This variation in development within the same ovule and cone is not in agreement with the conclusions of Buchholz (1945), who concluded that fertilization occurred in Pinus on nearly the same day within a l l the cones of a number of individuals of the same species growing on the same site. In tree. 1,. however, syngamy was first observed on May 26th, two weeks prior to this date. It was also quite evident that fertilization in Pseudotsuga takes place over a considerable period among different trees growing on the same site. - 30 -TABLE 6. DEVELOPMENT OF EMBRYOS AFTER SELF-POLLINATION. JUNE 9. TREE 1. Ovule No. of embryos No. of nuclei No. per in ovule terminal embryo 1 1 2 2 1 2 3 2 3 4 2 4 5 2 2 6 2 2 7 2 3 3 3 4 (2) June 16. seventy-3ix days after self-pollination. Eight ovules were sectioned and six of these contained embryos. In two of these ovules, however, the fi r s t signs of embryo collapse were noted. This early state of collapse was characterized by cel l shrinkage, excessive vacuolization, and very dense staining with haematoxylin. The entire cell contents were commonly stained densely as shown in figures 2 and 3, plate IV. Both of these collapsing embryos were small but s t i l l comparable in size to the other terminal embryos and i t i s significant that mitosis was observed in only one of these normal embryos. - 31 -TABLE 7. DEVELOPMENT OF EMBRYOS AFTER SELF-POLLINATION. JUNE 16. TREE 1. Ovule No. of embryos No. of nuclei No. per i n ovule terminal embryo 1 2 7 2 3 7 3 4 8 4 3 9 5 3 X 6 3 X X signifies early embryo collapse (3) June 21, eighty days after self-pollination. Thirty-one ovules were sectioned; twenty-five of these contained embryos and the remainder had not been f e r t i l i z e d . Embryo collapse had definitely increased as witnessed by the collapse of seven terminal embryos. In addition, some of the remaining embryos lacked vigour and mitosis was observed i n only seven of them; i n three of these mitosis appeared abnormal with considerable clumping of the chromosomes. Careful examination of the ovules throughout this investigation showed that collapse was general whenever the terminal embryo had collapsed, and that there was never any question of a second embryo taking i t s place. Some of the sound embryos had grown considerably since the last date and figure k, plate IV, illustrates one of the largest of these. Another terminal embryo which had stained abnormally and was on the point of collapse i s illu s t r a t e d i n figure 5; two collapsed embryos are shown i n figure 6. The.cells of the garnetophyte at this date showed evidence of becoming f i l l e d with densely staining material, and this was true regardless of the condition of the embryos. TABLE 8. DEVELOPMENT OF EMBRYOS AFTER SELF-POLLINATION. JUNE 21. TREE 1. Ovule No. of embryos No. of nuclei No. per i n ovule terminal embryo 1 1 4 2 2 7 3 2 8 4 4 11 5 5 11 6 3 12 7 3 12 8 2 14 9 3 14 10 3 15 11 2 16 12 3 19 13 2 24 14 5 24 15 3 26 16 2 30 17 4 33 18 4 34 19 1 X 20 1 X 21 1 X 22 2 X 23 2 X 24 3 X 25 4 X X signifies early embryo collapse (4) June 28. eighty-seven days after self-pollination. This sample was interesting since i t could be compared with one resulting from controlled cross-pollination eighty-four days previously. Twenty-one self-pollinated ovules were sectioned and - 33 -i n only three of these could the terminal embryos be considered normal. These embryos were small, no mitosis was observed i n them, and they were i n marked contrast to those resulting from cross-pollination. The largest of these latt e r embryos i s i l l u s t r a t e d i n figure 8, plate IV; i t showed a considerable number of mitoses and eighty-two nuclei i n one focal plane, compared with twenty-nine nuclei i n the largest of the embryos resulting from s e l f -pollination. Figure 7 i l l u s t r a t e s a selfed embryo i n a state of collapse; i t s cells and nuclei are abnormally large and the c e l l contents show considerable shrinkage. Fifteen of the remaining ovules contained embryos i n the advanced state of collapse (see figure 1, plate V). This condition i s characterized by complete c e l l collapse and erratic staining of the nuclei. The remains of three collapsed embryos can be seen i n figure 2; the gametophyte tissue, however, appears to be perfectly normal, i t s nuclei stain sharply and i t s cells contain a considerable amount of deeply-staining material. The data i n table 9 indicate! that a considerable difference exists i n the development of the dominant embryos i n different ovules of the same cone. In the embryos from the cross-pollinated ovules, for example, the number of nuclei v i s i b l e i n one optical plane varies from thirteen to eighty-two. Similar variation i n embryo size was reported for Pinus by Buchholz (1946), who suggested that i t may be due to genetic variation and, therefore, may be a measure of inherent embryonic vigour. - 34 -TABLE 9. DEVELOPMENT OF EMBRYOS AFTER SELF- AND CROSS-POLLINATION JUNE 28. TREE 1. Self-pollination Cro ss-pollination Ovule No. of No. of nuclei Ovule No. of No. of nuclei No. embryos i n terminal No. embryos i n terminal per ovule embryo per ovule embryo 1 4 11 1 2 13 2 3 19 2 1 16 3 1 29 3 1 16 4 1 X 4 4 40 5 1 X 5 4 40 6 2 X 6 2 82 7 2 Y 8 2 Y 9 2 Y 10 3 Y 11 3 Y 12 3 Y 13 3 Y 14 3 Y 15 3 Y 16 3 Y 17 3 Y 18 4 Y 19 4 Y 20 4 Y 21 4 Y X signifies early and Y advanced embryo collapse (5) July 7. ninety-six days after self-pollination The completeness of embryo collapse i n tree 1 after sel f -pollination i s well i l l u s t r a t e d by this sample for, although the twenty-four f e r t i l i z e d ovules which were sectioned a l l contained embryos, everyone of these had collapsed. Figures 3 and 4, plate V, . show the same section under different magnification; i t i s evident that the gametophyte cells are s t i l l f u l l of dark-staining material, yet the terminal and other embryos have a l l collapsed. A zone of apparently depleted c e l l s surrounds the corrosion cavity but such - 35 -was invariably present at this stage regardless of the type of pollination. The occurrence of a similar zone i n Larix was reported by Schopf (1943). Other collapsed terminal embryos appear i n figures 5 and 6; here the nuclei stain rather f a i n t l y and the garnetophyte cells are beginning to collapse. TABLE 10. DEVELOPMENT OF EMBRYOS AFTER SELF-POLLINATION. JULY 7. TREE 1. Ovule No. of embryos No. of nuclei No. per i n ovule terminal embryo 1 1 Y 2 1 Y 3 2 Y 4 2 Y 5 2 Y 6 2 Y 7 2 Y 8 2 Y 9 2 Y 10 2 Y 11 2 Y 12 2 Y 13 3 Y 14 3 Y 15 3 Y 16 3 Y 17 3 Y 18 3 Y 19 3 Y 20 3 Y 21 4 Y 22 4 Y 23 4 Y 24 5 Y Y signifies advanced embryo collapse (6) July 13, one hundred days after self-pollination. . Of the six ovules sectioned, the embryos of five had collapsed - 36 -completely; the garnetophyte tissue was beginning to break down and the dark staining material i n i t s c e l l s was oiminishing. One of these terminal embryos i s ill u s t r a t e d i n figure 7> plate Vj collapse i s complete and there i s very l i t t l e staining. In the remaining ovule, however, a well developed embryo was found; i t i s i l l u s t r a t e d i n figure 8 and was definitely past the stage where embryo collapse usually occurred. TABLE 11. DEVELOPMENT OF EMBRYOS AFTER SELF-POLLINATION. JULY 13. TREE 1. Ovule No. of embryos No. of nuclei No. per i n ovule terminal embryo 1 1 too numerous to count 2 2 Y 3 2 Y 4 2 Y 5 3 Y 6 4 1 Y signifies advanced embryo collapse (7) July 21. one hundred and eight days after self-pollination. Of the four ovules examined, two contained collapsed embryos; these were very d i f f i c u l t to find since the gametophyte cells did not stain and their nuclei had disappeared. Studies were discontinued beyond this date because embryo failure appeared to be complete. The results, which are summarized i n table 12, showed that no barrier to early embryo development existed, as the ninety ovules examined contained two hundred and thir t y - f i v e embryos. The very low y i e l d of viable seed from this - 37 -tree, however, indicated that very few embryos developed to maturity. TABLE 12. THE NUMBER OF EMBRYOS AFTER SELF-POLLINATION. TREE 1. Date of No. of No. of No. of Average no. of collection ovules ovules with embryos embryos per ovule sectioned embryos 9/6 10 8 15 1.9 16/6 8 6 18 3.0 21/6 31 25 67 2.7 28/6 21 21 58 2.8 7/7 26 24 63 2.6 13/7 6 6 14 2.3 102 90 235 2.6 - 38 -PLATE IV Figure 1. Vigorous young embryo sixty-eight days after self-pollination. June 9, 1952. X 345. Figure 2. Young terminal, embryo collapsing seventy-six days after self-pollination. June 16, 1952. X 345. Figure 3. The characteristic dense staining of the embryo cells in the early collapse state. June 16, 1952. X 890. Figure 4. Terminal embryo eighty days after self-pollination, June 21, 1952. X 345. Figure 5. Terminal embryo on the point of collapse. June 21, 1952. X 345. Figure 6. Two collapsed embryosj note the dense staining and cell shrinkage. June 21, 1952. X 345. Figure 7. Terminal embryo, eighty-seven days after self-pollination. June 28, 1952. X 345. Figure 8. A vigorous terminal embryo eighty-four days after cross-pollination, compare with figure 7. June 28, 1952. X 345. P L A T E W - 39 -PLATE V Figure 1. Terminal embryo eighty-seven days after s e l f -pollination and i n the advanced state of collapse. June 28, 1952. X 345. Figure 2. Three collapsed embryos; note the condition of the gametophyte. June 28, 1952. X 92. Figure 3. Two collapsed embryos ninety-six days after s e l f -pollination. July 7, 1952. X 345. Figure 4. The same two embryos under lower magnification; note the food material i n the cells of the gametophyte. X 92. Figure 5. A collapsed terminal embryo; note the erratic staining. July 7, 1952. X 345. Figure 6. A collapsed terminal embryo. July 7, 1952. X 345. Figure 7. A collapsed terminal embryo one hundred days after self-pollination. There i s l i t t l e staining and the gametophyte ce l l s are breaking down. July 13, 1952. X 345. Figure 8. A late embryo; note the cotyledons and plumule primordia. July 13, 1952. X 92. - 40 -III CYTOLOGICAL INVESTIGATION WITH TREE A Comparative Ovule Abortion After Self-Pollination The controlled self-pollination of 1952 had shown that there was a considerable variation in the amount of viable seed produced by two different trees. This variation was investigated further in 1954, as i t was considered desirable to obtain some comparable results from other trees. Six more trees were therefore selected for controlled self-pollination; three of these were near Victoria and the remainder farther north at Lake Cowichan. The female cones were isolated and self- and cross-pollinated in the manner already described. Sample cones were later removed at different dates from each tree and the ovules killed-and-fixed prior to embedding. The effect of self-pollination on these six trees was naturally not known at this time and would not have been known until their cones had been collected and the seed germinated. It was essential, however, that there should be some quick guide as to the effect of self-pollination on the different trees in order that a cytological study could be made on the most suitable tree. i It was found that a very useful guide was obtained by examining the condition of the ovules when the integuments had been removed prior to killing-and-fixing. There are two unproductive regions near the tip and base of the cones of Pseudotsuga and the ovules in this region do not develop past the rudimentary stage. Similar regions on the cones of Pinus resinosa Ait have been described by - 41 -Lyons (1956). Such ovules were discounted as their condition was not related to the effects of pollination. The condition of the ovule 1 i n the productive region, however, received particular attention. A healthy ovule i s firm and white i n colour (see figure 8, plate IX) but towards the end of June and the beginning of July, some of the ovules became limp and discoloured. This condition increased i n intensity and the ovules f i n a l l y collapsed and shrivelled up completely as shown i n figure 8, plate IX. Such ovules have been termed aborted by Lyons (1956) and this definition w i l l be used to distinguish between collapse of the ovules and collapse of the embryos. Ovule abortion can be caused either by lack of pollination and f e r t i l i z a t i o n or by the prior collapse of the embryo; i t was previously found that an ovule w i l l not persist to maturity i f there i s no embryo. The numbers of healthy and aborted ovules were recorded for each tree as soon as the f i r s t signs of abortion were observed. Table 13 shows these results and i t can be seen that ovule abortion i n cones that had intentionally been l e f t unpollinated and Abortion i n cones that had been pollinated, occurred at approximately the same time. The dates of t h i s abortion occurred several weeks after f e r t i l i z a t i o n i n a l l six trees. Ovule abortion after self-pollination was much more noticeable i n trees A, C, and D than i n trees F, H, and J, and these l a t t e r trees were clearly of l i t t l e value for the cytological investigation. It can be seen that self-pollination 1For convenience, ovule here refers to the intra-integumentary tissues and does not include the integuments. resulted i n complete ovule abortion for tree A and there was l i t t l e doubt that much of this must have been caused by the prior collapse of the embryo. Tree A was, therefore, selected for the cytological investigation and the results later showed that this supposition was perfectly correct. - 43 -TABLE 13. COMPARATIVE OVULE ABORTION IN POLLINATED AND UNPOLLBJATED CONES Self- Cross- Cones not Tree Date p o l l i n a t e d p o l l i n a t e d p o l l i n a t e d * of cone cones cones removal Days Healthy Aborted Healthy Aborted Healthy Aborted a f t e r Ovules Ovules Ovules Ovules Ovules Ovules p o l l i n . A 28/6 71 37 5 no collapse no collapse 2/7 75 35 8 46 3 no collapse 7/7 78 3 19 18 16 a l l collapsed 14/7 85 0 37 24 17 0 41 21/7 92 0 51 31 14 0 29 C 7/7 83 0 53 27 18 no 14/7 92 17 22 35 20 samples 26/7 104 2 22 25 12 D 28/6 74 4 15 19 0 no collapse 1/7 77 1 24 22 10 no collapse 7/7 80 3 30 11 22 21 5 14/7 no samples 0 39 F 11/7 68 no collapse no collapse 39 3 14/7 71 39 2 41 5 3 35 17/7 78 46 1 27 5 1 37 27/7 83 41 6 35 3 0 43 -13/8 105 19 28 42 17 0 43 H 14/7 76 74 1 47 8 29 41 17/7 79 50 0 22 29 1 69 22/7 84 62 3 58 0 0 67 27/7 89 37 11 49 3 0 66 13/8 106 7 . 