UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Meiosis and pollen stainability in Prunus avium L. cv. Lambert Whelan, Ernest David Pratt 1965

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

Item Metadata

Download

Media
831-UBC_1965_A4 W5.pdf [ 8.06MB ]
Metadata
JSON: 831-1.0302507.json
JSON-LD: 831-1.0302507-ld.json
RDF/XML (Pretty): 831-1.0302507-rdf.xml
RDF/JSON: 831-1.0302507-rdf.json
Turtle: 831-1.0302507-turtle.txt
N-Triples: 831-1.0302507-rdf-ntriples.txt
Original Record: 831-1.0302507-source.json
Full Text
831-1.0302507-fulltext.txt
Citation
831-1.0302507.ris

Full Text

MEIOSIS AND POLLEN STAINABILITY IN PRUNUS AVIUM L. cv. LAMBERT. by ERNEST DAVID PRATT WHELAN B. S.„A., Ontario Agricultural College, 1963 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE IN AGRICULTURE in the Division of Plant Science We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA Apr i l , 1965 I n p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f t h e r e q u i r e m e n t s f o r an a d v a n c e d d e g r e e a t t h e U n i v e r s i t y o f • B r i t i s h C o l u m b i a , I a g r e e t h a t t h e L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and s t u d y * I f u r t h e r a g r e e t h a t p e r -m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y p u r p o s e s may be g r a n t e d by t h e Head o f my D e p a r t m e n t o r by h i s r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t . c o p y i n g o r p u b l i -c a t i o n . o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l n o t be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n * XJKpQQfXXXJeoer Plant Science The U n i v e r s i t y . o f B r i t i s h C o l u m b i a , V a n c o u v e r 8 ? C a n a d a D a t e May 5. 1965 i i ABSTRACT Variation in pollen v i a b i l i t y and the meiotic behaviour of sweet cherry Prunus avium L. cv. Lambert were studied. Pollen v i a b i l i t y was estimated using aceto-carmine. Three types of pollen grain were observed; large, triangular; small, aborted; and intermediate. Only the f i r s t type were considered to have been viable at time of blossom harvest. No significant d i f -ferences were detected between trees, between branches, or within branches, except for a North and South branch of one tree. Highly significant differences were detected among blossoms in a bud. The relative frequency of abnormal pollen was 42.26 per cent. Premeiotic budwood was stored at -5°C. Subsequent forcing v at 10°, 15°, 20°, and 25°C to bring the buds to meiosis, and staining in alcoholic hydrochloric acid-carmine, revealed meiotic abnormalities at a l l temperatures except 15°C. Early meiotic division stages of additional material forced at this temperature were synchronous, but asynchrony developed after f i r s t metaphase. The f i r s t meiotic division was characterized by the frequent presence of a single pair of uni-valents, which appeared to arise from desynapsis of a bivalent i n diplo-tene. Meiotic abnormalities associated with the univalents were observed at later division stages. The metaphase-anaphase division was typified by sequential disjunction of the bivalents, characteristic for sweet cherry meiosis. The relative frequency of abnormal f i r s t meiotic divisions was 45.4^ 9 per cent. i i i ACKNOWLEDGEMENTS I would like to thank Dr. C. A. Hornby, Associate Professor, Division of Plant Science, University of British Columbia, under whose supervision this project was undertaken, for his encouragement and guidance i n both the project and the writing of the thesis; Dr. K. I. Beamish, Department of Botany, for her helpful suggestions i n the cytological aspects of the study; Dr. G. W. Roberts, Department of Poultry Science, for advice in the s t a t i s t i c a l analysis of the data. Grateful acknowledgement is also extended to the members of my thesis committee, Dr. V. C. Brink, Dr. D. P. Ormrod, and Dr. G. W. Eaton, Division of Plant Science, for their constructive criticisms. My special thanks to Dr. A. R. Forbes, Canada Depart-ment of Agriculture, for the use of the Leitz Ortholux micro-scope, without which the cytological study and photomicrography would have been very much more d i f f i c u l t , and to Mr. K. 0. Lapins, the Research Station, Summerland, for providing the materials used i n this project, and commissioning the studies under an Extra-Mural Research Grant from the Canada Department of Agric-ulture. And f i n a l l y , my thanks to my wife for her patience, encouragement and assistance during the project. i v TABLE OF CONTENTS INTRODUCTION 1 LITERATURE REVIEW 3 MATERIALS AND METHODS 13 A. Materials 13 B. Methods 13 I. Pollen Study . . . 13 1. Pollen Stainability 13 2. Sampling of Blossoms and Pollen 16 (a) Precision of pollen grain count 16 (b) The effect of sub-sample size 17 (c) Variation in pollen stainability within a tree 18 (d) Variation in stainability between trees • 19 (e) S t a t i s t i c a l analysis 19 II. Cytological Study 22 1. Forcing Temperature . . . . . . . . 22 2. Staining Technique . 23 3. Meiotic Analysis 25 RESULTS -26 I. Pollen Study 26 1. Pollen Stainability 26 2. Sampling of Blossoms and Pollen 28 (a) Precision of pollen grain count . . . . . . . 28 (b) The effect of sub-sample size 31 Experiment 1 31 V (c) Variation in pollen stainability within a tree 37 Experiment 2 37 Experiment 3 37 (d) Variation in pollen stainability between trees 42 Experiment 4 42 II. Cy to logical Study 46 1. Forcing Temperature 46 2. Stain Technique 49 3. Meiotic Analysis of Prunus avium L. cv. Lambert .50 (a) Synchrony of meiotic division . 51 (b) First meiotic division 51 (c) Second meiotic division 63 (d) Frequency of normal and abnormal f i r s t meiotic divisions 66 DISCUSSION . . . . . 72 I. Pollen Study 72 II. Cytological Study 74 SUMMARY AND CONCLUSIONS 79 LITERATURE CITED 82 v i LIST OF TABLES p a g e Table 1. Experimental materials used i n the pollen sampling study 14 Table 2. The expected mean squares of the analyses of variance . . . . 20 Table 3. Sources of variation for s t a t i s t i c a l analyses using the expected mean squares in Table 2 21 Table 4. Comparison of percentage normal pollen obtained from duplicate 500-grain counts for each of single-blossom and five-blossom sub-samples 29 Table 5. Comparison of the variances and means of each of the nine samples with duplicate 500-grain pollen counts 30 Table 6. Effect of single-, five-, and 10-blossom sub-sample size on estimated percentage normal pollen 32, 33 Table 7. Effect of the number of grains used on the estimate of percentage normal pollen. Single-blossom sub-samples . . . . . 35 Table 8. Effect of the number of grains used on the estimate of percentage normal pollen. Five-blossom sub-samples 36 Table 9. Effect of a North and South branch from a single tree on estimated percentage normal pollen 38 Table 10. Analysis of variance of five-blossom sub-samples from two locations within a North and South branch 39 Table 11. Effect of two randomly selected branches from a single tree on estimated percentage normal pollen 40, 41 Table 12. Analysis of variance of five three-blossom buds randomly selected from each of two branches from a single tree . . . . 43 Table 13. Analysis of variance of five three-blossom buds randomly selected from a single branch . 43 Table 14. Effect of two different trees (adjacent) on estimated percentage normal pollen 44 Table 15. Analysis of variance of randomly selected samples from two different trees 45 Table 16. Meiotic abnormalities present in the f i r s t meiotic division i n anthers of Lambert sweet cherry 67 v i i LIST OF TABLES - Continued Table 17. Possible univalent locations and their theoretical frequencies per meiotic event in f i r s t metaphase 69 Table 18. Number of pollen mother cells observed for each of the four possible univalent positions in f i r s t metaphase . . . 69 Table 19. Number of pollen mother cells containing univalents in f i r s t metaphase and subsequent observed f i r s t telophase distributions with related abnormalities . 71 v i i i LIST OF FIGURES Page Plate I. Fig. 1-4. Pollen grains of Prunus avium L. cv. Lambert stained with aceto-carmine . . . . 2 7 Plate II. Fig. 5-8. Pollen mother cells from material forced at 20° and 25°C 47 Plate III. Fig. 9-12. Pollen mother cells from material forced at 25°C 48 Plate IV. Fig. 13-16. Synchrony of division in early meiotic stages • 52 Plate V. Fig. 17-20. Pre-meiotic to pachytene stages 54 Plate VI. Fig. 21-23. Pachytene to diplotene 55 Plate VII. Fig. 24-27. Mid-diplotene to diakinesis 56 Plate VIII. Fig. 28-31. Diakinesis to late anaphase 58 Plate IX. Fig. 32-35. The four observed univalent positions during the f i r s t metaphase-anaphase division . . . . . . . . 60 Plate X. Fig. 36-39. Abnormal f i r s t meiotic divisions . 61 Plate XI. Fig. 40-43. Interkinesis to telophase II 64 Plate XII. Fig. 44-45. Abnormal second meiotic divisions 65 INTRODUCTION Sweet cherry trees (Prunus avium L.) grow very vigorously and become d i f f i c u l t to manage i n later years, thus increasing the costs of orchard management and harvesting. Dwarf or semi-dwarf trees would be highly desirable i f good productivity of high quality f r u i t could be coupled with this habit of growth. So far, conventional breeding programmes have not been able to combine these desirable characteristics. Furthermore, no satisfactory dwarfing rootstocks are available. Therefore plant breeders turned to induced mutations for the production of compact growth habit. Lapins (1963) produced potentially valuable mutants by exposing sweet cherry scions to ionizing radiation. Two of the mutants are of special interest because they have the desirable compact habit, and the f r u i t has the characteristics of the popular Lambert variety. This variety i s the most important i n the Pacific Northwest, accounting for 49.6 per cent of the sweet cherry trees in the Okanagan Valley of British Columbia (Anon., 1960), and 47 per cent of the sweet cherry trees in North Central Washington (Wilcox, 1955). The extensive use of Lambert warrants concern with the breeding behaviour of the two mutants. The normal habit Lambert is the obvious base of comparison for the mutants, but there i s no published report on meiosis and pollen status i n this variety. In order to study the usefulness of the induced mutations, i t was 2. necessary to make an i n i t i a l study of the normal type Lambert. The present investigation involved experiments (1) to study methods of pollen sampling, (2) to estimate pollen abortion in normal Lambert, (3) to find a temperature range within which dormant blossom buds could be forced without inducing chromosomal aberrations, (4) to establish a cytological technique for meiotic studies, and (5) to investigate whether any meiotic abnormalities were present that might account for the degree of pollen abortion that preliminary experiments had shown to be present in this variety. 