55 42 5 0 66 J 11/7 81 no collapse 42 11 9 42 15/7 85 40 1 28 19 0 42 17/7 87 31 5 38 9 0 61 27/7 89 25 13 14 32 0 46 13/8 114 12 42 43 6 0 45 * These cones had been isolated and intentionally l e f t unpollinated. F e r t i l i z a t i o n and Proembryo Development The results of the investigation with tree 1 had shown that f e r t i l i z a t i o n occurred i n the normal manner a f t e r s e l f - p o l l i n a t i o n - 44 -but that the subsequent collapse of the embryos was the cause of the low y i e l d of viable seed. It seemed probable that f e r t i l i z a t i o n would proceed normally after self-pollination i n tree A, but i t was thought desirable that this assumption should be checked. Sections were> therefore, made from eight ovules which had been removed f i f t y -one days after self-pollination. Syngamy and later stages of development were found i n five of these ovules and sperm nuclei were seen i n the archegonia of the remainder. The f i r s t division of the zygote i n one of these ovules i s ill u s t r a t e d i n figure 2, plate VI. In a l l of the eight cross-pollinated ovules sectioned, syngamy and later stages of development were observed. The development within the archegonia of the s e l f - and cross-pollinated ovules i s shown i n table 14 which indicates that development i n the la t t e r was more advanced. The cross-pollinated sample, however, had been pollinated two days earlier which could account for the more advanced development. The results definitely indicated that there was no physiological barrier to s e l f - f e r t i l i z a t i o n and supported the findings already reported for tree 1. TABLE 14. DEVELOPMENT IN ARCHEGONIA AFTER SELF- AND CROSS-POLLINATION TREE A Type Days No. of No. of Archegonia with these stages of after ovules archeg- Sperm Syngamy 1st & 2nd Pro embryo p o l l i n . p o l l i n . sampled onia. nuclei zygotic develop-present divisions ment Self 51 8 38 10 6 3 1 Cross 53 8 36 - 1 1 20 - 45 -Embryo Development The results of the previous examination showed that the impediment to normal seed development must occur after self-fertilization and during development of the embryo. The subsequent examinations, therefore, replicated those conducted with tree 1, comparisons with samples from controlled cross-pollination could also be made at most dates. (l) June 16. fifty-nine days after self-pollination. Of the fifteen ovules sectioned, fertilization had occurred in thirteen. Vigorous young embryos were observed in ten of these ovulesj one of these is illustrated in figure 3> plate VI; again, several stages of development were found within the same ovules as occurred in the case of tree 1. Syngamy, for example, was present in three ovules which also contained embryos; syngamy, proembryos, and embryos were a l l seen in another ovule. Figure 1, plate VI; shows a sperm nucleus lying within that of the egg, nine days after syngamy was firs t recorded from this tree. TABLE 15. DEVELOPMENT OF EMBRYOS AFTER SELF-POLLINATION. JUNE 16. TREE A Ovule No. of embryos No. of nuclei No. per in ovule terminal embryo 1 3 2 2 3 2 3 1 2 4 2 2 5 4 2 6 4 2 7 2 2 8 4 k 9 4 4 10 3 4 - 46 -(2) June 20, sixty-eight and sixty-five days respectively after s e l f - and cross-pollination.  Of the ten self-pollinated ovules sectioned, three contained embryos, the remaining seven had not been f e r t i l i z e d . The terminal embryos were growing vigorously and there was no sign of any collapse. On the other hand, of the fifteen cross-pollinated ovules sectioned, fourteen contained young embryos, one of which i s illus t r a t e d i n figure 4, plate VI. TABLE 16. DEVELOPMENT OF EMBRYOS AFTER SELF-- AND CROSS-POLLINATION JUNE 20. TREE A Self-pollination Cros s-pollination Ovule No. of No. of nuclei Ovule No. of No. of nuclei No. embryos i n terminal No. embryos i n terminal per ovule embryo per ovule embryo 1 1 7 1 4 2 2 3 7 2 1 2 3 2 11 3 3 3 4 5 4 5 5 4 6 3 4 7 3 4 8 3 4 9 2 4 10 1 4 11 4 4 12 5 4 13 3 5 14 4 6 (3) June 23, sixty-eight days after both s e l f - and cross-pollination.  Of the fourteen self-pollinated ovules sectioned, eleven contained embryos. The f i r s t signs of embryo collapse, however, appeared i n this 1 collection; three of the ovules contained terminal embryos i n the early state of collapse. The remaining eight ovules contained sound terminal - 47 -embryos and mitosis was observed in three of themj one such embryo is illustrated in figure 5> plate VI. Fifteen cross-pollinated ovules were sectioned and thirteen of them contained embryos. In one of the ovules the terminal embryo had collapsed but the remainder were perfectly normal. These embryos, moreover, were already developing more rapidly than those from self-pollination and the number of nuclei observable in one focal plane was generally higher; one such embryo i s illustrated i n figure 6. The contents of the gametophyte cells had previously not shown any unusual concentration of stainable material but from this date onward, the cell contents began to increase in stainability. This occurred without relation to the type of pollination. TABLE 17. DEVELOPMENT OF EMBRYOS AFTER SELF-- AND CROSS-POLLINATION JUNE 23. TREE A Self-pollination Cro s s-pollination Ovule No. of No. of nuclei Ovule No. of No. of nuclei No. embryos in terminal No. embryos in terminal per ovule embryo per ovule embryo 1 3 6 1 3 5 2 3 7 2 3 6 3 3 7 3 3 8 4 4 7 4 3 8 5 3 9 5 5 9 6 4 10 6 2 9 7 4 10 7 5 11 8 4 11 8 3 13 9 1 X 9 4 14 10 1 X 10 4 14 11 2 X 11 5 15 12 4 17 13 3 X X signifies early embryo collapse (4) June 28, seventy-one days after both self- and cross-pollination  Of the fifteen self-pollinated ovules sectioned, ten contained embryos. In six of these ovules the terminal embryos were in the early state of collapse as shown in figure 8, plate VI. Collapse was again characterized by the dense staining and shrinkage previously observed in the collapsing embryos of tree l j the embryo in figure 8, for example, i s very similar in appearance to that in figure 2, plate IV. Another embryo which i s on the point of collapse can be seen in figure 7; staining is particularly dense at the apex. Twenty-one cross-pollinated ovules were sectioned and nineteen of these contained embryos. In six ovules the terminal embryos were in the early state of collapse (see figure 1, plate VII). Their appearance is very similar to that of embryos collapsing after self-pollination. The remaining ovules contained healthy embryos which were considerably more developed than those from self-pollination; in one focal plane, their nuclei numbered between fifteen and forty-six compared with eight and sixteen for those resulting from self-pollination. The presence or absence of mitosis in these terminal embryos was also suggestive; i t was observed only in two embryos resulting from self-pollination compared to eleven resulting from cross-pollination. The cones from which both groups of ovules were taken were removed from adjacent branches so that the position on the tree could hardly have effected development within the ovules. - 49 -TABLE 18. DEVELOPMENT OF EMBRYOS AFTER SELF-- AND CROSS-POLLINATION JUNE 28. TREE A Self-pollination Cross-pollination Ovule No. of No. of nuclei Ovule No. of No. of nuclei No. embryos in terminal No. embryos in terminal per ovule embryo per ovule embryo 1 2 8 1 3 15 2 4 9 2 5 16 3 2 13 3 2 18 4 3 16 4 3 20 5 1 X 5 4 20 6 2 X 6 3 21 7 2 X 7 6 25 8 4 X 8 2 26 9 4 X 9 3 32 10 4 X 10 5 34 11 4 35 12 2 40 13 5 46 14 2 X 15 3 X 16 3 X 17 4 X 18 4 X 19 4 X X signifies early embryo collapse (5) July 2nd, seventy-five days after both self- and cross-pollination.  Of the thirty-four self-pollinated ovules sectioned, thirty-three contained embryos; in five of these the terminal embryos had collapsed. The majority of the terminal embryos in the other ovules were small and only one showed evidence of cell division. One such embryo is illustrated in figure 2, plate VIIj i t may be compared with figure 4 which shows an embryo resulting from cross-pollination and collected on the same date. A larger embryo resulting from self-pollination is illustrated in figure 3 but i t is s t i l l not - 5.0 -comparable i n vigour with that i n figure 4« Of the thirty-one cross-pollinated ovules sectioned, twenty-eight contained embryos. There was no collapse i n any of the terminal embryos which were in striking contrast to those resulting from self-pollination. The number of nuclei v i s i b l e i n one focal plane ranged between twenty and one hundred and thirty-eight, compared with seven and sixty-four i n the case of the embryos resulting from selfing. - 51 -TABLE 19. DEVELOPMENT OF EMBRYOS AFTER SELF- AND CROSS-POLLINATION JULY 2. TREE A Self-pollination Ovule No. of No. of nuclei No. embryos in terminal per ovule embryo 1 5 7 2 4 8 3 4 9 4 5 10 5 3 13 6 5 15 7 3 15 8 1 15 9 4 16 10 5 16 11 5 16 12 4 18 13 5 18 14 3 20 15 2 20 16 2 21 17 5 22 18 2 23 19 6 24 20 5 26 21 3 27 22 3 29 23 3 30 24 4 31 25 4 31 26 5 37 27 3 38 28 4 64 29 1 X 30 3 X 31 3 X 32 4 X 33 4 X Cro s s-pollination Ovule No. of No. of nuclei No. embryos in terminal per ovule embryo 1 5 20 2 4 32 3 6 39 4 4 u 5 4 44 6 5 49 7 6 50 8 5 52 9 5 54 10 4 56 11 3 56 12 2 62 13 4 65 14 2 69 15 3 69 16 4 71 17 4 74 18 3 78 19 4 86 20 6 87 21 3 88 22 5 88 23 4 90 24 3 97 25 5 98 26 5 103 27 4 126 28 4 138 X signifies early embryo collapse (6) July 8, eighty and seventy-five days respectively after self-and cross-pollination.  Only three self-pollinated ovules were sectioned at this date since they were the only sound ones observed when the seed coats were removed; none of the other thirty-eight aborted ovules was embedded. The three ovules a l l contained embryos in the advanced state of collapse and one of these is illustrated in figure 5, plate VII. There was no sign of collapse in the gametophyte cells, but the dark-staining cell contents were disappearing. Sixteen cross-pollinated ovules were sectioned. Fifteen of these contained terminal embryos with a number of visible nuclei ranging from thirty to one hundred and twenty-eight; numerous mitoses were also observed in a l l of these embryos. There was intense competition among the embryos in most of these ovules; i t was not unusual to find three or four developing in the same ovule, a l l with approximately the same number of nuclei. TABLE 20. DEVELOPMENT OF EMBRYOS AFTER SELF- AND CROSS-POLLINATION . JULY 8.. TREE A Self-pollination Cross-pollination Ovule No. of No. of nuclei Ovule No. of No. of nuclei No. embryos per ovule in terminal embryo No. embryos per ovule in terminal embryo 1 2 Y 1 3 30 2 2 Y 2 1 33 3 3 Y 3 2 39 4 6 48 5 4 52 6 2 54 7 6 57 8 2 67 9 4 80 10 2 93 11 1 94 12 3 95 13 1 101 14 5 112 15 3 128 Y signifies advanced embryo collapse - 53 -(7) July 14* Eighty-two days after cross-pollination. No self-pollinated ovules were sectioned at this date because they had a l l aborted. A l l of the five cross-pollinated ovules sectioned contained rapidly developing embryos. One of these is illustrated in figure 6, plate VII, and i t can be seen that there is already a considerable amount of embryonic organisation. (8) July 21, one hundred and five and eighty-nine days respectively after self- and cross-pollination. Two collapsed self-pollinated ovules were sectioned and both contained embryos. One of these terminal embryos i s illustrated in figure 7, plate VIIj i t can be seen that there is very l i t t l e staining in the garnetophyte cells which have begun to break down. An ovule in which fertilization had occurred, however, could s t i l l be distinguished from an unfertilized one at this date since the latter was in a much more advanced state of collapse. Three cross-pollinated ovules were sectioned; a l l of these contained embryos in advanced stages of development, one of which is illustrated in figure 8. The cytological study was not continued past this date since a l l of the self-pollinated ovules had aborted in the remaining samples collected on July 27th and August 8th. The results which are summarised