3. LITERATURE REVIEW Literature i s reviewed as follows: (1) methods used i n pollen v i a b i l i t y determination, (2) sampling of blossoms and pollen, (3) temperature effects on meiosis, (4) cytological techniques used to study meiosis i n Prunus species and (5) meiosis in Prunus avium L. 1. Methods Used i n Pollen V i a b i l i t y Determination There is a general lack of agreement as to the best method of estimating pollen v i a b i l i t y . Two of the standard methods are: (a) Staining — examination of pollen using some staining procedure to show differences among pollen grains within the sample. (b) In vitro determination — the germination of pollen on a r t i f i c i a l media. (a) Staining The most common methods involve staining in either aceto-carmine or iodine-potassium-iodide. The val i d i t y of either method has been questioned in recent years. Hauser and Morrison (1964) con-cluded that for a large number of species, a test based on the cyto-chemical reduction of nitro blue tetrazolium was preferable. King (1960) recommended a v i t a l test involving the oxidation of benzidine by peroxidase i n the presence of hydrogen peroxide i n preference to normal stain or i n vitro tests. He considered that the erratic response of pollens to germination in a r t i f i c i a l media necessarily made the germination 4. test unreliable, and did not consider normal staining methods measured pollen v i a b i l i t y . In sweet cherry, Galletta (1959) using a modified aceto-carmine glycerol j e l l y found a close agreement between shape and size of pollen, and stainability. He considered stainability a more accurate measure of functional pollen than germination tests for the clones investigated. Raptopoulos (1940) found a close agreement between pollen grain size and germination, and suggested that the diameter of the grain be adopted as the criterion of v i a b i l i t y i n cherries. His data showed that the percentage of pollen grains of non-hybrid cherries, having a diameter of 33-48 ^ i , did not di f f e r more than 5.6 per cent from their maximum germination percentage. (b) In vitro germination Pollen germination tests i n a r t i f i c i a l media have given highly variable results with many different plants. In sweet cherry, Raptopoulos (1940) investigated 41 diploid varieties. He obtained greater than 80 per cent germination in 24 varieties and less than 50 per cent i n only five cases. These results are in marked contrast to other investigators. Eaton (1961) obtained germination ranging from 45 to 61 per cent and 44 to 56 per cent i n two different experiments using a sample of commercial cherry pollen. Galletta (1959) reported germination ranging from 19 to 60 per cent from 23 clones i n one season, and the following year, no clone exceeded 12.3 per cent. He concluded that stainability was more indicative of differences i n potential pollen 5. v i a b i l i t y than i n vitro tests. None of these three investigators reported whether percentage germination was based on a l l grains seen, or whether obviously aborted grains were ignored. Such differences i n technique might account for part of the wide variation i n reported percentage pollen germination. It has been suggested that such variables as chemotropic factors (Rosen, 1961), pH level, boron-sugar content (Vasil, 1960) and pollen population effects (Brewbaker and Majumder, 1961; Brewbaker and Kwack, 1963) account, i n part, for the erratic responses of pollen germination in vitr o . 2. Sampling of Blossoms and Pollen Information i s scarce on both sampling procedure and the technique of making counts. (a) Microscope f i e l d Marks (1954) stated that empty grains, being lighter, tended to move towards the edges of the coverslip when i t was lowered. To overcome this d i f f i c u l t y he used aceto-carmine glycerol j e l l y which was more viscous than aceto-carmine, and prevented empty grains from d r i f t -ing towards the edges of the coverslip. Eaton (1961), in estimating pollen germination i n sweet cherry, examined the slide from side to side u n t i l 500 grains were counted. This method of examination was also used by Lapins (1964). 6. (b) Sample size Few reports have mentioned the i n i t i a l sample size, the number of blossoms i n the subsample, or the reason for selecting the number of grains counted. Hough (1939), de Nettancourt and Grant (1963), and Morris and Isikon (1964) reported results of pollen studies without stating the number of grains counted. Hauser and Morrison (1964) made counts which varied from 139 to 2,282 grains. Carnahan and H i l l (1955) counted from 100 to 1,200 grains for Lolium and Festuca hybrids. In sweet cherry pollen germination studies, Crane and Brown (1937) counted 600 to 2,000 grains, Raptopoulos (1940) 1,000 to 2,000, and Eaton (1961) used 500. Using stainability tests, Lenander (1962) counted 200 grains from a 10 to 20 blossom sample, and Galletta (1959) used 100 grains. Lenander (1962) i s the only investigator of sweet cherry to report the number of blossoms i n the i n i t i a l sample. 3. Temperature Effects on Meiosis Lenander (1962) reported that meiotic divisions were seen only when the temperature was greater than 9° or 10°C. In his study this temperature was rarely exceeded during meiotic development in one season, and meiosis was spread out over a three week period. In another season meiosis lasted only four days. During this time maximum 7. temperatures ranged from 13.8° to 20.1°C. Meiotic studies suggested that observed univalents were not due to low temperature. Temperatures greater than 17° to 18°C appeared to disturb meiosis. Varieties characterized by univalents were more susceptible to annual variation in pollen f e r t i l i t y than those i n which univalents were rarely seen. Several investigators have reported on the effect of temperature on meiosis i n higher plants. Pao and L i (1948) invest-igated the effects of temperatures ranging from 25° to 45° on meiosis in Triticum, Secale, Hordeum and Vicia species and found both direct and after effects of high temperature. Direct effects included lack of formation of the normal spindle, bivalents scattered randomly throughout the PMC (pollen mother c e l l ) , and "paralysis" of the centromeres resulting in non-disjunction and formation of one, two or three groups of bivalents at telophase. The major after effect was univalent formation. Univalent formation was attributed to the prevention of crossing-over between homologous chromosomes at early prophase. Univalent frequency reached a maximum two to four days after treatment in diploid and tetraploid Hordeum, three days i n Vicia faba, and almost immediately in V. cracca. Dollinger (1942) attributed lethal effects and meiotic irregularities including increased univalent frequency, and fewer quadrivalents and chiasmata to exposure to temperatures up to 37°C i n Dactylis glomerata L. Grun (1952) attributed to low temperature the increased univalent frequency observed in Poa hybrids grown at different elevations. 8. 4. Cytological Techniques Used for Studying Meiosis i n Prunus species The somatic chromosomes of Prunus species are only 1 to 3 p. long and are d i f f i c u l t to stain (Darlington, 1928). Darlington (1934), Raptopoulos (1941), Galletta (1959) and Keep (1963) used the squash technique in conjunction with B e l l i n i ' s (1926) iron aceto-carmine, or some modification thereof. These investigators commented on the small size but complex structure of the bivalents which made interpretation very d i f f i c u l t because good contrast was often prevented by cytoplasmic staining. Lenander (1962) studied sectioned material stained with crystal v i o l e t , and used Feulgen and Light Green staining for studying micronuclei. 5. Meiosis i n Prunus avium L. Darlington (1928) i n i t i a l l y reported sweet cherries as being aneuploids, having 16 to 19 chromosomes. The chromosome number is now known to be n=8 from the work of Kobel (1927), Darlington (1934) and Hruby (1939). Investigations on meiosis in sweet cherry have been reported by Hruby (1939), Raptopoulos (1941), Galletta (1959) and Lenander (1962). Only Lenander reported and discussed the varieties separately. None of these studies included the variety Lambert.-Hruby (1939) examined three varieties and considered meiosis to be characterized by nearly complete regular pairing (8 bivalents) with low chiasma frequency. He observed binucleate PMC with more than two 9. spindles, and believed that abnormal pollen tetrads having five, six or eight spores resulted from such irregularities. The number of PMC studied was not reported and differences among the varieties were not mentioned. Raptopoulos (1941) considered both meiotic div-isions to be normal. At diakinesis eight paired configurations, one of which was attached to the nucleolus, were always seen. Examination of the eight bivalents present at metaphase I showed four to seven were rods and one to six rings. . The chromosomes were joined by one to three chiasmata, usually subterminal. In a few cases two univalents were present which he believed resulted from failure of chiasmata at pachytene. No numerical data were given. He reported a tendency towards differential separation of the bivalents, the rods (terminal chiasmata) separating before the ring bivalents. A short interphase occurred, after which second division followed a normal course. Examination of 200 bivalents (25 PMC) showed a chiasma frequence per bivalent of only 1.34. Analysis of 120 tetrads showed 93.3 per cent normal. Cells containing five, six and eight daughter cells were also seen. A mean pollen grain diameter of 40 /i (500 grains per variety) was obtained i n an earlier study (Raptopoulos, 1940). These data were i n general agreement with Galletta (1959) who reported an average chiasma frequency per bivalent (6 clones) of 1.16 and pollen grain diameter of 39.5 fs. Galletta (1959) was the f i r s t investigator to report on prophase stages i n any detail, and commented on the d i f f i c u l t y of 10. interpretation. One to three nucleolar organizers, usually one, were observed suggesting that several or a l l chromosomes of the basic set were potential nucleolar organizers. Any small group of chromosomes isolated i n any one of the divisions appeared able to form a micro-nucleolus. The karyotype usually consisted of three long, three intermediate and two short bivalents. Due to the absence of highly contracted raetaphse I configurations, and the tendency for differential separation of the bivalents, the f i r s t division figures appeared as metaphase I - anaphase I configurations. Only one PMC with two uni-valents was seen. Possible univalents on, or i n the region of the metaphase plate, were interpreted as being due to one or two bivalents which showed a tendency to undergo precocious disjunction. Meiotic abnormalities seen included laggard chromosomes at either division, segregation of an entire bivalent to one pole (non-disjunction), s p l i t metaphase plates, multiple spindles, syndiploidy (unreduced gametes), failure of or unequal second division, multipolar arrangement of chromosomes on metaphase I plate and possible misalignment of second division plates. These observations were based on only 30 PMC per variety. His investigation of 130 sporads (tetrads) for each of three varieties (Bing, Napoleon and Ord) and six seedlings resulted i n a modal class of four spores per sporad in a l l cases. Sporads containing two to ten spores were seen. Good agreement was obtained between per cent normal sporads (four spores per PMC) and per cent 11. stainable pollen. A study of pollen shape by Galletta (1959), revealed triangular, circular, oval and rectangular grains. He considered that the proportion of the different grains was sufficiently distinct between clones to serve as a method of clonal identification. Within any group (sweet, sour and Duke cherries), the more f e r t i l e forms had the greater number of triangular grains. Lenander*s (1962) investigations are an outstanding con-tribution to an understanding of meiosis in sweet cherry. His description of the meiotic cycle i s similar to that reported by Raptopoulos (1941) and Galletta (1959) but Lenander observed the pre-sence of univalents and recognized their importance. Investigation of 500 to 2,000 PMC for each of 17 different varieties over a two year period showed that the varieties could be divided into two major groups: (1) a stable group of varieties showing a low frequency of univalents (10 per cent or less) and high pollen stainability (80 per cent or greater), and (2) a labile group with high univalent frequency (up to 66 per cent) and low pollen stainability (50 to 66 per cent). Univalents were seen in both meiotic divisions but were observed mostly i n the f i r s t division. The most common univalent frequency was two, but three or four and rarely five or six were also observed. Dividing univalents were also seen. It was suggested that univalent formation and irregular disjunction could lead to the formation 12. of micronuclei. A few cases of aneuploid pollen grains were reported. Reports on meiotic studies rarely comment on the synchrony of division. Zeuthen (1964) pointed out that an anther at the right stage of division frequently contained a high proportion of cells at the same stage of meiosis, that is highly synchronous. A high degree of synchrony of division has been shown to exist in Trillium  erectum (Sparrow and Sparrow, 1949) and Lolium longiflorum anthers (Erickson, 1948). Sparrow, Hoses and Steele, (1952) point out that such information can be of interest when investigating radiation effects. Galletta (1959) reported that meiotic stages usually ranged from prophase to early tetrads in dividing anthers in a l l the material he studied, thus suggesting a lack of synchrony in meiosis. 