in table 21 again showed that no barrier to early embryo development existed as the seventy-two ovules examined contained two hundred and thirty-one embryos. A test of Chi square, moreover, showed that there was no significant difference between the number of embryos produced after self- or cross-pollination. TABLE 21. NUMBER OF EMBRYOS AFTER SELF- AND CROSS-POLLINATION TREE A Self-pollination Cross-pollination Date of No. of No. of Average No. of No. of Average collection ovules embryos no. of ovules embryos no. of with embryos with embryos embryos per ovule embryos per ovule 16/6 10 30 3.0 20/6 3 6 2.0 14 46 3.3 23/6 11 32 2.9 13 47 3.6 28/6 10 28 2.8 19 67 3.5 2/7 33 122 3.7 28 116 4.1 8/7 3 7 2.3 15 45 3.0 14/7 - - - 5 20 4.0 21/7 2 6 3.0 3 7 2.3 72 231 3.2 97 348 3.6 - 55 -PLATE VI Figure 1. The sperm lying in the egg nucleus fifty-nine days after self-pollination. June 16, 1954. X 890. Figure 2. First division of the zygote f i f t y days after self-pollination. June 7, 1954. X 345. Figure 3. Vigorous young embryo fifty-nine days after self-pollination. June 16, 1954. X 345. Figure 4. Young embryos sixty-five days after cross-pollination. June 20, 1954. X 345. Figure 5. A terminal embryo sixty-eight days after self-pollination; note the mitoses in the cells. June 23, 1954. X 345. Figure 6. A terminal embryo sixty-eight days after cross-pollination, compare its development with that in figure 5. June 23, 1954. X 345. Figure 7. A temlnal embryo about to collapse, seventy-one days after self-pollination. June 28, 1954. X 345. Figure 8. A collapsed terminal embryo after self-pollination; note the characteristic dense staining. June 28, 1954. X 345. P L A T E VI - 56 -PLATE VII Figure 1. A collapsed terminal embryo seventy-one days after cross-pollination; note the similarity with figure 8, plate VI. June 28, 1954. X 345. Figure 2. A terminal embryo seventy-five days after self-pollination. July 2, 1954. X 92. Figure 3. A terminal embryo after self-pollination, compare development with figure 4. July 2, 1954. X 345. Figure 4. A terminal embryo seventy-five days after cross-pollination. July 2, 1954. X 345. Figure 5. A collapsed terminal embryo eighty days after self-pollination. July 8, 1954. X 92. Figure 6. A terminal embryo eighty-two days after cross-pollination; note the depleted cell layers of the gametophyte. July 14, 1954. X 92. Figure 7. A collapsed terminal embryo, one hundred and five days after self-pollination. July 21, 1954. X 345. Figure 8. A terminal embryo eighty-nine days after cross-pollination. July 21, 1954. X 92. P L A T E VII - 57 -Comparison of Embryo Collapse in Trees 1 and A It was quite evident that the low viability of seed after self-pollination in both trees 1 and A was due to some inhibition to embryo development. Pollen germination and fertilization a l l compared favourably with that following cross-pollination, and the early development of the embryos was also quite normal. These embryos, however, were soon completely outgrown by those resulting from cross-pollination. This is vividly illustrated in plate VIII, which shows the comparative embryo growth following both kinds of pollination. The points on the graphs are based on the mean number of nuclei visible in one focal plane in the terminal embryos; the very rapid growth of those resulting from cross-pollination is in striking contrast to that of the embryos resulting from self-pollination. A few sections were made from the self -pollinated ovules of tree D in which ovule abortion had also been high. The normal stages of fertilization were observed and subsequent embryo collapse occurred at approximately the same time as in tree A. It seems reasonable to conclude, therefore, that embryo collapse is the cause of the low seed yields after self-pollination in Pseudotsuga. There were some interesting differences between the course of embryo collapse in trees 1 and A. It should be remembered, however, that the investigation on these two trees was carried out in different years and that trees 1 and A were growing in Vancouver and Victoria respectively. Syngamy in both trees - 58 -occurred between fifty-one and fif t y - f o u r days after self-pollination, but the subsequent development of these embryos was much slower i n tree 1 and they never reached the size of those i n tree A. In both cases, very few viable seeds were produced, but there would appear to have been a higher degree of embryo collapse i n tree 1, as only two viable seeds were produced from thirty-nine cones i n contrast to sixteen from eighty-two cones i n tree A. Development of the embryos resulting from crossing also was slower i n the case of tree 1 than of tree A; the largest embryo from tree 1, eighty-seven days after pollination, for example, showed only eighty-two nuclei i n one focal plane whereas the nuclei i n an embryo of similar age from tree A were too numerous to count. The characteristic dense staining and c e l l shrinkage of the embryo i n the early collapse state, however, were common to both trees and the physical appearance of the ce l l s of the garnetophyte was identical. No collapse was observed i n these cells prior to the collapse of the embryo which invariably occurred i n the trees investigated when the ce l l s of the gametophyte were f u l l of stainable material. i PLATE VIII Comparative embryo growth following self-and cross-pollination of trees 1 and A. The number of nuclei refers to the maximum number visible in one focal plane, and not to the total number contained by the embryo. PLATE YHI COMPARATIVE EMBRYO GROWTH FOLLOWING S E L F AND CROSS POLLINATION. 80 701 60-D >-tt m Z W 5 50-- i u o At 82days,embryos too -f large to count. .SELF POLLINATED PROGENY TREE A .CROSS . S E L F „ „ 1 NUMBERS REFER TO HEALTHY APICAL EMBRYOS. DAYS AFTER POLLINATION. - 60 -IV THE POSSIBLE CAUSES OF EMBRYO COLLAPSE AFTER SELF-POLLINATION Self-Incompatibility An important question is posed; i s embryo collapse after self-pollination part of a genetic system of incompatibility or is i t the .result of some different cause? The conclusions of several well known investigators of the incompatibility systems in the Angiosperms definitely suggest that embryo collapse i s not part of such a genetic system since fertilization i s able to take place. East (1929), one of the early investigators of incompatibility, stated: Various species of hermaphroditic animals and plants exist in which a union between the male and female gametes produced by a given individual is difficult or even impossible, although each type of gamete is functional in other unions. This phenomenon i s usually called self-sterility. Mather (1943) pointed out that: "Incompatibility may be defined as the failure, following mating or pollination, of a male and a female gamete to achieve fertilization,...." According to Lewis (1944): Sexual reproduction is essentially the formation of a new individual resulting from the fusion of two gametic nuclei. Any hindrance to this fusion, within the regular mating group, except when i t is due to a defect in the nuclei themselves, is said to be due to incompatibility. Bateman (1952) stated: "The primary function of self-incompatibility is the avoidance of self-fertilization". It wi l l be recalled that Sears (1937), classified the incompatibility systems of the Angiosperms into three distinct stages (page 6 of this dissertation). In one of these which concerned Gasteria, the pollen tubes grew - 61 -normally, reached and fertilized the ovules, but no seed developed. Mather (1944), however, did not consider that this is a true self-incompatibility system and pointed out: It is said that in some plants, like Gasteria ... incompatibility manifests itself as a breakdown of development after fertilization. Such a situation can, however, scarcely be described as due to incompatibility in the present sense, for i t omits the essential selective advantage of conservation of female gametes, and hence must have arisen in some other way. In fact, i t i s difficult to see how such a system could arise by direct selection. Lewis (1954) later wrote: An extreme case was described by Sears in Gasteria in which the incompatible pollen tubes grow at the same rate as the compatible tubes, reach the ovary, and fertilize the endosperm, but the egg nucleus is not fertilized. Unfortunately nothing is known about the incompatibility system in this species. If these definitions of incompatibility are accepted there i s clearly l i t t l e justification for applying this term to embryo collapse after self-pollination in Pseudotsuga. since fertilization i s in no way inhibited. Embryo collapse, therefore, must be ascribed to some other cause. Bateman (1952) pointed out, however, that there is no reason why a true incompatibility system should not be present in the Gymnosperms. It is , therefore, interesting to note that Forshell (1953) recently claimed that such a system does exist in Pinus sylvestris and that fusion of the sex cells is partly inhibited. She states that her theory of incompatibility i s not contrary to what is known about the genes of self-sterility in the Angiosperms in which pollen-tube growth, after self-- 62 -pollination, is restrained by these genes. Her study i s the only other known investigation on self-incompatibility in the Gymnosperms and her conclusions have been carefully considered as i t is unlikely that the behavior of two species of the Pinaceae should be radically different. Forshell's conclusions, apparently, are not based upon cytological evidence but on an earlier study by Branscheidt (1939), and on X-ray photographs of seeds resulting from both self- and cross-pollination of Pinus sylvestris. In his study, Branscheidt (1939) succeeded in growing pollen of Taxus baccata on an a r t i f i c i a l medium and followed the complete development from germination to the maturity of the male sex cells. Under these a r t i f i c i a l conditions, the pollen tubes did not go through the customary rest period which is usually six weeks in duration. Branscheidt referred to a previous study on Taxus  baccata by Jagerl who found that the several archegonia in the same ovule were often at different stages of development. A young developing archegonium, for example, could be found near one from which the embryo has already developed. Branscheidt concluded that the immature archegonium or garnetophyte must produce some inhibiting substance which causes the rest period of the pollen tube and that this substance must be eliminated when the archegonium i s mature. Forshell added that, in Pinus sylvestris. this inhibiting effect of the archegonia does not prevent the pollen tubes from growing after self-pollination since a large number of seeds are found which are either empty or contain only gametophyte tissue. Such seeds can "'"jSger L. Beitrage zur Kenntnis der Endospermbildung und zur Embryologie von Taxus baccata. Flora 86. 1899. - 63 -develop only i f the integument and gametophyte tissue are stimulated by a pollen tube. The archegonia, however, do prevent fusion of the male and female nuclei in those ovules where the inhibiting effect does not cease following the rest period of the pollen tube. Forshell pointed out that the reason why self-sterility is not complete and why a large or small number of embryos are produced after self-fertilization in Pinus sylvestris, is explained by Branscheidt. He concluded that as the several archegonia in an ovule were at different stages of development, the inhibiting substance would be eliminated at different timesfor each archegonium. Furthermore, several pollen tubes at different stages would surround archegonia which were also in different stages of development. Forshell, therefore, stated that the formation of viable seed after self-pollination i s due to the archegonia losing their inhibiting effect, either partly or completely at a late stage so that they can then be fertilized. The "self-sterility" of Pinus  sylvestris i s , therefore, not complete according to this concept. There appears to be l i t t l e evidence to support this hypothesis. Branscheidt (1939) based a l l of his conclusions on the previous study by Jager and on the finding that pollen, growing on an a r t i f i c i a l medium, did not go through the rest period which was usual under natural conditions. Neither he nor Forshell presented any experimental evidence that an inhibiting substance is produced by the archegonia. Forshell, moreover, presented no cytological evidence to support her statements that fusion of the male and female nuclei i s prevented in certain ovules. She, therefore, had no proof that there - 64 -was less fertilization after self-pollination than after cross-pollination. Anhaeusser (1953), recently conducted a much more intensive study on pollen germination and development on a r t i f i c i a l media in the Gymnosperms. He believed that Branscheidt did not select a typical representative in regard to the rest period of the pollen tubes and delay in fertilization since the interval between pollination and fertilization i s only six weeks. A much clearer picture of the problem would have been given by a genus such as Pinus. Regarding the X-ray photographs of the seed from self- and cross-pollination in Pinus sylvestris, there