13. MATERIALS AND METHODS A. MATERIALS I. Pollen Study Cherry blossoms at the balloon stage of development were obtained on April 27th, 1964 from three six-year-old Lambert trees at the Research Station, Summerland, B. C. The selected trees had the Summerland Code designations, 20-13, 20-22, and 20-23. A l l trees were i n a single North-South orchard row; numbers 20-22 and 20-23 were adjacent trees. Blossom samples were coded from I to X and divided into sub-samples (Table 1). Aceto-carmine was used to estimate pollen stainability. II. Cytological Study Twigs from several Lambert trees were harvested at the Research Station, Summerland, in late February of 1964, and stored with moist paper towels i n sealed polythene bags at -3° to -5°C until required for study. B. METHODS I. Pollen Study 1. Pollen Stainability An estimate of pollen v i a b i l i t y was obtained by examination Table 1. Experimental materials used i n the pollen sampling study Experiment Tree Sample Sample No. of Blossoms/ Total No. of 500-No. No. Location Code No. Sub-samples Sub-sample blossoms grain counts 1. 20-23 Random I-A 10 1 10 2 Variation among Random 10 10 sub-sample I-B 1 1 sizes 25 Random II 5 5 2 Random III 5 10 50 2 2. 20-22 South, inside IV 5 5 25 2 Variation within a tree South, outside V 5 5 25 2 North, inside VI 5 5 25 2 North, outside VII 5 5 25 2 3. 20-13 Random, Variation within branch 1 VIII 15 1 15 1 a tree Random, branch 2 IX 15 1 15 2 4. 20-23 Random II 5 5 25 2 Variation between trees 20-22 Random X 5 5 25 2 15. of shape, size, and stainability of pollen grains using aceto-carmine. The rate of pollen grain expansion after the addition of aceto-carmine, subsequent changes i n shape or stainability after prolonged staining, and the different shapes and sizes of pollen grains were investigated. Since subsequent analysis of the data obtained showed a high degree of precision (repeatibility), the technique used i s given i n detail. Blossoms with the petals removed were allowed to dry at room temperature (about 22°C) in shell vials or petri dishes. When dry, the containers were corked and the blossoms stored at room temperature t i l l examined (eight months). Prior to pollen extraction, the blossoms were examined under a dissecting microscope and signs of frost injury (damaged p i s t i l s ) recorded. Blossoms were emasculated and the anthers placed i n a small shell v i a l . Pollen was extracted by shaking the v i a l 10 times. Anther debris was removed by inverting the v i a l , leaving the extracted pollen adhering to the bottom and sides. A pollen sample was taken from the v i a l using a dissecting needle, stirred into two drops of stain on a slide and a coverslip added. Both the bottom and sides of the v i a l were sampled. Care was taken to avoid movement of the coverslip or entrapment of a i r to prevent the tendency of light, aborted grains congregating around the edges of such areas. The slide was examined at a magnification of X 100 using an ocular grid in one eye-piece. In counting, the slide was moved across the f i e l d of view 16. and every pollen grain passing within a designated area of the grid was counted. The results were recorded every 100 grains u n t i l a total of 500 pollen grains had been counted. When making duplicate counts of any sample, two separate slides were made. 2. Sampling of Blossoms and Pollen A series of experiments was designed to investigate: (a) precision of pollen grain count, (b) the effect of subsample size on the accuracy of estimated normal pollen, (c) variation in pollen stainability within a tree and (d) variation i n pollen stainr a b i l i t y between trees. (a) Precision of pollen grain count Using the approximate proportion of pollen stainability, the standard error was used to estimate the number of grains required for counting to give a 100 (1 -<£•) per cent confidence interval of P t 5 per cent (Steel and Torrie, 1960). A pilot sample of 200 grains was counted in which 60 per cent of the grains were stained. It was calculated that 369 and 637 pollen grains should be counted for the 95 and 99 per cent confidence intervals respectively. In view of the large number of samples to be examined, i t was decided to accept the 95 per cent confidence interval. A count of 500 grains was therefore more than adequate and was used for a l l further work. 17. For each of the nine samples i n which duplicate 500-grain counts were made (Table 1), the precision of counting was investigated (1) by testing the equality of variances (F test) for the f i r s t 500-and second 500-grain counts from the sub-samples, and (2) by comparing the f i n a l mean for the f i r s t count with that for the second count (Student 1s t test). (Steel and Torrie, 1960). (b) The effect of sub-sample size Experiment 1 Tree 20-23 was the source of material for this experiment. Sub-samples of single-, five-, and 10-blossoms were investigated. A random sample of 20 blossoms (I) at the balloon stage was harvested for the single blossom sub-samples. Subsequently, the petals were removed, each blossom placed i n a small shell v i a l , l e f t at room temperature u n t i l dry and then the v i a l corked. For estimation of pollen stainability the 20 vials were randomly divided into two equal lots designated I-A and I-B, thus each lot contained 10 sub-samples. The procedure for five-blossom sub-samples (II) was the same as for the previous experiment except that 25 blossoms were used to make up the i n i t i a l sample, which was subsequently divided into five five-blossom sub-samples. The procedure for the 10-blossom sub-samples (III) used 50 blossoms which were sub-divided into five 10-blossom sub-samples. 18. Duplicate 500-grain counts were made for a l l sub-samples except for samples I-B and VIII for which single 500-grain counts were made. (c) Variation i n pollen stainability within a tree Since cherry trees are vegetatively propagated, genetic differences within or among trees were not expected. The following experiments were designed to investigate whether significant environ-mental variation did occur. Experiment 2 This experiment was designed to investigate: (1) variation; between a North and South branch on tree 20-22, (2) variation between two locations for each branch, and (3) variation among samples for each location, designated samples IV to VII (Table 1). Each branch was divided into two main locations. The f i r s t location, designated "inside", consisted of the main branch and i t s blossom spurs, below secondary branching. The remainder of the branch was designated as "outside." A 25-blossom random sample was harvested from each location for both branches and divided into five-blossom sub-samples. A 25-blossom random sample was harvested from the complete tree prior to sampling the North and South branches. Only a single blossom was collected from any one spur. Duplicate 500-grain counts were made for a l l sub-samples. 19. Experiment 3 This experiment was designed to investigate: (1) variation among buds (spurs), and (2) variation within buds (blossom). No attempt was made to investigate possible variation within a blossom. Two branches were selected at random from tree 20-13. Buds containing three blossoms were randomly selected from five d i f -ferent spurs for each branch, thus providing 15 sub-samples of single blossoms from each branch. (d) Variation in pollen stainability between trees Experiment 4 The five five-blossom sub-samples selected from tree 20-22 (X) and from tree 20-23 (II) were used to investigate stainability of pollen from different trees. The material used, sample and sub-sample size, and the number of counts for a l l pollen stainability experiments are sum-marized i n Table 1. (e) Statistical analysis The following general model was assumed to describe the sources of variance i n the 14 different analyses of variance calculated: Yijkm " ? + a i + b i j + C i j k + dijkm 20. Table 2. The expected mean squares of the analyses of variance Source of variation df Expected mean squares (n-1) n(p-l) np(q-l) npq(r-l) <5 + r o i + qrcr + pqr 0~ -S- rcr + qrcr d c ^ b 2 2 cr + rcr d c O 2 d Table 3. Sources of variation for statistical analyses using the expected mean squares in Table 2 Sources of variation Experiment No. A B C D 1 Samples Sub-samples Counts 2 Branches Locations Sub-samples Counts 3 Branches Buds Sub-samples 3 Buds Sub-samples Counts 4 Trees Sub-samples Counts Any sample with duplicate 500-grain counts Sub-samples Counts 22. Where the main effects A : a ( i = l,2,...,n), B : b (j = l,2,...,p), C : c (k = 1,2,....,q) and D : d (m = l, 2 , . . . . , r ) , were a l l assumed to be random effects. The expected mean squares, based on Eisenhart*s random model (No. 2) (Eisenhart, 1947), are shown in Table 2. The specific sources of variation are shown in Table 3. II. Cytological Study 1. Forcing Temperature The appropriate stage of meiosis was not available i n the budwood at the time of harvest, therefore buds had to be forced a r t i f i c i a l l y to obtain material for examination of meiosis. Experiments were done to determine the most useful temperature for forcing in order to eliminate any of the abnormalities which can result from high temperatures. Twigs were removed from storage, kept at 6° to 8°C for 48 hours to precondition the twigs prior to forcing, and then forced at 10°, 15°, 20°, and 25°C, the ends being freshly cut under water at the start of the forcing period. Only twigs with no deterioration of buds were used, and repetitive sampling was done every 12 hours u n t i l no buds remained. For each twig, the sample included a l l blossoms from five buds. The one exception was twigs forced at 10°C 23. which were sampled every 24 hours. Samples were stored separately i n fixative. 