is certainly evidence that there were fewer embryos in the self-pollinated ovules as compared with those cross-pollinated. These photographs, however, appear to have been taken when the embryos were at a late stage of development as the cotyledons in some cases were already developed. This investigation with Pseudotsuga. however, has shown that the embryos are just as numerous after self-fertilization as after cross-fertilization but that they collapse at an early age before organisation of the surviving embryos occurs. The collapsed embryos gradually disintegrate and are soon very difficult to find. Photographs by X-ray taken shortly after fertilization in the case of Pinus could have presented a totally different picture since at that time, there may have been just as many embryos in the self-pollinated ovules as there were in the cross-pollinated. A recent study on Pinus monticola by Bingham and Squillace (1955), also suggests that i t is unlikely that a self-incompatibility system i s operative in Pinus. They self-pollinated several trees for successive years and in one tree, the yield of viable seed varied considerably from year to year. Such variation would be unlikely i f any known incompatibility system were present. The fact that certain Gymnosperms can produce variable amounts of viable seed after self-pollination must make the terms "self-sterile" and "self-incompatible" superfluous. It i s , therefore, suggested that neither of these terms should be used in their present sense for the Gymnosperms until well substantiated evidence has been presented. Inbreeding Effect The most likely explanation of embryo collapse is that i t is caused by inbreeding which results in the increased homozygosity of recessive deleterious genes. Dobzhansky (1952, page 220), demonstrated that in time, any normal breeding population will build up a mass of deleterious recessive genes by mutation. Many of these genes, moreover, are lethal or semi-lethal but they are sheltered from selection pressure by the normal dominant alleles in a heterozygous organism. The situation, however, i s completely changed by inbreeding because many of these recessive genes will then be homozygous and consequently exposed. It is logical to assume that there will be both lethal and semi-lethal genes in Pseudotsuga. and that they are capable of destroying the embryo soon after fertilization when in a homozygous state. The number of such genes will undoubtedly vary greatly from tree to tree and the effects of self-pollination in terms of embryo collapse will vary accordingly. The degree of collapse may again be dependent upon certain - 66 -recombinations which explains why repeated self-pollinations of a tree do not always give the same result. This has already been demonstrated in Pinus monticola by Bingham and Squillace (1955). The exact conditions which bring about embryo collapse, however, will probably remain obscure. Wardlaw (l955> page 10), pointed out that growth from the zygote or spore to the adult state is characterized by an orderly sequence of development in which factors in the hereditary constitution are involved. These factors or genes, moreover, determine and control metabolic processes, and are, therefore, involved in a l l growth and cellular activity. Wardlaw stated that the time of action of particular genes is of special importance in embryogenesis as only by their activation can orderly development follow. He added, however; "But how the genes act, individually and collectively, is a problem that i s as difficult as i t i s important". It is evident, therefore, that any drastic change such as inbreeding a normally outbreeding organism could very easily upset the orderly sequence of development in the young embryo. Brink (1952) pointed out that the early embryo i s dependent upon the endosperm for certain metabolites which, in i t i a l l y , the embryo is quite incapable of synthesizing. It later becomes progressively less dependent by acquiring for itself the synthetic capabilities previously limited to the nurse tissue. The cause of embryo collapse in Pseudotsuga would appear to be from some failure in this vital relationship between the early embryo and the gametophyte - 67 -tissue. These embryos regularly collapsed at an early stage of development and, consequently, before they were less dependent upon the gametophyte. Every case of collapse, moreover, occurred in ovules whose gametophyte cells were f u l l of dark-staining material apparently of l i t t l e or no value to the embryos. Embryos from cross-pollination, however, were able to develop in the normal manner under what appeared to be identical conditions in the gametophyte. Wardlaw (1955, page 12), pointed out that in the case of plants, very l i t t l e is known of the chemical composition of the zygote, the young embryo, the surrounding tissues and the metabolic processes involved in growth and development. At the present time, therefore, i t can be concluded only that embryo collapse in Pseudotsuga is more likely to be the result of inbreeding than of some unknown incompatibility system. There is other evidence from the Gymnosperms that embryo development is dependent upon the inheritance of the parents. Buchholz (1945) effectively demonstrated that hybrid vigour, the antithesis to inbreeding, can be observed in the early development of embryos from a hybrid pine as compared with the development of those from the parent trees. He decided, however, that this hybrid vigour was nothing that might be detected by a morphological study of the embryos themselves since its nature was definitely a physiological vigour of growth. This same physiological vigour of growth was definitely exhibited in the early development of the embryos from cross-pollination as compared with those from self-pollination during the cytological investigation of tree A. Embryo collapse and its possible connection with inbreeding - 68 -has been reported in some Angiosperms. Brink and Cooper (1940), for example, found that in Medieago sativa, the endosperm grew faster after cross-pollination than after self-pollination. An embryo from cross-pollination was accompanied, at a given early stage of development, by an endosperm more advanced than that associated with an embryo from self-pollination. They suggested that the reduced rate of early endosperm growth was an inbreeding phenomenon associated with the increasing homozygyosity of numerous recessive genes, each having a small adverse effect on development. Cooper and Brink (1940) later concluded that the collapse of the fertible ovules in Medieago sativa. which occurred so freely after self-pollination, could be due to some of the genes conditioning pollen-tube behaviour, these being also responsible for ovule collapse. On the other hand, i t could be due to recessive genes at other loci which, when brought together in a homozygous state by inbreeding, adversely affect the early development of the zygote. Cooper and Brink (1940) pointed out that although the evidence was limited, i t indicated that the collapse of these fertile ovules was an inbreeding effect rather than a direct manifestation of self-incompatibility. Further cytological investigations on the effects of self-pollination in other Gymnosperms are obviously desirable in order to augment present limited information. - 69 -V PARTHENOCARPY AND AGAMOSPERMY The development of non-pollinated cones with fully formed but empty seeds has been termed parthenocarpy and appears to have been first investigated in the Gymnosperms by Kurdiani (see Hummel, 1930). He concluded that i t was a damaging influence in the forest as the development of cones and empty seed took nutrients away from the tree which would otherwise have been used for growth or for the formation of flower primordia the following year. Kurdiani found that parthenocarpy was of common occurrence in Taxus, Thuja, Larix. Abies, Picea, and Juniperus. It was found that the non-pollinated cones of Pinus sylvestris, however, usually dropped off although in later studies, Pinus eldarica produced cones of normal size but the seed coats were absent, Syrach Larson (1937, page 73), later found that parthenocarpy was also of normal occurrence in Chamaecyparis, Cryptomeria, Tsuga, and Pseudotsuga and i t i s obviously widespread in the Gymnosperms, a fact, however, which i s s t i l l not always recognized. Anderson (1947), for example, making controlled crosses on Picea, noted: The most peculiar finding of a l l i s that the isolated control flowers grew out into perfectly normal cones ... Formerly i t was generally expected that a certain number of egg-cells must be fertilized in the female inflorescence to ensure cone development. Again, Isaac (1949, page 9), discussing flowering and seed formation in Pseudotsuga stated: " . . . i f not successfully pollinated or fertilized, the conelet like the male flower, withers and disappears". The possibility that parthenocarpic cones of some of the Gymnosperms might also produce viable seed is of considerably more - 70 -interest as, although i t s occurrence has not been definitely established, there are indications that i t may occasionally occur. This phenomenon, previously known as parthenogenesis, i s of fairly common occurrence in many of the Angiosperms. It is now better known as apomixis, which Stebbins (1950, page 380) defined as covering a l l types of asexual reproduction which tend to replace or to act as substitutes for the sexual method. Gustafsson (1946, page 7) further subdivided apomixis into two main groups, agamospermy where reproduction is from seeds and vegetative ' reproduction where i t i s from vegetative formations. The various ways in which agamospermy can arise were described in detail but no reference was made to the Gymnosperms. There have, however, been several reports that viable seeds have occasionally been found in parthenocarpic cones. Dengler (1932), for example, isolated cones on Pinus sylvestris and obtained one parthenocarpic cone that contained two seeds which germinated before extraction. Allen (1942) isolated cones of Pseudotsuga  menziesii and germinated the three viable seeds obtained from two of the trees. The cones, moreover, had been isolated a week before they had emerged from the bud scales and before pollination had begun. Orr-Ewing (1954) obtained twenty-two f i l l e d seeds containing embryos from isolated cones on another tree of the same species. Twelve of these seeds germinated and seven of the seedlings were alive in 1955. One isolated cone was also removed a few weeks before fertilization would normally have occurred and forty-eight of the ovules were sectioned and stained, but there was no trace of any - 71 -pollen* It is improbable that these seedlings could have resulted from normal pollination since the female buds were a l l isolated before the cones had emerged* It is evident, however, that direct cytological evidence must be obtained before agamospermy can be accepted as an established fact; evidence to date is s t i l l by no means conclusive but, on the other hand, i t can not be lightly dismissed. The earliest report of agamospermy, (or parthenogenesis as i t was then called), was by Saxton (1909), in an investigation of Pinus pinaster. He found some ovules with proembryos in a l l stages of development but either there was no trace of pollen tubes in the nucellar cap or the tubes extended through only a part of i t . Saxton claimed that, in one section, the egg nucleus had begun to divide before the pollen tube had reached the archegonium and no sign existed of either the second sperm or the tube nuclei. The number of chromosomes in the egg nucleus could not be counted but Saxton thought that they were less than the diploid number. Coulter and Chamberlain (1917, page 300) supported these findings, and wrote: It has long been expected that parthenogenesis would be found to occur among the Pinaceae and recently Saxton has reported that Pinus pinaster ..