2. Staining Technique Walker*s (1957) technique was used for "quick checking" anther development. Two slight modifications were introduced: (a) omission of maceration in concentrate hydrochloric acid gave better staining, and (b) Swaminathan*s et ail (1954) propiono-carmine stain was used i n place of aceto-carmine. When this technique showed that anthers contained meiotic divisions, blossoms were harvested for detailed examination. A slightly modified Snow's (1963) alcoholic hydrochloric acid-carmine stain solution proved most satisfactory for detailed examination of meiosis. Maximum contrast and staining were obtained by adjusting the stain solution to a pH of 2.25 to 2.45. Blossoms were fixed i n Carnoy's 6:3:1 fixative (Darlington and La Cour, 1960) with the modification of the one part of glacial acetic acid being replaced by a saturated solution of f e r r i c acetate i n glacial acetic acid. Blossoms were stored i n the fixative i n a refrigerator* Using the modified Snow's (1963) solution, the following procedure was employed. Blossoms were removed from the refrigerator, the fixative discarded, washed twice i n 70 per cent alcohol for one hour per wash, 24. and stained overnight in alcoholic hydrochloric acid-carmine at 60°C. After staining was completed, the stain was poured off, saved for further use, the material washed with 70 per cent alcohol to remove excess stain and then stored in the f i n a l wash. When examining individual blossoms, a rough squash of five anthers was made to ascertain whether the meiotic stages present were suitable for further study. If so, a further five anthers were selected and a fresh slide made for detailed study. The small size of the anthers necessitated a l l squash preparations being done under a dissecting microscope. The "rough" squash procedure was as follows. A single blossom was removed from the v i a l and placed i n several drops of 70 per cent alcohol on a dissecting slide. Several anthers were obtained, five transferred to two drops of 45 per cent acetic acid on a glass slide, a coverslip added, and the anthers squashed. If the anthers failed to squash readily, the slide was gently heated and further pressure applied. The slide was examined for meiotic stages present and degree of division synchrony. If the rough squash revealed three of the five anthers contained meiotic stages from diplotene onwards, a fresh slide for detailed examination was then made as follows. A further five anthers were selected and placed i n a small drop of 45 per cent acetic acid on a clean glass slide. Using a tungsten needle, most of the liquid was drawn away from the anthers which were then broken-up using a spear-point needle. Some of the acetic acid held by the tungsten needle was 25. allowed to flow back over the broken-up anthers, floating the released PMC away from the anther debris. This liquid containing the PMC was drawn away from the anther debris and the procedure repeated several times u n t i l i t was considered that most of the released PMC had been floated off. The anther debris was then removed from the slide. A further drop of 45 per cent acetic acid was added to that already on the slide, a coverslip added, and slight pressure applied between the pages of a book of bibulous paper to flatten the PMC and remove excess liquid. The slide was then examined under the microscope and, i f required, further pressure applied to increase flattening of the PMC and/or the slide heated slightly to decrease cytoplasmic staining. Temporary slides were sealed using a mixture of gum mastic and paraffin wax. This method of sealing kept slides i n satisfactory condition for two to six months, and stain intensity gradually decreased. 3. Meiotic analysis. Several hundred slides and thousands of PMC were examined. Analyses of meiotic stages were carried out under o i l immersion (X 1125) using a Leitz Ortholux microscope with a Leitz Orthomat for photo-micrography. Kodak High Contrast Copy film was used i n conjunction with a Leitz yellow-green glass f i l t e r . Film was developed i n Kodak D-8 Developer for three minutes at 20°C and printed on Kodabromide paper. 26. RESULTS I. Pollen Study 1. Pollen Stainability I n i t i a l examination of several slides indicated that the pollen grains attained their characteristic shape almost immediately (within 5 seconds) after the addition of the stain. No further change in shape was apparent after several hours. Staining was very rapid and contrast persisted although stain intensity increased to some degree with time. The i n t i a l examination showed four apparent types of grains, triangular, oval, round and empty (Fig. 1, 2). The f i r s t three types stained with aceto-carmine. However, by slight move-ment of the coverslip, the large, stained oval grains were shown to be triangular (Fig. 3, 4). Thus, only three types of pollen grain were present i n the variety Lambert. These types can be seen in Fig. 2, and are described as: (a) Normal. Triangular grains, 40-47 ja in diameter, and stained. (b) Round. Intermediate size, round or triangular stained grains, appreciably smaller than the triangular grains of type (a). (c) Empty. Small, oval to round unstained grains. Only the 27. PLATE I . 1-4. P o l l e n g r a i n s o f Prunus avium L. c v . Lambert s t a i n e d w i t h a c e t o - c a r m i n e . F i g . 1. P o l l e n g r a i n sample i l l u s t r a t i n g t r i -a n g u l a r , o v a l , and a b o r t e d p o l l e n g r a i n s . F i g . 2. Normal ( a ) and abnormal (b and c ) p o l l e n g r a i n s . Types ( b ) and ( c ) a r e c o n s i d e r e d t o be n o n - v i a b l e . F i g . 3, 4. A s i n g l e group o f p o l l e n g r a i n s a f t e r s l i g h t l a t e r a l movement o f t h e c o v e r s l i p . The a p p a r e n t l y o v a l g r a i n was i n f a c t t r i a n g u l a r . A l l X 160. 28. large, triangular grains of the f i r s t category (type a) were con-sidered normal. The diameter of 200 grains of each major category (normal and abnormal) was measured. The diameter was considered to be the widest measurement obtainable for any grain. The normal grains ranged from 38.5 to 51.2 p i n diameter with 47 per cent in the modal class of 45.8 p. The abnormal ranged from 19.7 to 36.4 u with 32.5 per cent i n the modal class of 33.3 p. Blossom examination revealed some frost injury in 25 blossoms. Average pollen stainability of these blossoms was within 0.21 per cent of the f i n a l average stainability for a l l experiments. The range was 7.6 per cent and similar to that for undamaged blossoms. 2. Sampling of Blossoms and Pollen (a) Precision of pollen grain count The estimated mean percentage normal pollen obtained for the f i r s t and second 500-grain counts for any sub-sample showed l i t t l e variation. This observation applied to a l l experiments, regardless of the number of blossoms in the sub-sample. Typical data obtained are shown for single-blossom sub-samples (I-A) and for five-blossom sub-samples (II) in Table 4. Irrespective of sub-sample size, no significant differences Table 4. Comparison of percentage normal pollen obtained from duplicate 500-grain counts"for each of single-blossom and five-blossom sub-samples Sample Code Blossoms/ 500-grain No. sub-sample count no. Percentage normal pollen I-A 1 1 50.8 55.5 53.0 52.4 54.8 58.6 60.6 57.2 55.8 55.0 2 55.0 57.0 48.4 55.6 55.4 59.0 59.8 55.8 54.4 55.2 II 5 1 54.0 59.6 57.0 56.2 55.8 2 55.6 61.2 58.0 55.4 57.2 30. Table 5. Comparison of the variances and means of each of the nine samples with duplicate 500-grain pollen counts Sample Blossoms/ Calculated df Calculated df Code No. sub-sample F t I-A 1 1.102 9 and 9 0.134 18 II 5 1.321 4 and 4 0.976 8 III 10 1.158 4 and 4 0.813 8 IV 5 1.238 4 and 4 0.433 8 V 5 1.742 4 and 4 1.751 8 VI 5 2.189 4 and 4 0.689 8 VII 5 5.478 4 and 4 0.496 8 LX 1 1.296 14 and 14 0.264 28 X 5 1.208 4 and 4 0.157 8 I 31. were detected between the variances or f i n a l means of the f i r s t and second 500-grain counts for any of the nine samples on which duplicate 500-grain counts were made (Table 5). The lack of s i g n i f i -cant variation suggested a high degree of precision repeatability for pollen grain counts. (b) The effect of sub-sample size Experiment 1 Estimated percentage normal pollen for the single-blossom, five-blossom and 10-blossom sub-samples are shown in Table 6. There were no significant differences between sample means for the single-, five- or 10-blossom sub-samples with duplicate 500-grain counts. A further analysis revealed highly significant differences among single-blossom sub-samples, and significant differences among five-blossom and 10-blossom sub-samples. There was a significant difference between the extreme sub-sample values i n each sub-sample group (Duncan*s multiple range test). Differences were greatest for the single-blossom sub-samples. Comparison of the sample mean for each of the two groups of single-blossom sub-samples (I-A, duplicate counts, I-B single count) showed a difference of 0.41 per cent (Table 6) and suggested that single 500-grain counts gave as accurate an estimate of mean percentage normal pollen as did duplicate 500-grain counts. Table 6. Effect of single-, five-, and 10-blossom sub-sample size on estimated percentage normal pollen No. of Sample No. of Blossom/ Total 500-grain 1 Code No. Sub-samples Sub-sample blossoms pollens Percentage normal pollen of the sub-samples I-A 10 1 10 2 50.7 52.9 54.0 55.1 55.1 55.1 56.3 56.5 58.8 60.2 a ab ab be be be bed bed cd d I-B 10 1 10 1 51.6 53.6 55.0 55.8 56.0 56.2 56.6 57.0 58.2 58.8 II 5 5 25 2 54.8 55.8 56.5 57.5 60.4 a ab ab b c III 5 10 50 2 53.2 54.9 57.6 59.5 60.5 a ab be be c 1. Means of sub-samples followed by different letters are significantly different at the 5 per cent level.(Duncan 1s multiple range test). 2. S i g n i f i c a n t differences at the 5 per cent level. **Significant differences at the 1 per cent level. Table 6. (Continued) F-test Sample Standard among deviation sub-samples2 mean 55.47 2.804 5.317** 55.88 2.112 57.00 2.142 10.52* 57.14 3.066 7.266' 34. Increasing from single- to five-blossom sub-sample (number of blossoms) size decreased the range of means and standard deviation, and this trend was expected. However, the increase to 10-blossom size resulted in the highest standard deviation. Since single 500-grain counts proved highly accurate in estimating percentage normal pollen, the accuracy of pollen grain counts of different size was investigated for the single-blossom (Table 7) and five-blossom sub-samples (Table 8). Subsequent variation in percentage normal pollen, after 200 grains + had been counted, never exceeded - 5 per cent for any single-blossom sub-sample, and after 400 grains had been counted, never exceeded t 2 per cent (Table 7). The mean percentage normal pollen for five such sub-samples < < # > 5 and X(y.... ,IQ) did not change by more than 1.0 per cent from the f i n a l mean of the two groups of five blossoms after 400 grains per sub-sample were counted. A 200-grain count of a l l 10 single-blossom sub-samples resulted i n the mean percentage normal pollen (Xlj.,,.,10) being within * 0.5 per cent of the f i n a l mean of 55.47 per cent based on examination of 10,000 pollen grains. Similar results were obtained with the 5-blossom sub-samples for 400-grain counts (Table 8). Slightly greater variation was observed for this sub-sample size,, after 200 grains had been counted, than in the case of the single-blossom sub-samples. Accuracy of the estimated mean percentage normal pollen increased as the number of