* i s parthenogenetic ... The series was close enough, and the comparison with stages in the normal embryo complete enough to make the conclusion reasonably safe. It must be admitted, however, that this study is open to considerable criticism since i t was based on sections from ovules which had not been isolated from pollination. Pollen tubes and their nuclei are not always easy to locate and without the necessary control, normal fertilization could easily have occurred and been overlooked. - 72 -There i s , however, stronger evidence that the ventral canal cell nucleus may occasionally fertilize the egg nucleus which would result in a diploid embryo. Chamberlain (1935, pp. 336 and 340), for example, wrote: It is a curious fact that the fi r s t two cases of fertilization described for conifers were abnormal. Strasburger, as early as 1878, figured two gamete nuclei of about equal size in Picea vulgaris. Nearly 20 years later, Coulter described.two fusing gametes of about equal size in Pinus laricio. There can hardly be any doubt now that the "sperm", in both these cases, was an enlarged ventral canal nucleus, since normal fertilization has been observed so often in both these species, and the sperm nucleus has been proved to be very small in comparison with that of the egg. Chamberlain later added: "My own work on Pinus made this interpretation practically a certainty". Hutchinson (1915), reported that in Abies balsamea the ventral canal cell nucleus occasionally broke through the cell wall and was seen moving towards the egg nucleus. Johansen (1950, page 18), referring to an early report in Gingko, stated: "Fusion of the ventral canal nucleus with the nucleus of the archegonium has been described, but whether this can result in the development of a normal embryo i s unknown although not beyond the realm of possibility". The most conclusive evidence has been presented by Sedgewick (1924), in his investigation of Erieephalartos, a genus in the Cycadaceae. He stated: Fertilization was not observed, and apparently fertilization by a sperm is not necessary in Encephelartos for the successful development of embryos. In greenhouse material and material collected in the field, i t is found that the ventral canal nucleus instead of degenerating, has enlarged and is approaching the egg. He added, however, that the two nuclei were not found in contact but were very close together. Sedgewick later stated that: -73 -An additional reason ... is based on Chamberlain's field observations. In collecting Encephalartos ... , he procured only one staminate cone in the neighbourhood of plants bearing ovulate cones. A further reason i s found in careful study of a l l the preparations. In not a single free nuclear proembryo has a sperm sheaf or ciliated band been found. He then added: The cytoplasmic sheaf of the sperm, and especially i t s ciliated spiral band, are so conspicuous that failure to find them in complete serial sections is practically a proof that they are not present ... ,therefore, i t i s safe to conclude that E. villosus and E. friderici  guilielmi can produce normal embryos in the absence of pollination and fertilization by a sperm. Chamberlain (1935, pp« 122) added: "The fact that embryos have been found in conservatory material where no male cones were present indicates that such a fertilization might occur". Chamberlain also stated that the reported hybrids between Ceratozamia longifolia and Ceratozamia mexicana, also support this conclusion as three-year-old pollen was used and cycad pollen does not retain i t s vitality for more than a month. He added: "It i s possible that the mere irritation of dead pollen in the pollen chamber might stimulate development; but i t is certain that the three year old pollen was dead". Another possible manner in which agamospermy could occur was described by Johansen (1950, page 79), for Ephedra campylopoda: Both microgametes enter an archegonium; one fuses with the ventral canal nucleus and the other with the archegonial nucleus ... moreover, nuclei from the cells surrounding the archegonium may become freed to migrate into the latter and the union of the archegonial nucleus with such a free nucleus was actually observed in an unfertilized archegonium. Whether such an occurrence can give rise to an embryo i s not known. These varied reports suggest that agamospermy could occur and i t was hoped that a more detailed cytological study could be made on tree 2 from which the viable seeds from isolated and unpollinated cones were first obtained. It produced no cones, however, in either 1953 or 1954, and, therefore, some cones were isolated and left unpollinated on five of the trees ar t i f i c i a l l y pollinated in 1954. Sample cones were removed at regular intervals from the time when fertilization would normally be expected to occur. The ovules were then embedded and some were later sectioned and stained. In the f a l l , the remaining cones were collected, and the seed extracted and cleaned. The purpose of cleaning the seed was to remove as much of the empty seed as possible and i t can be seen from table 22 that very l i t t l e seed remained. The results of the germination test, however, showed that this seed was a l l empty. TABLE 22. GERMINATION TEST OF SEED FROM UNPOLLINATED CONES Tree No. of -No. of No. of Germination Cutting test unpollinated seeds cleaned of cones seeds ungermin. seed A 38 1419 2 Nil Empty D 35 966 7 Nil Empty F 53 2368 9 Nil Empty H 30 1821 0 Nil Empty J 56 2719 5 Nil Empty The results of the cytological study proved more interesting. The ovules aborted in the five trees three to four weeks after fertilization would normally have occurred. There were two exceptions, however, two healthy ovules were collected from trees A and J at the end of July and the beginning of September, respectively. Both ovules were later sectioned and each contained diploid embryos which are illustrated in figures 5 and 6, plate IX. It is realized that the possibility of pollen contamination can not be entirely discounted, but i t i s considered to be very remote as a detailed record was kept of the date of isolation of these cones on both trees where these embryos were found. The cone from which the embryo originated on tree A was isolated on March 22nd, and pollen dissemination from this tree was not observed until May 1st; the earliest dissemination in the vicinity began on April 10th. The other cone on tree J was isolated on April 1st; the earliest pollen dissemination from this tree did not take place until May 5th which was earlier than that from the other two trees, F and H. In both cases, isolation took place at least two weeks before the young cones had emerged from the bud scales. The viscose casing bags were securely tied to the branchlets and were not removed until the cones scales had closed completely. The greatest care was taken to remove any male buds present on the branchlets prior to isolation and i t is certain that a l l were picked off. Unfortunately, none of the sections from the unpollinated samples gave any conclusive evidence as to how these two embryos originated. However, one particularly interesting section was obtained from an unfertilized ovule which had been removed at the time when fertilization was taking place in pollinated cones. This section, illustrated in figure 7> plate IX, was stained with Feulgen and showed what appears to be an enlarged ventral canal cell nucleus in close proximity to that of the egg, with chromatic figures - 76 -visible in both nuclei. The series of sections of this ovule were carefully checked and there was no other trace of the ventral canal cell nucleus. It is presumed, therefore, that i t may have broken through the cell wall although no sign of the remnants of this cell wall could be found. It is also possible that the egg nucleus had divided and that these two nuclei are the result of this division. Both nuclei, are very similar in appearance to those shown by Coulter and Chamberlain (1917, figure 298, page 267) for Pinus laricio and interpreted by them as the ventral canal and egg nuclei. The cytological evidence from this investigation with Pseudotsuga i s by no means conclusive but i t indicates a possible manner in which -agamospermy could arise and result in a diploid embryo. One interesting feature of the unfertilized ovules in Pseudotsuga in this study was the comparatively long time that elapsed before they finally aborted. Figure 1, plate IX, for example, illustrates an ovule with archegonia and egg nuclei, the jacket cells s t i l l stain sharply and there i s considerable mitosis both in these and in the cells of the gametophyte. This ovule, however, was removed on June 20th, at least two weeks after fertilization was first noted in the tree. Figure 2, plate IX, illustrates an ovule removed on June 23rd, the archegonium has finally broken down and its former position i s marked by a dark staining and shrunken mass. The nuclei of the gametophyte, however, s t i l l stain sharply and there are numerous mitoses. Figure 3;, plate IX, illustrates another ovule removed on June 28thj i t is aborting and the cell walls are breaking down while the nuclei stain with abnormal intensity. Ovule abortion in unpollinated cones generally occurs about four weeks after the date of normal fertilization, the nuclei at first stain very intensely and finally do not stain at a l l . Some ovules, however, do not abort until much laterj figure 4, plate IX, for example, shows an intact ovule removed on July 14th, nearly six weeks after the normal date of fertilization. Figure 3 and 4 show that there is a definite corrosion cavity in these unpollinated ovules which forms even though fertilization has not occurred. Buchholz (1918) reported that he examined many ovules of Pinus in which the gametophytes had well developed corrosion cavities without any traces of embryos, indicating that the archegonia may excrete the digestive enzyme even though the eggs have not been fertilized. Schopf (1943) found that the gametophytic tissue below the archegonium in Larix broke down at the regular time whether an embryo was present or not. He noted that "In the "automatic" corrosion cavities lacking embryos, the break-down of tissue left a residue very similar to that formed in the old archegonia of fertile gametophytes". This residue i s clearly visible in figure 2 and 4, plate IX, and was commonly but not invariably found in unfertilized ovules. It was never present, however, in those that had been fertilized and was one means of distinguishing between them. The more significant but later difference between the two was the complete absence of stainable material in the cells of the gametophyte in unfertilized ovules. Stainable material began to accumulate in the cells of fertilized - 78 -ovules a few weeks after syngamy and the cells soon became f i l l e d with dark material, as illustrated in figure 4, plate V. The cells of the unfertilized ovule in figure 4, plate IX, however, are almost empty. Allen (1943), in his investigation of Pseudotsuga. found that prior to elongation of the primary suspensor cells, cells in the region of the main axis of the ovule began to accumulate granules of reserve food material. This starch zone, moreover, became more and more extensive and was formed whether or not the ovule was fertilized. The zone was also observed in this investigation but i t did not persist. There seems l i t t l e doubt that fertilization must in some way stimulate the gametophyte to provide an adequate store of food material. It i s possible that the embryo itself i s the centre of this stimulus as the cells of the gametophyte are only f u l l of dark staining material i f embryos are present. If they are not present, the whole ovule finally aborts. This explanation would certainly seem applicable to Pseudotsuga. It, however, would not appear to be applicable to Pinus as Forshell (1953) stated that neither the integument nor the female gametophyte in Pinus sylvestris can develop without the stimulus of the pollen tubes. Forshell referred to Goebel: (1932, page 1799) who pointed out that the conveyance of the male gametes to the egg cell i s by no means the only function of the pollen tubes. He added that in the Coniferae. the ovules are in most, cases pollinated before the megaspores have developed and that the germinating pollen grains stimulate their further development. Goebel,,- however, presented no experimental evidence to support this assumption. It i s interesting to note that Forshell ( 1 9 5 3 ) isolated some young cones of Pinus sylvestris and that they usually did not develop, A few of these young cones did develop to maturity and either produced fully formed but empty seed or seed containing embryos, Forshell, however, considered that this was the result of faulty isolation of the cones, in other words some uncontrolled pollination had taken place. It does indeed seem strange why the behaviour of these two genera in the Panaceae should be so different and further studies with the genus Pinus would seem desirable. There i s ample evidence, however, to show that in Pseudotsuga the female gametophyte can develop up to the time of fertilization, irrespective of whether pollination has occurred or not. There is also no question that the integument can develop without the stimulus of pollination. - 80 -PLATE IX Figure 1. Two archegonia with unfertilized egg nuclei; note the sharply staining jacket cells. Tree A. June 20, 1954. X 92. Figure 2. The remains of an archegonium in an unfertilized ovule; note the empty cells of the gametophyte and the cell break-down in the corrosion cavity. Tree D. June 23, 1954. X 345. Figure 3. An unfertilized ovule in the first stages of abortion. June 28, 1954. Tree A. X 92. Figure 4. An unfertilized ovule, note the empty cells of the gametophyte. Tree A. July 14, 1954. X 345. Figure 5. An embryo found in an ovule from an unpollinated cone of tree A. July 27, 1954. X 92. Figure 6. An embryo found in an ovule from an unpollinated cone of tree J. September 1, 1954. X 92. Figure 7. A nucleus thought to be that of the ventral canal cell approaching the egg nucleus. Tree 1. June 1, 1954. X92. Figure 8. A healthy ovule compared with others which have aborted. P L A T E IX PART 2. I. RESULTS OF THE 1953 PROGENY TEST Comparative Height and Diameter Development These results are related to the comparative development of the seedlings resulting from the self, controlled cross and wind-pollination of trees 1 and 2. The earlier information pertinent to the cone and