sub-samples was increased (Table 8). Table 7. Effect of the number of grains used on the estimate of normal pollen. Single-blossom sub-samples Percentage Normal Pollen Sub-sample Number of Pollen Grains Counted No. 100 200 300 400 500 600 700 800 900 1000 1 55.0 50.0 52.3 51.8 50.8 53.3 53.7 54.0 53.7 52.9 2 53.0 55.0 55.7 55.3 55.6 55.3 54.6 55.7 55.9 56.3 3 45.0 48.0 51.3 52.5 53.0 52.2 51.0 50.3 51.1 50.7 4 50.0 50.5 52.3 53.0 52.4 53.8 54.3 54.0 54.0 54.0 5 61.0 58.0 55.7 55.8 54.8 54.5 53.9 53.0 54.6 55.1 6 69.0 62.0 59.3 58.5 58.6 59.3 58.9 58.6 58.6 58.8 7 59.0 61.0 59.0 59.5 60.6 58.3 58.4 59.6 60.3 60.2 8 71.0 60.0 59.0 58.3 57.2 57.5 56.7 57.0 56.8 56.5 9 59.0 58.0 55.7 55.0 55.8 55.0 55.3 55.4 55.2 55.1 10 58.0 56.0 59.0 56.5 55.0 53.7 54.6 55.0 54.9 55.1 XV S 52.80 52.30 53.50 53.65 53.32 53.83 53.49 53.40 53.84 53.80 ^6,....,10 6 3 , 2 0 5 9 , 4 0 5 8 , 4 0 5 7 , 5 5 5 7 , 4 4 5 6 , 7 7 5 6 , 7 7 5 7 , 1 3 5 7 , 1 6 5 7 , 1 4 XT 1 0 58.00 55.85 55.93 55.60 55.38 55.30 55.13 55.26 55.50 55.47 Table 8. Effect of the number of grains used on the estimate of percentage normal pollen. Five-blossom sub-samples Percentage Normal Pollen Sub-sample No. 100 200 Number of 300 400 Pollen Grains Counted 500 600 700 800 900 1000 1 46.0 50.5 56.0 53.8 54.0 54.5 54.4 55.1 55.3 54.8 2 68.0 61.5 61.3 61.0 59.6 60.7 60.7 61.0 59.9 60.4 3 55.0 56.5 57.0 57.3 57.0 57.3 57.3 57.1 57.0 57.5 4 67.0 63.0 59.0 54.5 56.2 59.0 56.6 56.1 56.3 55.8 5 69.0 60.0 58.3 56.8 55.8 57.3 57.1 57.5 56.7 56.5 x l , 2 57.00 56.00 58.67 57.37 56.80 57.58 57.57 58.06 57.61 57.60 Xi 3 56.33 56.16 58.11 57.33 56.87 57.50 57.48 57.75 57.41 57.57 Xi 4 59.00 57.88 58.33 56.63 56.70 57.88 57.25 57.35 57.14 57.13 X, 5 61.00 58.30 58.33 56.65 56.52 57.77 57.23 57.38 57.04 57.00 37. (c) Variation i n pollen stainability within a tree Experiment 2. Variation between and within a North and South branch of tree 20-22 The data for this experiment are shown in Table 9. The estimated percentages of normal pollen for the sub-samples of the South branch (samples IV and V) were consistently higher than those for the North branch (samples VI and VII). The f i n a l means differed by 3.09 per cent. Significant differences among sub-samples were detected only for North, inside (sample VI). Analysis of the data (Table 10) for the North and South branch showed highly significant differences between branches, and among sub-samples within locations. No significant differences were detected between the two locations (inside and outside) within a branch. Experiment 3. Variation between and within two randomly selected branches from tree 20-13. The data for the 15 single-blossom sub-samples for each branch are shown i n Table 11. The f i n a l estimate of percentage normal pollen for the two branches differed by only 0.1 per cent. As i n Experiment 1, the single 500-grain count appeared to have a high degree of accuracy. Highly significant differences were Table 9. Effect of a North and South branch from a single tree on estimated percentage normal pollen Sample Percentage normal pollen of Sample Standard F test among Code No. Branch Location the five-blossom sub-samples^ Mean Deviation sub-sample^ IV South Inside 57.9 60.0 60.3 61.3 61.8 60.36 VI VII South Outside 58.6 59.9 60.5 61.5 61.7 60.44 North Inside 55.2 56.0 57.9 58.2 60.1 57.48 a ab be be c North Outside 55.4 56.2 56.8 56.9 60.4 57.14 1.526 1.265 1.934 1.918 3.622 2.698 7.475* 3.921 1. Means of sub-samples followed by different letters are significantly different at the 5 per cent level (Duncan's multiple range test). 2. Significant differences at the 5 per cent level. 39. Table 10. Analysis of variance of five-blossom sub-samples from two locations within a North and South branch Source df MS Branches 95.551 353.89 Locations/ branches 0.270 0.048 Sub-samples/ locations branches 16 5.673 4.308 Counts/sub-samples/ locations/ branches 20 1.317 Total 39 1. **Significant differences at the 1 per cent level. Table 11. Effect of two randomly selected branches from a single tree on estimated percentage normal pollen Sample Branch Code No. No. Percentage normal pollen of the single blossom sub-samples VIII 1 52.6 54.0 54.8 54.8 57.0 57.0 57.6 57.6 59.4 60.0 60.4 60.6 60.8 63.6 64.2 IX 2 54.2 54.5 55.6 55.7 55.7 56.0 56.1 57.7 58.9 60.3 60.8 61.3 61.9 62.3 64.9 a a ab ab ab ab b be cd de de de e e f 1. Means of sub-samples followed by different letters are significantly different at the 5 per cent level (Duncan's multiple range test). 2. Significant differences at the 1 per cent level. Table 11. (Continued) F-test Sample Standard among ^ mean deviation sub-samples 58.29 3.410 J> 58.39 2.714 23.57** 42. detected among sub-samples within sample IX. Analysis of the data (Table 12) for both branches did not reveal any significant differencesbetween branches nor among buds within either branch. Because three-blossom buds had been randomly selected from five different spurs of each branch at time of harvest, a further analysis was done for data from sample IX with duplicate 500-grain counts (Table 13). No significant differences were detected among buds (spurs), but highly significant differences were present among sub-samples (blossoms) within a bud. (d) Variation in pollen stainability between trees Experiment 4 The data (Table 14) from this experiment are from the random samples from each of trees 20-22 and 20-23. Estimated percentage normal pollen for the samples differed only by 0.5 per cent between the two trees. Significant differences were detected among the five-blossom sub-samples for each tree. In a further analysis (Table 15), no significant dif-ferences were detected between trees but there were highly significant differences among sub-samples. Combined data for a l l experiments showed an estimated percent normal pollen of 57.74 per cent and standard deviations of sample means of * 1.58, based on 75,000 pollen grains. 43. Table 12. Analysis of variance of five three-blossom buds randomly selected from each of two branches from a single tree Source df MS F Branches 1 0.53 0.087 Buds/branches 8 6.835 0.467 Blossoms/buds/ branches 20 14.631 Total 29 Table 13. Analysis of variance of five three-blossom buds randomly selected from a single branch Source df MS F Buds 4 4.1175 0.1407 Blossoms/buds 10 29.263 23.574 Counts/ blossoms/buds 15 1.2413 Total 29 1. Significant differences at the 1 per cent level. Table 14. Effect of two different trees (adjacent) on estimated percentage normal pollen Percentage normal pollen^of Sample Standard F test amonj Sample Code No. Tree No. five-blossom sub-samples Mean Deviation sub-samples' II 20-23 54.8 55.8 56.5 57.5 60.4 57.00 2.142 10.52 a ab ab b c X 20-22 55.3 55.9 58.0 58.3 60.0 57.50 1.908 10.883* a ab be c c 1. Means of sub-samples followed by different letters are significantly different at the 5 per cent level (Duncan*s multiple range test). 2. Significant differences at the 5 per cent level. 45. Table 15. Analysis of variance of randomly selected samples from two different trees Source df MS F 1 Trees 1 Sub-samples/trees 8 Counts/sub-samples/ trees 10 Total 19 1. **Significant differences at the 1 per cent level. 1.25 0.152 8.22 10.675 0.77 46. II. Cytological Study 1. Forcing Temperature The influence of forcing temperatures revealed the following effects on meiosis. Examination of several squashes of material forced at 10°C showed that meiotic divisions were not present after 96 hours at this temperature. Further forcing resulted i n visual signs of blossom and anther deterioration in some buds as evidenced by general discoloration. Buds forced at 15°C showed meiotic divisions after 36 to 72 hours. Divisions seemed normal, the only apparent defect being the presence of one pair of univalents. Material forced at 20° and 25°C showed meiotic abnormalities, especially at the higher temperature. At 20°C a second nucleolus was frequently observed as early as leptotene (Fig. 5) and the chromosomes appeared more diffuse than normal. In metaphase I -anaphase I laggards, dividing univalents and chromatin bridges (Fig. 6) were present. The latter defect was never seen at 15°C. The spindle mechanism also seemed to be affected because random groups of chromosomes were present in telophase I in some PMC (Fig. 7). Meiotic abnormalities restricted to material forced 47, PLATE II © © Fig. 5-7. Pollen mother cells from material forced at 20°C. Fig. 5. An additional nucleolus and decreased condensation of the chromatic material i n leptotene. Fig. 6. A chromatin bridge, laggard, and divided univalent i n anaphase I. Fig. 7. Random groups of chromosomes at telophase I. A l l X 1180. Fig. 8. A pollen mother c e l l from material forced at 25°C with an additional nucleolus and multiple association of chromosomes i n diplotene; X 1180. 48. Fig. 9-12. Pollen mother cells from material forced at 25°C. Fig. 9. Chromosome clumping and chromatin bridges in metaphse I. Fig. 10. Complete breakdown of anaphase II. Fig. 11. Random grouping of chromosomes and chromosome elimination in telophase II. Fig. 12. Formation of eight to nine micro-nuclei in late telophase II, prior to cytokinesis. A l l X 1180. 49. at 25°C included multiple association of chromosomes in diplotene (Fig. 8), clumping of chromosomes and chromatin bridges i n metaphase I (Fig. 9), complete breakdown of normal division and segregation in anaphase II (Fig. 10), and random groups of chromo-somes in early telphase II (Fig. 11) which resulted i n the formation of up to eight micronuclei i n late telophase II prior to cytokinesis (Fig. 12). It is suggested that such abnormalities could give r i s e to multi-spore sporads (tetrads). The foregoing observations resulted i n a l l material for meiotic analysis being forced at 15°C. , 2. Staining Technique Propiono-carmine proved extremely useful for "quick-check" techniques on meiotic stages i n anther development. A l l stages from pachytene to finished tetrads could be clearly dis-tinguished. The PMC in Fig. 13 and 14 were stained with propiono-carmine. _ The alcoholic hydrochloric acid-carmine gave excellent results, and was used for a l l detailed meiotic studies. Some of the specific advantages of this technique were: 1. The method was extremely simple, and gave easily repeatable and high quality results. 50. 2. The hot staining resulted i n a combined stain-maceration; thus the anthers could be easily squashed without further treatment. 3. Stained material could be kept in 70 per cent alcohol for several weeks without any apparent deterioration in subsequent results. The slight toughening that occurred from the alcohol storage could be overcome by a brief maceration i n 45 per cent acetic acid at 60°C, i f toughness became a problem. Slight destaining occurred during this maceration, but as the cytoplasm also destained, contrast remained good. 4. The anthers and contents were already stained, therefore dissection could be carried out in a clear liquid which greatly increased the ease of dissection of the small anthers. 5. When making a rough squash, the PMC were readily released and remained i n the immediate v i c i n i t y of the anther, thus permitting a general estimate of the synchrony of meiotic divisions in an anther. 6. The stained PMC could readily be separated from the anther debris when making slides for detailed examination. 3. Meiotic Analysis of Prunus avium cv. Lambert The results of the meiotic analysis were based largely on observations of the f i r s t meiotic division. There was extreme d i f f i c u l t y interpreting the second meiotic divison; therefore, only general observations could be made about this stage. 51. (a) Synchrony of meiotic division In the early stages of meiosis there appeared to be a high degree of synchrony, that i s , most PMC were at the same division stage. Synchrony was greater within an anther than among anthers i n a blos-som. Prior to pachytene, the PMC adhered to each other on squashing the anther (Fig. 13), and remained clumped together. The large, darkly staining nucleolus was readily seen, and examination at a higher mag-nification (Fig. 14) revealed that chromosomal organization had begun. When the PMC were at this stage, diplotene or later stages were never seen. From late pachytene onwards, the prominent hyaline callose PMC wall was present around each PMC, and cells were readily separated on squashing. The high degree of synchrony was present through diplotene (Fig. 15) and metaphase I (Fig. 16), but became less obvious towards the end of the f i r s t division. From late anaphase I onwards, the divisions became increasingly asynchronous. By the time tetrads were present, most meiotic stages could be found but the majority of PMC were in the second meiotic division. The gradual breakdown of division synchrony may have been due to the variation in time required for the different meiotic stages. I n i t i a l meiotic stages were markedly slow, as the sampling of many twigs forced at 15°C for up to 36 hours, revealed pachytene as the most advanced meiotic stage present, and further samples taken 12 hours later contained PMC in late telophase II. (b) Fir s t meiotic divison 52. PLATE IV Fig. 13-16. Synchrony of division in early meiotic stages. Fig. 13. The "grape stage" during early meiosis, the PMC tending to adhere to each other prior to formation of the callose special PMC wall; X 225. Fig. 14. The same stage at higher magnification. Prominent nucleoli are present, and chromosomal organization has commenced; X 445. Fig. 15. Early diplotene to diakinesis. The broken callose special PMC walls can be clearly seen; X 445. Fig. 16. Diakinesis to the start of f i r s t anaphase division; X 445. 53. Interpretation of the early meiotic stages was d i f f i c u l t due to the small size and complex structure of the chromosomes. The following observations were made. 1. Prophase stages A series of prophase stages showed the following features. A preleptotene PMC, illustrated i n Fig. 17, contained a prominent nucleolus, but no chromosomal organization was apparent. In leptotene (Fig. 18), the single stranded chromosomes were v i s i b l e . Their organization appeared to be associated with three to four hetero-pycnotic bodies, two of which were attached to the nucleolus. Synapsis of the homologous chromosomes occurred during zygotene (Fig. 19), and the synapsed chromosomes were clearly evident by pachytene. Throughout pachytene, a longitudinal contraction of the chromosomes occurred (Fig. 20 and 21) with a consequent increase i n c l a r i t y . By late pachytene, the callose wall surrounding each PMC was clearly evident. Only a single large nucleolus was usually present, but an additional one (Fig. 21), or two micronucleoli were also to be seen i n a few cases. These micronucleoli always appeared to be associated with a heteropycnotic body (Fig. 21). The double stranded nature of the pachytene chromosomes could be seen (Fig. 21). After pachytene, the heteropycnotic bodies were no longer evident. Examination of early diplotene stages revealed a low frequency of definite ring bivalents (one in Fig. 22, two i n Fig. 23). 54. PLATE V Fig. 17-20. Pre-meiotic to pachytene stages. Fig. 17. A pre-meiotic PMC containing a single nucleolus. Organization of the chromatic material is not yet apparent. Fig. 18. Leptotene. Start of chromosomal organization. The typical four heteropycnotic bodies are present, two of which are associated with the nucleolus. Fig. 19. Zygotene. Start of synapsis of the homologous chromo-somes. Fig. 20. Late pachytene. Due to longitudinal contraction the chromosomes have become clearly v i s i b l e . Heteropycnotic bodie are s t i l l present, three of which can be seen. A l l X 1180. 55. Fig. 21-23. Pachytene to diplotene. Fig. 21. Late pachytene. A second nucleolus i s present and associated with one of the heteropycnotic bodies. The double stranded nature of the chromo-somes can be seen. Fig. 22, 23. Diplotene il l u s t r a t i n g the low frequency of definite ring bivalents. A highly condensed chromosome is associated with the nucleolus i n both cases. Two bivalents are part i a l l y hidden by the nucleolus i n Fig. 22. A l l X 1180. 5 6 . PLATE VII Fig. 24-27. Mid-diplotene to diakinesis. Fig. 24. Seven bivalents and one pair of univalents are present. Several bivalents appear to be at maximum repulsion, the chromatin material joining the homologous chromosomes being only just v i s i b l e . Fig. 25. Late diplotene. Seven bivalents, one pa r t i a l l y obscured by the nucleolus, and two possible univalents. Fig. 26. Start of diakinesis; six bivalents highly contracted and two s t i l l contracting. Fig. 27. Diakinesis with seven bivalents and one pair of univalents. A l l X 1180. 57. There appeared to be a very pronounced repulsion between the homologous chromosomes in diplotene. The chromatin joining the condensed areas of the bivalents could barely be seen (Fig. 24). Univalents were* evident for the f i r s t time in this stage (Fig. 24). It seemed likely that they arose from desynapsis of a specific bivalent as more than two univalents were never found. No unpaired chromosomes were found in early diplotene, therefore, univalent origin due to asynapsis appeared unlikely. A gradual thickening and shortening of the bivalents occurred during diplotene (Fig. 24 and 25). By diakinesis (Fig. 26) the bivalents were highly contracted. Univalents, i f present, could be clearly seen as i s shown in Fig. 27. Towards the end of diakinesis only the remains of the prominent nucleolus were evident (Fig. 28). 2. Metaphase I - telophase I Due to the tendency towards differential separation, that i s sequential disjunction of the bivalents, a clearly defined f i r s t metaphase configuration was rarely seen. The nearest approach to this i s illustrated in Fig. 29, but even->here, the tendency for one of the bivalents to undergo early disjunction can be seen. Detailed examination of 227 PMC without univalents in metaphase I -anaphase I indicated a sequential pattern of disjunction which seemed to be associated with bivalent size and number of chiasmata. This pattern is illustrated i n Fig. 30. Apparently one small bivalent (a) underwent disjunction soon after the bivalents became oriented on 58 • PLATE VIII Fig. 28-31. Diakinesis to late anaphase. Fig. 28. Diakinesis with eight bivalents and the remains of the nucleolus. Fig. 29. Metaphase I. One of the bivalents has already started to under-go disjunction. Fig. 30. Metaphase-anaphase. Differential separation of the bivalents characteristic of this division stage. Five of the bivalents have undergone disjunction (a,b,c, d and e), and three are s t i l l on the metaphase plate. Fig. 31. Late anaphase with the last bivalent (h) undergoing disjunction after the other chromosomes have reached the poles. A l l X 1180. 59. the metaphase plate, and was rapidly followed by two more (b and c). By the time these had started to move towards the poles, two more bivalents (d and e) could be seen to have separated, leaving only three bivalents on the metaphase plate. Of these three, the largest bivalent (h) was characteristically last. By the time this bivalent had undergone disjunction, the other chromosomes were already at the poles (Fig. 31). This tendency towards sequential disjunction was.also found to occur when univalents were present (Fig. 32 and 33). No PMC were seen in which a l l bivalents underwent simultaneous disjunction. The three smaller bivalents (a, b and c) which were the f i r s t to undergo disjunction appeared to be rods with a single chiasma. Two univalents were present in some PMC in the f i r s t meiotic division, being f i r s t evident i n diplotene. It was f i r s t postulated that the univalents might assume either one of two positions during the metaphase I - anaphase I division, namely; one on each side of the metaphase plate (Fig. 32)> or both on the same side (Fig. 33). After inspection of several hundred PMC i t was evident that two further locations were possible. One univalent might be included with the seven bivalents on the metaphase plate (Fig. 34 and 39), or both univalents included (Fig. 35). The frequency of these four possible types is shown i n Table 16. If these four possible univalent positions were genuine, abnormal divisions showing (1) unequal 7:9 chromosome distribution 60 . 1 PLATE IX -( § ) ® . 32-35 . The f o u r observed u n i v a l e n t p o s i t i o n s d u r i n g the f i r s t metaphase-anaphase d i v i s i o n . F i g . 3 2 . One u n i v a l e n t at each p o l e . F i g . 33 . Two u n i v a l e n t s at the same p o l e . F i g . 34. One u n i v a l e n t at each p o l e , and the o t h e r on the metaphase p l a t e . F i g . 3 5 . Both u n i v a l e n t s on the metaphase p l a t e . A l l X 1180. 6 1 . PLATE X Fig. 36-38. Abnormal meiotic divisions due to univalents at earlier division stages. Fig. 36. A 7:9 telophase I as a result of two univalents at the same pole i n metaphase I. Fig. 37. A univalent dividing equationally late in anaphase as a result of one or two univalents on the metaphase plate. Fig. 38. Two univalents dividing equationally late in anaphase as a result of two uni-valents being on the metaphase plate. A l l X 1180. Fig. 39. A diakinetic chromosome that failed to become oriented on the metaphase plate. Two univalents are also present, one of which is included with the bivalents on the metaphase plate. X 1180. 62. i n telophase I (2) one or two univalents dividing equationally in the f i r s t meiotic division, and (3) laggards would be expected. A l l these categories were seen. Of the telophase I divisions seen without dividing univalents or laggards, (102), 24.5 per cent were abnormal, seven chromosomes being at one pole and nine at the other (Fig. 36). A single univalent on the metaphase plate was seen in 43 FMC and one univalent dividing equationally (Fig. 37) occurred in 21 PMC. Both univalents on the plate were seen less frequently (27 PMC), and only nine PMC were observed in which equational division of both univalents occurred (Fig. 38). In a l l cases observed, univalent division occurred after a l l the bivalents had undergone disjunction and the normal chromosomes were already congregated at the poles (Fig. 37 and 38). Ten PMC were seen i n which laggards (univalents) were present that had not yet divided. The position of the divided univalents at the end of anaphase I (Fig. 37 and 38) was such that i t appeared unlikely they would be eliminated at the end of the f i r s t division. However, i t appeared that chromosome elimination could occur at this stage. It is suggested that Fig. 44 illustrates a PMC in interkinesis where univalents were present. An unequal segregation of 6:9 has occurred, and one chromosome, possibly a univalent, i s not included in either of the chromosome groups. Resulting micronuclei could eventually be apparent as aborted pollen grains. Two further abnormalities were observed. In 18 PMC, one bivalent failed to become oriented on the metaphase plate (Fig. 39). 