seed numbers, germination and the development of the seedlings to the f a l l of 1953 has already been published (Orr-Ewing, 1954). In the spring of 1954 the seedlings were removed from the pots and planted out i n a random design i n seed beds at the B. C. Forest Service nursery at Duncan and kept shaded and watered during the summer months. The heights and diameters were measured i n the late f a l l and the results were similar to those of the previous year. There were highly significant differences i n height between the inbred seedlings and those resulting from both cross and wind-pollination on tree 2. The two inbred seedlings obtained from tree 1, also were both smaller than those that resulted from cross- and wind-pollination on the same tree. In the spring of 1955 the seedlings were a l l carefully moved to.a permanent location i n the nursery and planted out at eight-foot spacing.., The heights and diameters were measured i n the f a l l ; the difference between the heights of the inbreds and the seedlings from controlled cross-pollination on tree 2 was again found to be - 82 -highly significant. The development of the seedlings from tree 2 between 1953 and 1955 i s graphically illustrated in plate XIII. These graphs show that the seedlings from cross-pollination are superior to the inbreds in both height and diameter whereas those from wind-pollination are intermediate between them. This i s not unexpected since, in the case of wind-pollination, the pollen parent or parents are unknown, and uncontrolled self-pollination can easily occur. Statistical comparisons of the inbred seedlings with those from other types of pollination on the same tree are by no means the sole purpose of inbreeding studies, although they serve to emphasize the practical aspects of self-pollination in forestry practice. It is quite evident that the inbreeding of a naturally outbreeding species, such as Pseudotsuga will ordinarily result in some loss of vigour and that the seedlings will usually compare unfavourably with those from cross-pollination. Mather and Edwardes (1943) have pointed out that when an outbreeding system i s changed artif i c i a l l y to inbreeding, the inbred lines are much more variable and less vigorous than the outbreeding population from which they came. The polygenic systems controlling heights, vigour and f e r t i l i t y are adjusted to outbreeding, the system in force, and not to inbreeding. The frequency distributions and coefficients of variation of both the heights and diameters of the progeny of these trees are shown in table 23 and 24. These indicate that much more variation existed among the inbred seedlings of tree 2 than among those produced by cross-pollination. Considerable variation among the seedlings from wind-pollination was found and i t is very probable that a considerable amount of uncontrolled self-pollination took place in this tree. Some of the variation in the seedlings from wind-pollination is illustrated in figures 3 and 4, plate XI. These two seedlings were the same age and had been grown under identical conditions, yet their heights were six and thirty-four inches respectively. TABLE 23. FREQUENCY DISTRIBUTIONS AND COEFFICIENTS OF VARIATION BY HEIGHT CLASSES Height classes in inches (at age three years) Tree Pollin. 2- 7- 12- 17- 22- 27- 32- Total Mean C.V. 6 11 16 21 26 31 36 No. ht. 1 Self 1 1 2 11.6 _ Cross 6 5 2 13 23.0 — Wind 2 1 3 22.8 -2 Self 5 14 9 5 4 37 13.1 48.3 Cross 5 14 21 12 4 56 24.1 20.6 Wind 3 5 9 12 9 5 1 44 19.0 36.6 Non 5 2 7 10.2 28.7 TABLE 24. FREQUENCY DISTRIBUTIONS AND COEFFICIENTS OF VARIATION BY DIAMETER CLASSES Diameter classes in millimeters (at age three years) Tree Pollin. 3 - 7- 11- 15- 19- 23- Total Mean C.V. 6 10 14 18 22 26 No. diam. 1 Self 1 1 2 9.5 Cross 1 2 k 7 13 18.1 -Wind 1 2 3 18.7 -2 Self 9 8 13 3 4 37 10.6 45.9 Cross 1 7 16 27 5 56 18.2 19.7 Wind 3 8 12 7 12 2 44 14.7 36.9 Non 2 2 3 7 9.1 29.4 - 84 -Variation in Form Shull (1952, page 24) stated that in his early inbreeding studies with Zea mays, the difference between the self- and cross-fertilized rows were obvious. A very different explanation of the facts, however, was forced on him when i t was found that several self-fertilized rows differed widely from each other in morphological characteristics, indicating that they belonged to distinct elementary strains. He then realized that selfing merely served to purify the strains (since the homozygous lethals died) and that his comparisons were not properly between self-.and cross-fertilization but between pure strains and their hybrids. It is appreciated that distinct elementary strains can hardly be expected to appear after only one generation of inbreeding but much variation in form already can be found in the inbreds from tree 2. Figures 1, 2, 3 and 4, plate X, show four distinct forms, figure 1 illustrates a dwarf form which was only.two inches high and i t i s interesting to find that Jones (1924, page 310) reported similar dwarf forms when inbreeding Zea mays. Figure 2, illustrates a more vigorous but bushy form with no true leader, while the seedling in figure 4 is much larger and has a wide spreading form with considerable branch growth. This branching habit, moreover, was apparent in 1953. The seedling in figure 3, on the other hand, has a much better form and compared favourably with any of those from cross-pollination. Figures 3 and 4 both demonstrate that inbreeding does hot always result in reduction of vigour. This is - 85 -in agreement with results from the inbreeding of Zea mays (Hayes et al, page 77, 1955). The amount of variation wi l l depend entirely upon the genetic background of the parent trees which, in this investigation, is not' known; and could only be evaluated by an analysis of the progenies when they were older. East and Jones (1919, page 139) have pointed out: ... that inbreeding has but one demonstrable effect on organisms subject to i t s action - the isolation of homozygous types. The diversity of the resulting types depends directly upon number of heterozygous hereditary factors present in the individuals with which the process is begun; i t is likely, therefore, to vary directly with the amount of cross-breeding experienced by their immediate ancestors. The parent trees of the seedlings in both the 1952 and 1955 progeny tests were a l l from second growth stands and they could have resulted from either selfing, related or unrelated crosses. Inbreeding is the more effective method of exposing certain recessive characters which would rarely appear as a result of outbreeding. One such abnormal form, exposed by the self-pollination of tree 2, is illustrated in figure 5, plate XII. This small seedling developed with twin terminal buds, both of which have since produced shoots; figures 6 and 7 illustrate the subsequent development in 1954 and 1955. This abnormal form was not observed in any of the seedlings from cross-pollinations. Inbreeding also resulted in the appearance of other forms which were inherently weak and unable to survive once they were planted out. One type of seedling common to both the 1952 and 1955 progeny tests seemed unable to develop beyond the cotyledon stage and never produced - 86 -true leaves; one of these is illustrated in figure 2, plate XII. Inbreeding, therefore, can be expected to result in increased mortality when i t occurs in normal outbreeding species such as Pseudotsuga. It was found that the mortality of the inbreds in this progeny test was considerably higher than that of the seedlings resulting from controlled cross-pollination. Twelve out of a total of forty-nine inbred seedlings have died since 1953 as compared with only three out of a total of fifty-nine crossbreds. The mortality of the seedlings from wind-pollination was also high, eleven dying out of a total of fifty-five, this suggests that some uncontrolled inbreeding must have occurred. Variation between Progenies It i s hardly possible to compare the progenies of trees 1 and 2 owing to the small number of seedlings from the former tree. One striking difference, however, has been consistently observed and i t is quite independent of the type of pollination. The progeny from tree 1 regularly a l l flush at least a week before any of those from tree 2. This was observed in the springs of 1954 and 1955 in spite of transplanting. This early flushing habit has made the progeny of tree 1 particularly susceptible to late spring frosts which have already caused some damage. It was not surprising to find that the parent tree also flushed at least a week earlier than tree 2. Seedlings from Unpollinated Cones It will be recalled that twelve seeds obtained from the isolated and unpollinated cones of tree 2 germinated and the earlier information on their development has been published (Orr-Ewing, - 87 -1954). Only seven of these seedlings now remain and as can be seen from the graphs in plate XIII, they are inferior in both height and diameter to a l l the other progenies of tree 2. The number of samples is admittedly small, but i t can be concluded that the heights are significantly less than those of the seedlings from both cross-and wind-pollination and apparently less than the inbreds. The seedlings are extremely uniform and show l i t t l e variation in either height or diameter (see tables 23 and 24). This uniformity is illustrated in figures 1 and 2, plate XI; according to Warmke (1952), i t is characteristic of apomictic offspring as they are not subject to the effects of synapsis, segregation and recombination. Smears from the root tips have indicated that these seedlings are a l l diploid. It can not definitely be concluded that these seven seedlings are apomictic in origin although the evidence does suggest that this may be the case. - s a -i l RESULTS OF THE 1955 PROGENY TEST Seed Extraction and Gleaning The cones from the six trees were collected in the f a l l of 1954; those from trees A, C, and D were removed ten to twelve days earlier than those from the trees at Lake Cowichan. The cones were air-dried for several weeks and the seed was then extracted and dewinged. It was carefully cleaned to remove as much empty seed as possible and then counted before being stored in sealed containers at 4° C. Great care was taken throughout to ensure that no contamination occurred in any of the seed lots. The results of the seed extraction and cleaning are shown in table 25; the average number of seeds is based on the number found in the productive region of the sample cones removed for the cytological investigation. The flat and misshapen seeds, which are often found at the tip and base of a cone, are not included in this figure. The number of seeds remaining after cleaning indicates that much less seed is obtained after self-pollination than after either controlled cross- or wind-pollination. Very few cleaned seeds were obtained after the self-pollination of trees A, C, and D; this, however, was not unexpected since the amount of ovule collapse had also been very high. Selfing, however, had much less effect on trees F, H, and J; in tree F there was actually a higher percentage of cleaned seeds after self-pollination than after cross-pollination. In every other case, however, controlled cross-pollination yielded more f i l l e d seed. - 89 -TABLE 25 RESULTS OF SEED EXTRACTION AND CLEANING AFTER DIFFERENT POLLINATIONS. 1955. Tree Type of pollin. No. of Cones Aver. no. seeds per cone No. of seeds No. of cleaned seeds Per cent of cleaned seed A Self 82 37.4 3066 90 2.9 Cross 47 1756 634 36.1 Wind - 47 1756 456 26.0 C Self 47 45.4 2134 56 2.6 Cross 25 1135 598 52.7 Wind. 53 2406 856 35.6 D Self 16 27.6 442 6 1.4 Cross 24 662 217 33.0 Wind 73 2015 1375 68.2 F Self 72 44.7 3218 959 29.8 Cross 33 1475 420 28.5 Wind 168 7509 2935 39.1 H Self 56 60.7 3400 445 13.1 Cross 37 2246 1078 48.0 Wind 99 6010 1886 31.4 J Self 63 48.6 3061 454 14.8 Cross 41 1992 1051 52.8 Wind 56 2721 668 24.5 Sowing The different seed lots were fi r s t soaked in water for twenty-four hours and then stratified for a period of five weeks by the method recommended for Pseudotsuga by Allen and Bientjes (1954). The seed was then ready for sowing. A randomised block, split plot design was used in order that the environmental effects on the seed lots would be equalised as far as possible. Each of the eighteen seed lots was lightly dusted with a fungicide and divided - 90 -into equal numbers of seed i n order that they would be a l l represented i n each of the four standard seed beds at the nursery-Each of these four seed beds was divided up into six plots and each plot was divided into three sub plots. The s i x parent trees were assigned at random to each of the six plots i n each seed bed, the three types of pollination of each parent tree were then assigned at random to the three sub plots within the parent tree plot. One of the four seed beds i s i l l u s t r a t e d i n figure 1, plate XII. The method of seeding was modified from that reported by Liddicoet and Kimbrough (1950). The seed was individually sown i n spots which were six inches apart and seven to a row; the spots at the end of each row were four inches from the sides of the seed bed and the rows were six inches apart. The location of each spot within a row was marked mechanically by large headed nails, set at the appropriate intervals on the underside of a f l a t board which could be l a i d across the seed bed. Pressure was applied to this board, then i t was removed, and the position of each seed spot was visible on the s o i l surface. The seed was sown i n early May. The number per spot varied from one to three depending upon the amount of seed available; i f more than one was sown i n the same spot, half an inch was l e f t between each seed. The seeds were then covered by a thin layer of s o i l and their location marked with a tooth pick. This method of sowing had many advantages since records could be kept of each seed spot and the apparent germination readily obtained. The seed beds - 91 -were covered as a protection against birds and animals by fine meshed netting and poison bait was used as an added precaution against any possible rodent damage. The seed beds were shaded throughout the summer months and watered when necessary. Germination The seed spots were checked for germination at intervals of three days. The fi r s t germination took place sixteen days after sowing. Some delayed germination occurred owing to unusually cold weather but the time of germination of seeds from different pollinations did not differ markedly. At the end of June, a l l seeds which had not germinated were uncovered and examined to see whether or not they contained embryos; they were then recorded as either f i l l e d or empty. Germination percentage was generally high in a l l of the beds which indicated that the seed had been thoroughly cleaned. It was found that the precautions taken to equalise the environmental effects on the different seed lots in the four seed beds were justified. An analysis of variance showed that there were significant differences between beds, between trees and between pollinations. None of the interactions was significant and for this reason, the germination of each seed lot has been summarised in table 26. It was of some interest to find that isolation of the female cones must have interfered with the l i f e cycle of Megastigmus  spermotrophus. the Douglas f i r seed chalcid which can cause considerable damage. According to a recently revised publication - 92 -(Anonymous 1954), attack takes place in the spring when the female forces her long ovipositor through the bracts and scales of the cone and lays her eggs in the ovule. The eggs later hatch and the larvae feed on the contents of the seed, finally emerging the next spring as adults through a small hole. The results from table 26, however, indicate that only the seeds from wind-pollinated cones contained larvae of this seed chalcid. TABLE 26 SUMMARY OF GERMINATION RESULTS. 1955 Tree Type of Seeds Seeds Appar. Ungerminated seeds pollin. sown germin. germin. per Filled Empty Not With cent found larvae A Self 90 16 17.8 _ 62- 12 -Cross 420 330 78.6 7 61 22 -Wind 420 166 39.5 131 59 51 13 C Self 56 23 41.1 30 3 _ _ Cross 420 394 93.8 6 10 10 -Wind 420 222 52.9 29 121 28 20 D Self 6 4 66.7 1 1 _ — Cross 217 208 95.8 2 5 2 -Wind 420 381 90.7 1 23 15 — F Self 959 820 85.5 35 92 12 — Cross 420 400 95.2 4 13 3 -Wind 420 358 85.2 10 47 4 2 H Self 445 430 96.6 10 4 1 Cross 420 406 96.7 8 5 1 -Wind 420 395 94.1 12 7 4 2 J Self 454 363 79.9 21 63 7 _ Cross 420 397 94.5 7 3-3 3 -Wind 420 352 83.8 14 30 9 15 Seedling Mortality Records were kept of the mortality that occurred during 1955, almost a l l of which was caused by damping-off fungi. Mortality i n the four seed beds i s summarised i n table 27; the data indicates an apparent relationship between first-year survival and the type of pollination. Mortality of inbred seedlings was consistently higher than that of the crossbred seedlings, and losses of the seedlings from wind-pollination were intermediate. The higher mortality rate of the inbred seedlings was not unexpected since many obviously were far less vigorous than the crossbred seedlings. This i s clearly illustrated i n figure 2, plate XII, which shows an inbred and a crossbred seedling from the same parent and of almost identical ages. The mortality of the inbreds and of the crossbreds was compared by the Chi square method and the difference was found to be highly significant for those from trees A, D and H. TABLE 27 SUMMARY OF SEEDLING MORTALITY. 1955. Tree Self-pollination Cross-pollination Wind-pollination Total Dead Percent Total Dead Percent Total Dead Percent A 16 4 25.0 330 32 9.7 166 28 16.9 C 23 3 13.0 394 46 11.7 222 23 10.4 D 4 3 75.0 208 9 4.3 380 35 9.2 F 820 83 10.1 400 28 7.0 358 29 8.1 H 430 75 17.4 406 47 11.6 395 36 9.1 J 363 27 7.4 397 21 5.3 351 27 7.7 i >tal 1656 195 11.7 2135 183 8.6 1872 178 9.5 Comparative Height Development Thirty seedlings from each pollination type, for each of the six trees, were selected at random i n each of the four seed beds and their heights measured from ground level to the t i p of the terminal bud. In cases where there was only a few seedlings representing a particular pollination, they were a l l measured. It was again found that the precautions taken to equalise the environmental effects on the different seed lots in the seed beds were justified. An analysis of variance showed that there were significant differences between beds, between trees and between pollinations. None of the interactions were significant, however, so the heights from each pollination type have been summarised. The frequency distributions and coefficients of variation for the progeny of a l l six trees are shown in table 28. Selfing has again produced seedlings which are smaller than those from controlled cross-pollination" while those from wind-pollination are intermediate between them. The coefficients of variation for any one inbred line differed considerably, those for trees A and F, for example, being very close to those of the seedlings from controlled cross-, pollination. In the case of tree J, however, there i s a much greater difference between the two values. This variation is probably related to the undetermined genotype of the parent tree. Figures 2 and 3," plate XII, illustrate the variation in any one inbred line, the seedling in figure 2 i s inherently weak and unlikely to survive whereas the seedling in figure 3 is vigorous. This seedling already has a tendency to branch and should be compared with the seedling in figure 4 which has resulted from controlled cross-pollination of the same parent tree. It is of some considerable interest that in five of the trees, controlled cross-pollination has resulted in a higher proportion of large seedlings than were obtained from wind-pollination although no attempt was made to select particularly vigorous pollen parents. The comparative results of controlled cross-pollination on different trees are also of interest. The most vigorous crosses appeared to be those in which both parent trees came from different localities. The crosses A x F and J x C were both made with Victoria and Lake Cowichan trees, the superiority of their progeny at the present time over those from wind-pollination of the same parent trees is quite marked. It moreover can not be matched by any other cross between parents from the same locality. It remains to be seen whether such differences persist as i t is fully realised that hasty conclusions should not be made at this early stage of development. Several investigators such as Righter (1945) have effectively demonstrated that seed weight in the Panaceae is of considerable importance in i n i t i a l growth, and that the difference in inherent growth may be concealed by the effects of i n i t i a l seed weight. The effects of seed weight are only temporary but.they should not be overlooked at this stage of development. - 96 -TABLE 28 FREQUENCY DISTRIBUTIONS AND COEFFICIENTS OF VARIATION BY HEIGHT CLASSES Height classes in millimeters at end of fi r s t year Tree Pollen 10- 30- 50- 70- 90- 110- 130- Total Mean C.V. parent 29 49 69 89 109 129 149 no. ht. A Self 2 4 4 2 12 88.8 18.1 X F 1 5 22 45 41 6 120 101.2 18.6 Wind 3 9 13 37 45 11 2 120 83.9 28.0 C Self 1 3 9 3 2 1 19 63.7 34.5 X B* 2 7 24 44 27 16 120 80.7 28.2 Wind 4 20 47 38 9 2 120 83.2 23.8 D Self 1 1 X C 2 7 29 51 29 2 120 96.1 19.1 Wind 17 40 50 12 1 120 86.7 19.3 F Self 2 21 57 31 9 120 62.2 25.9 X H 1 5 32 41 37 4 120 77.8 24.7 Wind 2 24 46 21 24 3 120 66.8 29.7 H Self 3 61 39 15 1 1 120 50.8 31.4 X J 6 40 44 22 8 120 75.9 24.8 Wind 5 23 48 27 16 1 120 63.1 31.8 J Self 21 9 46 32 12 120 59.3 40.5 X C 1 6 12 37 41 19 4 120 88.9 25.2 Wind 7 10 42 46 13 2 120 67.5 31.4 * Tree B was also included in the investigation but was damaged later by fir e . - 97 -PLATE X Figure 1. A dwarf form of inbred seedling from tree 2. October, 1955. Figure 2. A bushy form of inbred seedling from tree 2. October, 1955. Figure 3. An inbred seedling from tree 2, comparable in form to those from cross-pollination. October, 1955. Figure H» A vigorous inbred seedling from tree 2 with a very branching habit, i t is twenty-six inches in height. October, 1955. P L A T E X - 98 -PLATE XI Figures 1 Two seedlings from tree 2, believed to be apomictic; and 2. note the similarity i n height and form. October, 1955. Figure 3. A seedling from the wind-pollination of tree 2) compare with figure 4. October, 1955. Figure 4. A vigorous seedling from wind-pollination of tree 2j i t i s thirty-four inches i n height. October, 1955. P L A T E X T -5= - 99 -PLATE XII Figure 1. One of the four seed beds used for the 1955 progeny-test, the laths : separate the progeny of one parent tree from another. October, 1955. Figure 2. Variation in vigour between an inbred and a seedling from controlled cross-pollination of the same parent tree, both are the same age and from tree C. October, 1955. Figure 3. A vigorous inbred from tree C with a pronounced branching habit. October, 1955. Figure 4* A seedling from controlled cross-pollination of tree C, note the improved form compared to figure 3. October, 1955. Figure 5. Twin terminal buds, possibly representing a recessive character revealed by inbreeding tree 2. October, 1953. Figure 6. The same seedling, one year later, October, 1954. Figure 7. The same seedling two years later, note the persistence of the two leaders. October, 1955. P L A T E XII - 100 -PLATE XIII The comparative heights and diameters of the progeny of tree 2 from 1953 to 1955. PLATE i m COMPARATIVE HEIGHTS OF S E L F E D , CROSSED, WIND POLLINATED AND "APOMICTIC" PROGENY FROM T R E E 2. 1953 1954 1955 YEAR OF MEASUREMENT. COMPARATIVE DIAMETERS OF SELFED, CROSSED, WIND POLLINATED AND "APOMICTIC" PROGENY FROM TREE 2. 1953 1954 Y E A R OF M E A S U R E M E N T . - 101 -III SOME PRACTICAL ASPECTS OF SELF-POLLINATION Natural Regeneration The fact that Pseudotsuga can be self-pollinated and produce seedlings which are generally inferior to those resulting from cross-pollination is of significance in planning for natural regeneration of cutover areas. Johnson (1945) previously pointed out the possibility that self-pollination will occur i f a few widely spaced individual trees are left to reseed a cutover area. He added that the silviculturist should, among other things, leave sufficient seed trees to provide for cross-pollination among individuals of the same species. It is possible that a loose form of inbreeding could take place on cutover. areas in the Douglas f i r region where such areas have regenerated from single or a small number of seed trees. This important aspect of regeneration has been given scant attention and no studies have yet been conducted on the crucial matter of distance of pollen flight in Pseudotsuga. Several misconceptions, moreover, have arisen in regard to. self-pollination in Pseudotsuga. It was formerly considered that viable seed could not be obtained from selfing and, indeed, that self-pollination itself was almost impossible, owing to the different times of maturation of the male and female cones. Syrach Larson (1937, page 67) believed that Pseudotsuga is markedly metandric and that self-pollination can occur to only a limited extent. This statement was corrected by Allen (1942), who pointed out that many trees showed no evidence of dichogamy and that there was much variation from tree to tree. Some years later, however, - 102 -Isaac (1949, page 49) wrote:. Since pistillate flowers of Douglas f i r receive their pollen from staminate flowers of another Douglas f i r tree, there is certain to be cross-fertilization with other good trees in the plantation. Consequently, the line degeneration with attendant loss of vigor, which usually accompanies self-fertilization is automatically and naturally avoided. This statement is, of course, both incorrect and misleading and can not be substantiated by experimental evidence. On the contrary, this study has shown that there are definite opportunities for natural self-pollination in some trees. Particular attention was given to the development of both the male and female cones and, in several trees, pollination was in progress at the time when the female cones had reached the peak of the receptive stage, in addition, the appearance of some of the seedlings produced by wind-pollination was very similar to that of the inbreds and in a l l probability they were the result of self-pollination. It can be definitely concluded