63. This was not considered to be an artifact due to squashing, since the PMC were not distorted and the other bivalents appeared undisturbed. These chromosomes were recorded as "Diakinetic" chromosomes, and were expected to result in abnormal divisions. The other abnormality involved unequal division in metaphase I - anaphase I when several of the bivalents had not yet undergone disjunction. It seemed likely that this type of division resulted from two univalents being present on one side of the metaphase plate prior to the start of disjunction, and a telophase distribution of 7:9 would develop. However, the actual presence of univalents could not be identified with sufficient cert-ainty to permit these divisions to be categorized as two univalents at one pole. Interkinesis (Fig. 40) appeared to be very brief and was rarely seen. There was no evidence of cytokinesis. (c) Second meiotic division Due to the d i f f i c u l t y of obtaining accurate chromosome counts, no numerical data were obtained for this division stage. However, the following observations were noted. The normal four groups of eight chromosomes (Fig. 41) were usually present. Several PMC were seen i n which the chromosome grouping was much more random, five or six groups being present. It was impossible to t e l l whether the scattering of the chromosomes was due to squash technique or was genetic in origin. It would appear li k e l y that the chromosome distribution of 7:7 + 9:9 shown in 64. PLATE XI Fig. 40-43. Interkinesis to telophase II. Fig. 40. Chromosome elimination at interkinesis. Fig. 41. Normal telophase II with tour groups of eight chromosomes; X 1180. Fig. 42, 43. A single PMC at different levels of magnification showing abnormal telo-phase II with chromosome groups of 7:7 and 9:9; both X 900. 65. PLATE XII Fig. 44. Abnormal PMC at late telophase II with two macro- and one micro-nuclei. Fig. 45. Abnormal sporad (tetrad) containing six spores, one of which contains two nucleoli. The broken PMC callose wall i s clearly evident; both X 1180. 66. Fig. 42 and 43 for a single PMC arose as a result of a 7:9 telophase I. The position of the chromosomes i n each of the four groups suggested that two monosomic and two trisomic pollen grains would be formed from such a PMC. A 7:9 telophase I division was assumed to produce inviable pollen. At the end of telophase II the individual chromosomes were no longer distinct, and prominent nucleoli were again present (Fig. 44). Cytokinesis i s detectable only after this stage has been completed (Fig. 45). Sporads (tetrads) containing from two to eight spores were seen. Some of the spores contained more than one nucleus (Fig. 45). No completely satisfactory method was found i n the present study for obtaining accurate counts of the number of spores per sporad The majority of the sporads examined were ruptured and one or more spores were missing. (d) Frequency of normal and abnormal f i r s t meiotic division Twenty-four separate categores were designated for recording the meiotic events seen, and divided among the appropriate meiotic stages from diakinesis to telophase i (Table 16). Twelve of the categories were considered to be abnormal. Three of the meiotic events were never seen; namely, four univalents in diakinesis, simultaneous disjunction of a l l bivalents, and 6:10 segregation of chromosomes in telophase I. Of the remaining 21 categories, nine were concerned directly with univalent occurrence 67. Table 16. Meiotic abnormalities present in the f i r s t meiotic division in anthers of Lambert sweet cherry Meitoic event No. of PMC Frequency 7. DIAKINESIS 8-II 1 7 TII + 2-1' 6-II + 4-1 METAPHASE I 18 29* __* 2.748 4.427 8-II 7-II+2-I 2-1 At the same pole 2-1 At opposite poles 1- 1 On the metaphase plate 2- 1 On the metaphase plate 9<L 58" 36 43* 27* 15.115 8.855 5.496 6.565 4.122 ANAPHASE I 1 Diakinetic chromosome 8-II undergoing disjunction 7-II + l - I I undergone disjunction 6-II + 2-II undergone disjunction 5-II + 3-II undergone disjunction 4-II + 4-II undergone disjunction 3-II + 5-II undergone disjunction 2-II + 6-II undergone disjunction l - I I + 7-II undergone disjunction Unequal division TELOPHASE I Normal 8:8 Abnormal 7:9 6:10 Laggards 1 univalent dividing 2 univalents dividing NORMAL MEIOTIC DIVISIONS ABNORMAL MEIOTIC DIVISIONS TOTAL PMC COUNTED 18" 42 37 20 14 6 3 6. 46 * 77 25* ~* I? 21 9* 357 298 655 2.748 6.412 5.649 3.053 2.137 0.916 0.458 0.916 7.023 11.756 3.817 1.527 3.206 1.374 54.504 45.496 100.0 1. Bivalents designated as II. 2. Univalents designated as I. Considered to be abnormal meiotic divisions. 68. in diakinesis or metaphase, and subsequent fate i n telophase. A tenth category, unequal division, was also probably associated with the presence of univalents. It is to be noted that 298 of the divisions in Table 16 were considered to be abnormal, a frequency of 45.49 per cent. Because only two univalents were present, a theoretical classification was designed to predict the probability of their assuming any one of nine possible locations in metaphase I (Table 17). Since no cytological distinction was possible between the two univalents, only four general positions were recognizable. The observed frequencies of these four positions are shown in Table 18. In only one case did the observed data f i t the frequencies expected on theoretical grounds (Chi-square test). This may have been due to possible factors which were not included i n the theoretical classification. One such factor was possible movement of the uni-valents to positions other than the i n i t i a l ones, due to flattening of the spherical PMC in slide preparation. Also, univalents are not governed to the same extent as the bivalents by the physiological forces that result in chromosomal orientation and alignment on the metaphase plate. Only three of the four general univalent positions might result in an recognizable atypical telophase I, since the PMC with one univalent at each pole would probably result in a normal 8:8 segregation. It was suggested that (1) two univalents at the same 6 9 . Table 17. Possible univalent locations and their theoretical frequencies per meiotic event i n f i r s t metaphase Univalent positions Univalent in metaphase I i n PMC cations Total possible Expected relative Pole 1 Plate Pole 2 locations frequency 2-1 at the same pole AB -7 AB 2 2/9 1-1 at opposite poles A B 2 2/9 B A 1-1 on the plate A B — 4 4/9 B A B A — A B 2-1 on the plate -- AB — 1 1/9 1. One univalent i s designated A, the other B. Table 18. Number of PMC pollen mother cells observed for each of the four possible univalent positions in f i r s t metaphase Probability Univalent position of event Observed Expected 1 2 Chi-square 2-1 at the same pole 2/9 58 36.5 12.66* 1-1 at each pole 2/9 36 36.5 0.007 1-1 on the plate 4/9 43 72.8 12.2* 2-1 on the plate 1/9 27 18.2 4.25 1. Expected frequencies and probabilities based upon Table 17. ic 2. Significant Chi-square at the 5 per cent level. 70. pole would result in a 7:9 telophase I, (2) one univalent on the plate and one at the pole i n seven 7 + 8 + the \ arising from equational division of the univalent on the plate, and (3) two univalents on the plate i n 7 + ^ + %: 7 + % + % i f both uni-valents divided equationally. The frequency of observed PMC with univalents in metaphase I, and related meiotic abnormalities i n telophase I are shown in Table 19. The data indicate a 2:1 frequency for the f i r s t two categories and a 3:1 for the third, and suggest that not a l l univalents present on the plate at f i r s t metaphase underwent equational division i n anaphase. These overall observations resulted i n a relative frequency of abnormal f i r s t meiotic divisions of 45.49 per cent (Table 16). 71. Table 19. Number of pollen mother cells containing univalents in f i r s t metaphase, and subsequent observed f i r s t telo-phase distributions with related abnormalities Univalent positions in metaphase I No. of PMC observed Chromosome distribution in telophase I No. of PMC observed 2-1 at the same pole 58 7:9 25 1-1 on the plate 43 7 + %: 8 + \ 21 2-1 on the plate 27 7 + % + %: 7 + % + \ 9 72. DISCUSSION I. Pollen Study A precise and valid method of determining the pollen v i a b i l i t y was desirable since i t was hoped to relate meiotic abnormalities with non-viable pollen. The conflicting opinions on stainability and in vitro methods have been pointed out. Since germination tests frequently lack precision, only marked variation from the normal can be detected. Furthermore, prolonged pollen storage which was unavoidable i n this study, frequently affects pollen germination thus reducing the v a l i d i t y of the test. Since Raptopoulos (1940) obtained a close agreement (it 5.6 per cent) between size and shape of pollen and percentage germination, and Galletta (1959) reported a general relationship between pollen shape and pollen f e r t i l i t y , the use of an aceto-carmine stainability test appeared j u s t i f i e d . The ease of categorizing the pollen and the low fre-quency of intermediate grains (type b) may account for (1) the high degree of precision obtained i n the pollen sampling study which resulted in subsequent detection of relatively minor differences i n pollen stainability within a tree, and (2) the accuracy obtained from single 500-grain counts. 73. The significant differences frequently detected among sub-samples may have prevented further differences being detected at a higher level such as "within branches". However, when even relatively minor differences were consistently present, as in the North and South branch in Experiment 2, highly significant differences were detected. Although s t a t i s t i c a l l y highly significant, the difference was only 3.09 per cent and i t is doubtful whether such a slight difference i s of biological or horticultural importance. Light distribution studies i n apple using uranyl oxalate actinometers (Heinicke, 1963) revealed l i t t l e variation between the average per-centage f u l l sun on the North and South sides for both a 17- and 35-year old tree. It would therefore seem unlikely that variation i n available carbohydrates at the time of blossom development was responsible for the significant differences found. No explanation is offered for the unexpected high range and standard deviation obtained for the 10-blossom sub-sample size in Experiment 1. A high degree of precision was obtained for these data, thus suggesting that pollen sampling from the v i a l was accurate. The data obtained suggested that marked variation was not present either within or between the trees studied. The lack of such variation was reflected i n the accuracy of estimated percentage normal pollen from single-blossom and five-blossom sub-samples after relatively few grains had been counted. These data suggest that a precise estimate of normal pollen can be obtained for six-year-old 74. Lambert trees by counting 400 grains of either five single-blossom sub-samples or two five-blossom sub-samples. The single-blossom sub-samples have the added advantage of indicating blossom to blossom variation. II. Cytological Study The necessity of forcing cherry buds to the appropriate stages for observations of meiosis made the choice of temperature in the laboratory an important consideration. The experiments showed that the use of temperatures above 15°C resulted i n meiotic abnormalities, and the severity and extent increased directly with temperature. These temperature effects were not surprising because abnormalities observed i n the genus Saxifraga (Beamish, 1961, 1964), in