that self-pollination does occur under natural conditions in Pseudotsuga. that self-pollination does result in variable amounts of viable seed, and that the resulting seedlings will usually be less vigorous than those produced by cross-pollination. These proven facts should be of some significance in the selection of silvicultural methods for regenerating cutover land. Artificial Regeneration The knowledge that Pseudotsuga can produce viable seed after self-pollination should have obvious implications in regard to - 103 -cone collections. The necessity of not collecting cones from isolated trees and for ensuring that there has been adequate cross-pollination i n young stands producing cones needs no further elaboration. The deleterious results of self-pollination could also be a matter of some significance i n the establishment of seed orchards. There has been considerable emphasis i n recent years on the necessity for improving the inherent quality of seed for reforestation purposes and seed orchards have been recommended for this purpose. These would consist of a number of clones taken from selected trees thought to be inherently superior i n growth and form to the remaining trees i n a stand. It i s hoped that the clones from these selected trees w i l l cross-pollinate freely and produce abundant superior seed. Various suggestions for the formation of such orchards have been proposed by Isaac (1949), Duffield (1950), and Meyer (1951), but these were a l l based on the improbability of viable seed being produced after self-pollination. It i s clear, however, that much careful planning of these orchards w i l l be necessary i f inbreeding i s to be reduced as much as possible. This danger i n seed orchards has already been appreciated i n Europe. Langner (1953), for example, has recently made a detailed study of the most desirable designs for planting out ten clones from each of ten selected trees i n order to reduce self-pollination. There has i n addition been some controversy i n regard to the number of clones that should be represented i n a seed orchard. In Sweden, Gustaffson (1950) has stated that the clones - 104 -should not be too few i n number i n order to prevent a depauperization of genes as well as future risks of homozygotization and inbreeding depression of variable intensity. He has suggested that there should be at least twenty to th i r t y clones from as many parent trees to represent each provenance zone. Jensen (1954> page 63), on the other hand, has pointed out that he has been repeatedly c r i t i c i z e d for selecting only nine parent trees to represent each provenance zone i n Sweden. It was f e l t that this small number could lead to inbreeding deterioration. In his opinion there was no theoretical proof that inbreeding appeared as a result of the crossing of a limited number of parent individuals i n a natural population. Such controversies show that the formation of a seed orchard i s no easy matter. In conclusion i t should be pointed out that inbreeding could also be of value to future forestry practice. Ehrenberg et a l (1955) have recently pointed out that a seed orchard u t i l i z i n g the phenomenon of hybrid vigour could be established on the basis of inbred lines. The intercrossing of such lines, moreover, might i n some cases effect a higher heterosis than i n the other types of heterotic orchards. Wright (1953) again has concluded that recent developments such as the induction of precocious reproduction, indicate that an inbreeding program started now could lead to very useful results within the next half century. He also pointed out that several tree breeders at the present time are maintaining selfed lines for the use of their successors. Such long term planning presents a challenge to a l l concerned with programs i n forest genetics. - 105 -BIBLIOGRAPHY Allen, G.S. 1942. Parthenocarpy, parthenogenesis and s e l f - s t e r i l i t y of Douglas f i r . J. For. 40: 642-644. . 1943. The embryogeny of Pseudotsuga t a x i f o l i a (Lamb.) B r i t t . Amer. J. Bot. 30: 655-681. . 1945. Embryogeny and the development of the apical meristems of Pseudotsuga t a x i f o l i a (Lamb.) B r i t t . Diss., Univ. of California, Berkeley. Allen, G.S. and W. Bientjes. 1954. Studies on coniferous tree seed at the University of British Columbia. For. Chron. 30: 183-196. Andersson, E. 1947. Pollen and seed setting studies of an asyndetic spruce and some normal spruces; and a progeny test of spruces. Medd. Fdren. V&ctfflradl. Skogstrfld. 45: 1-22. Svensk PappTidn. 4/7. Anhaeussei*, H. 1953. Keimung und Schlauchwachstum des Gymnospermenpollens unter besonderer Berttcksichtigung des Wuchsstoffproblems. Beitr. B i o l . Pflanzen 29: 297-338. Anonymous. 1954. Megastigmus f l i e s attacking conifer seed. Leafl. 8. For. Comm. Lond. Bateman, A.J. 1952. Self-incompatibility systems i n angiosperms. I. Heredity. 6: 285-310. Bingham, R.T. and A.E. Squillace. 1955. Self-compatibility and effects of s e l f - f e r t i l i t y i n western white pine. For. S c i . 1: 121-129. Branscheidt, P.C. 1939. Befruchtungsphysiologische Untersuchungen an Taxus baccata. Ber. dtsch. Bot. Ges. 57s 495-505. Brink, R.A. 1952. Inbreeding and cross breeding i n seed development. In Heterosis, ed. by J.W. Gowen, pp. 81-97, Iowa State College Press., Ames. Brink, R.A. and D.C. Cooper. 1940. Double f e r t i l i z a t i o n and development of the seed i n Angiosperms. Bot. Gaz. 102: 1-25* Buchholz, J.T. 1918. Suspensor and early embryo of Pinus. Bot. Gaz. 66: 185-228. . 1945. Embryological aspects of hybrid vigor i n pines. Science. 192: 135-142. - 106 -Buchholz, J.T. 1946. Volumetric studies of seeds, endosperms and embryos., i n Pinus ponderosa during embryonic differentiation. Bot. Gaz. 108: 232-244. Chajaberlain, C.J. 1935* Gymnosperms, structure and evolution. 484 pp. Univ. of Chicago Press., Chicago. Coleman, L.C. 1938. Preparation of leucobasic fuchsin for use i n the feulgen reaction. Stain Tech. 13: 123-124. Cooper, D.C. and R.A. Brink. 1940. Partial self-incompatibility and the collapse of f e r t i l e ovules as factors affecting seed formation i n a l f a l f a . J. Agric. Res. 60: 453-472. Coulter, J.M. and C.J. Chamberlain. 1917* Morphology of Gymnosperms. 466 pp. Univ. of Chicago Press., Chicago. Cumming, W.C. and F.I. Righter. 1948. Methods used to control pollination of pines i n the Sierra Nevada of California. Circ. 792. U.S. Dep. Agric. Darwin, C. 1900. The effects of cross- and s e l f - f e r t i l i z a t i o n i n the vegetable kingdom. 487 pp. J. Murray., London. Dengler, A. 1932. Kttnstliche Bestfluberungsversuche en Kiefern. Z. Forst.u. Jagdw. 64: 513-555. • . 1939. Uber die Entwicklung kttnstlicher Kiefernkreuzungen. Z. Forst. u. Jagdw. 71: 457-485. Dobzhansky, T. 1951. Genetics and the origin of species. 364 pp. Columbia Univ. Press., New York. . 1952. Nature and origin of heterosis. __ Heterosis ed. by J.W. Gowen, pp. 218-223. Iowa State College Press., Ames. Duffield, J.W. 1950. Techniques and po s s i b i l i t i e s for Douglas f i r breeding. J. For. 48: 41-45. . 1954. Studies of extraction, storage and testing of pine pollen. Z. Forstgenet. 3* 39-45. East, E.M. 1929. S e l f - s t e r i l i t y . Bibliogr. Genet. 5: 331-368. East, E.M. and D.F. Jones. 1919. Inbreeding and outbreeding. 277 pp. J.B. Lippincott Co., Philadelphia. Ehrenberg, C , A. Gustafsson, C. Forshell, and M. Simak. 1955. Seed quality and the principles of forest genetics. Hereditas, Lund. 4: 291-366. - 107 -Forshell, CP. 1953. Kottens och frflets utbildning efter sjftlv-och korsbefruktning hos t a l l (Pinus s i l v e s t r i s L.) Medd. Skogsfflrsknlnst, Stockh. 43: 1-42. Goebel, K. 1932. Organographie der Pflanzen. 3. 1379-1820. G. Fischer., Jena. Gothe, H. von. 1952. Ein Kreuzungsversuch mit Larix europea D.C., Herkunft Schlitz, und Larix leptolepis Gord. 1. Z. Forstgenet. 1: 108-110. . 1953. Ein Kreuzungsversuch mit Larix europea D.C., Herkunft Schlitz, und Larix leptolepis Gord. 2. Z. Forstgenet. 2: 122-125. Gustafsson, A. 1946. Apomixis i n higher plants. I. The mechanism of Apomixis. Lunds Univ. Arsskr. N.F. Avd. 2, 42 (3): 66 pp. . 1950. Conifer seed plantations: their structure and genetical principles. 3rd World For. Congr., 117-119. Helsinki. Hayes, H.K., F.R. Immer, and D.C. Smith. 1955. Methods of plant breeding. 551 pp. McGraw-Hill Book Co., New York. Hummel, 0. 1930. Aus der Biologie des Samentragens der Waldbflume. Einiges tiber Parthenokarpie und Parthenospermie. Z. Forst. u. Jagdw. 62: 365-371. Hutchinson, A.H. 1915. F e r t i l i z a t i o n i n Abies balsamea. Bot. Gaz. 60: 457-472. Isaac, L.A. 1949. Better Douglas f i r from better seed. 64 pp. Univ. of Washington Press., Seattle. Jensen, H. 1954. The establishment of forest tree seed orchards at Ramsldsa 1941-1954. Goteborg. Bot. Tradgard. Meddel. 19: 157-192. Johansen, D.A. 1940. Plant microtechnique. 523 pp. McGraw-Hill Book Co., New York. • 1950. Plant embryology. 305 pp. Chronica Botanica Co., Waltham. . 1954. Letter to the writer. February 26. Johnson, L.P.V. 1945. Reduced vigour, chlorophyll deficiency and other effects of s e l f - f e r t i l i z a t i o n i n Pinus. Canad. J. Res. C. 23: 145-149 - 108 -Jones, D.F. 1925. Genetics i n plant and animal improvement. 568 pp. J. Wiley & Sons., New York. Kolessnikoff, A.T. 1929. Uber die Notwendigkeit die probleme der forstlichen Genetik und Veredlung durch Forstversuchsanstalten zu studieren und fiber einige Result ate der i n der Ukraina zur Erforschung dieser Probleme angestellten Versuche. Verhandl. Int. Kong, f o r s t l . VersAnst., 427-430, Stockholm. Langlet, 0. 1940. 0m utvecklingen av granar ur frfl efter sjfllvbefruktning och efter vindpollinering. Medd. Skogsfdrsknlnst., Stockh. 32: 1-22. Langner, W. 1951. Kreuzungsversuche mit "Larix europaea D.C. und Larix leptolepis Gord. 1.'. Z. Forstgenet. 1: 2-18. — . 1952. Kreuzungsversuche mit Larix europaea D.C. und Larix leptolepis Gord. 2. Z. Forstgenet. 1: 40-56. — . 1953* Die Klonanordnung i n Samenplantagen. Z. Forstgenet. 2: 119-121. Lawson, A.A. 1909. The gametophytes and embryo of Pseudotsuga  douglasii. Ann: Bot. Lond. 23: 163-180. Lewis, D. 1944. Incompatibility i n plants. Nature, Lond. 153J 575-578. . 1954. Comparative incompatibility i n angiosperms and fungi. In Advances i n Genetics 6. ed. by M. Demerec, pp. 235-285. Academic Press., New York. Liddicoet, A.R. and E.F. Kimbrough. 1950. Pinus: Propagation of experimental stock. For. Res. Note 56. C a l i f . For. Range Exp. Sta. Lyons, L.A. 1956. The seed production capacity and efficiency of red pine cones (Pinus resinosa Ait.) Canad. J. Bot. 34J 27-36. McLean, R.C. and W.R. Ivimey-Cook. 1951. Textbook of theoretical botany. 1. 1069 pp. Longmans, Green and Co., London. Mather, K. 1943. Specific differences i n Petunia. I. Incompatibility. J. Genetics. 45* 215-235. . 1944. Genetical control of incompatibility i n angiosperms and fungi. Nature, Lond. 153: 392-394. Mather, K. and P.M.J. Edwardes. 1943. Specific differences i n Petunia. III. Flower colour and genetic isolation. J. Genet. 45: 243-260. - 109 -Meyer, H. 1951. Aufgaben und Wege der Douglasienzttchtung. A l l g . Forstzeitschr. Nr. 27/28: 281-283. Nienstaedt, H. and H.B. Kriebel. 1955. Controlled pollination of eastern hemlock. For. Sci. 1: 115-120. Orr-Ewing, A.L. 1954. Inbreeding experiments with the Douglas f i r . For. Chron. 30: 7-16. Piatnitsky, S.S. 1936. Experiments on self-pollination of Larix, Acer and Quercus. U.S. For. Serv. Transl. 290. 22 pp. Righter, F.I. 1939. A simple method of making germination tests of pine pollen. J. For. 37: 574-576. . 1945. The relationship of seed size and seedling size to inherent vigour. J. For. 43: 131-137. Sass, J.E. 1940. Elements of botanical microtechnique. 222 pp. McGraw-Hill Book Co., New York. Saxton, W.T. 1909. Parthenogenesis i n Pinus pinaster. Bot Gaz. 47: 406-409. Scamoni, A. 1950. fiber die weitere Entwicklung kunstlicher Kiefernkreuzungen i n Eberswalde. Zttchter 20: 39-42. Schopf, J.M. 1943. The embryology of Larix. 111. Bi o l . Monogr. 19: 5-97. Sears, E.R. 1937. Cytological phenomena connected with s e l f -s t e r i l i t y i n the flowering plants. Genetics. 22: 130-181. Sedgwich, P.J. 1924. Lif e history of Encephalartos. Bot. Gaz. 77: 300-310. Shull, G.H. 1952. Beginnings of the heterosis concept. In Heterosis ed. by J.W. Gowen, pp. 14-48. Iowa State College., Ames. Stebbins, G.L. 1950. Variation and evolution i n plants. 643 pp. Columbia Univ. Press., New York. Sylven, N. 1910. 0m pollineringsfd'rso'k med t a l l och gran. Medd. Skogsfdrsknlnst, H. Stockh. 7: 219-228. Syrach Larsen, C. 1937. The employment of species, types and individuals i n forestry. Roy. Vet. Agr. C o l l . Y.B. 154 pp. Reitzel., Copenhagen. Wardlaw, C.W. 1955. Embryogenesis i n plants. 381 pp. Methuen & Co.: Ltd., London. - 110 -Warmke, H.E. 1952. Aponiixis i n grasses with special reference to Panicum iraximum. 209-215. Proc. Sixth Int. Grassland Congr. Whitehouse, H.L.K. 1950. Multiple-allelomorph incompatibility of pollen and style i n the evolution of the angiosperms. Ann. Bot. Lond. 14: 199-216. Wright, J.W. 1953. A survey of forest genetics research. J. For. 51: 330-333. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0106274/manifest

Comment

Related Items