Triticum, Secale, Hordeum and Vicia species (Pao and L i , 1948), Dactylis glomerata (Dollinger, 1942), and Tradescantia (Sax, 1937) were attributed to high temperature. Lenander (1962) suggested that sweet cherry varieties could be divided, into two major groups; those in which univalents were common, the " l a b i l e " group, and those in which univalents were rare, the "stable" group. His data suggested that the labile group varied from year to year in univalent frequency and pollen f e r t i l i t y as a result of temperature differences. Such variation might be expected to occur in Lambert as this variety exhibits a relatively 7 5 . high univalent frequency; however, the present studies used material from one season only thus no "season effects" could be observed. However, the harvest at premeiotic stages, and the use of uniform forcing temperatures would eliminate the effects of diurnal temp-erature fluctuations. This removal of seasonal effects should result in more precise studies from year to year. Although no information is available, l i t t l e clonal variation is anticipated due to the vegetative method of propagation. The callose wall surrounding each PMC f i r s t became evident in mid-pachytene, and appeared to- break down after tetrad formation prior to maturation of the resulting pollen grains of Lambert. This development appeared similar to the pattern of synthesis of callose wall i n Cannabis sativa reported by Heslop-Harrison (1964). In Lambert PMC this wall is clearly evident when present and i t is suggested that the character can be used to make instantaneous decisions when searching for meiotic patterns between diplotene and tetrad formation. Meiosis in Lambert sweet cherry was characterized by the frequent presence of one pair of univalents. Many PMC were seen in which evidence of these univalents could be detected. Examination of meiotic stages prior to metaphase I suggested that the univalents arose as a result of desynapsis of one of the smaller bivalents rather than from asynapsis. Apart from the PMC containing univalents, meiotic behaviour was similar to that described for sweet cherry by Hruby 76. (1939), Raptopoulos (1941), Galletta (1959) and Lenander (1962). It would appear l i k e l y that the high percentage of aborted pollen grains seen in Lambert i s associated with the high frequency of PMC containing one pair of univalents, thus particular attention was paid to univalent behaviour. Univalent distribution at f i r s t metaphase did not appear to be random in Lambert. The most common univalent position was two univalents on one side of the metaphase plate; that i s , at the same pole. This distribution occurred in 58 (35.4 per cent) of those PMC seen in metaphase I containing one pair of univalents. It seemed l i k e l y that the category "unequal division" (Table 16) in metaphase-anaphase originated as two univalents on the same side of the metaphase plate. If the 46 PMC from this category are added to the above figure (58), the frequency of this univalent position is increased to 49.5 per cent as compared to a frequency of 17.1 per cent for one univalent at each pole, 20.5 per cent for one univalent in-cluded with the bivalents on the metaphase plate, and 12.9 per cent for both univalents on the plate. These unequal distributions of univalents suggest that these chromosomes are not randomly distributed in meiosis. The significance of univalents can be summarized from several sources. White (1961) and Burnham (1962) pointed out that unpaired chromosomes such as univalents, frequently become attached to the spindle somewhere between the equator and the pole. Univalents may pass to one pole without dividing at the f i r s t meiotic divison, and 77. then divide normally at the second; but there i s a strong tendency for them to lag at the f i r s t division, and lag at the second. In either case, the lagging chromosomes are usually not included in the nuclei resulting from meiosis, but appear as micro-nuclei in the quartet of spores. The behaviour of univalents during meiosis in Lambert sweet cherry agrees with the foregoing general pattern. A total of 70 PMC were observed in which one or both univalents were in-cluded with the bivalents on the metaphase plate. Subsequently, only 30 (42.9 per cent ) PMC were observed in late anaphase I in which univalents were dividing equationally. Presumably, not a l l univalents on the plate divide. Their position (Fig. 37, 38) was such that i t appeared unlikely that divided univalents would be eliminated prior to the second meiotic division. The present study indicated that univalents, which passed undivided to the poles in the f i r s t meiotic division, divided normally i n the second division. These divided univalents were included with the normal chromosomes at the respective poles (Fig. 42, 43). The frequency of such divisions was not known because the second division was so d i f f i c u l t to inter-pret. White (1960) and Burnham (1962) point out that elimination of lagging chromosomes is likely to occur in either division, resulting in micro-nuclei in the quartet of spores. These various meiotic abnormalities would lead to the production of a large number of aneuploid gametes. The v i a b i l i t y of such aneuploid gametes or zygotes would be expected to be low. Only one aneuploid cherry clone has been 76-reported (Galletta, 1959) which would tend to suggest that there i s a low v i a b i l i t y of the aneuploids under f i e l d conditions. It is evident that Lambert has a high degree of abnormal meiosis. The meiotic observations resulted i n an estimated abnormal f i r s t meiotic division frequency of 45.49 per cent (Table 16), and the stainability tests gave an estimated frequency of 42.26 per cent abnormal pollen grains. It is recognized that the v a l i d i t y of extra-polating from abnormal f i r s t meiotic divisions to f i n a l pollen abortion may be questionable, but for the materials and methods used i n these experiments there appears to be a close agreement between abnormal meiosis and aborted pollen. Both meiotic and pollen studies have been shown to be precise, and have provided the information on normal habit Lambert sweet cherry. Thus a basis has been established for future studies of irradiated clones of the Lambert variety. 79. SUMMARY AND CONCLUSIONS A study was made of the meiotic behaviour of Prunus  avium L. cv. Lambert, and the variation in pollen stainability between and within trees. I. Pollen Study 1. Examination of 75,000 pollen grains , stained with aceto-carmine, revealed that three major types of grains were present; namely, large, triangular, stainable; small, oval, and non-staining and intermediate sized, stainable grains. The f i r s t two types ac-counted for over 95% of the pollen grains seen. Only the large, t r i -angular, stainable grains were considered to have been viable at time of blossom harvest. 2. The method of counting proved to be very precise, and frequently resulted i n significant differences i n pollen stainability being detected among the single-, five-, and 10-blossom sub-samples. No significant variation was detected between trees, between branches, or within branches, except for a North and South branch in one tree tested. Highly significant differences were detected between blossoms within three-blossom buds. It is suggested that variation within buds was the major source of detectable variation in pollen stain-a b i l i t y tests, however, the significant differences were small and 80. did not appear great enough to be of any horticultural importance. 3. Analysis of the data of 11 experiments revealed only slight variation from an estimated mean percentage abnormal pollen of 42.26 per cent. II. Cytological Study 1. No meiotic abnormalities other than those associated with uni-valents were observed in material forced at 15°C. Abnormalities were induced by forcing temperatures of 20° and 25°C, frequency and severity increasing with temperature. 2. Synchrony of meiotic divisions within an anther was observed up to diakinesis, asynchrony gradually developing after this division stage. 3. The meiotic cycle was characterized by the frequent presence of one pair of univalents. It is suggested that they arose from de-synapsis of a specific bivalent in diplotene. More than two univalents were never seen. 4. There appeared to be non-random distribution of the univalents i n metaphase I, two univalents on one side of the metaphase plate being the most frequently observed position. Univalents included with the bivalents on the metaphase plate frequently divided equationally. 5. The bivalents underwent sequential disjunction in the metaphase-anaphase division. The order of disjunction was apparently 81. associated with size and shape of the chromosomes. 6. An abnormal f i r s t meiotic division frequency of 45.49 per cent was i n close agreement with an estimated 42.26 per cent abnormal pollen obtained from stainability tests. 62 LITERATURE CITED Anon., Orchard Survey of the Okanagan Valley. 1960. B r i t . Columbia Dept. Agr. Beamish, K. I. 1961. Studies of meiosis i n the genus Saxifraga of the Pacific Northwest. Can. Jour. Bot. 3 9 : 567-80. 1964. Personal communication. Becker, C. L. 1932. Studies of pollen germination in certain species and interspecific hybrids of Prunus. Proc. Amer. Soc. Hort. Sci. 29: 122-6. Belling, J . 1926. The iron-acetocarmine method of fixing and staining , chromosomes. B i o l . B u l l . 50: 160-2. Brewbaker, J . L. and Kwack, B. J . 1963. The essential role of calcium ion in pollen germ-ination and pollen tube growth. Amer. Jour. Bot. 50: 859-65. , and Majumder, S. K. 1961. Cultural studies of the pollen population effect and the self-compatibility inhibition. Amer. Jour. Bot. 48: 457-64. Burnham, C. R. 1962. Discussions in cytogenetics. Burgess Publishing Co., Minneapolis. Carnahat> J . L. and H i l l , H. D. 1955. Lolium perenne L. Tetraploid Festuca elatior L. t r i p l o i d hybrids and colchicine treatments for inducing autoallohexaploids. Agron. Jour. 47: 258-62. Crane, M. B. and Brown, A. G. 1937. Incompatibility and s t e r i l i t y i n the sweet cherry, Prunus avium, L. Jour. Pomol. Hort. Sci. JL5: 86-116. 83. Darlington, C. D. 1928. Studies i n Prunus. I and II. Jour. Genet. 19: 213-56. 1934. Studies i n Prunus. IV. Jour. Genet. 28: 327-8. , and LaCour, L. F. 1960. The handling of chromosomes. George Allen and Unwin Ltd., London. de Nettancourt, D., and Grant, W.F. 1963. The cytogentics of Lotus (Leguminosae). II. A diploid interspecific hybrid between L. tenuis and L. f i l i c a u l i s . Can. Jour. Genet. Cytol. 5: 338-47. Dollinger, E. J. 1942. Effect of heat treatment on meiotic irregularity in Dactylis glomerata. M.S. Thesis, The Pennsylvania State University. (Cited by H i l l and Carnahan, 1963). Eaton, G. W. 1961. Germination of sweet cherry (Prunus avium L.) pollen in vitro as influenced by fungicides. Can. Jour. P l . Sci. 41: 740-3. Eisenhart, C. 1947. The assumptions underlying the analysis of variance. Biometrics 3: 1-21. Erickson, R. 0. 1948. Cytological and growth correlations in the flower bud and anther of Lilium longiflorum. Amer. Jour. Bot. 35: 729-39. Galetta, G. J. 1959. Comparative breeding, morphological, and cytological studies of polypooid sweet cherries (Prunus avium L) PhD. Thesis, University.of California at Davis. Grun, P. 1952. Effect of environment upon chromosomal pairing of some species and hybrids of Poa. Amer. Jour. Bot. 39: 318-23. 84. Hauser. £. J. P. and Morrison, J. H. 1964. The cytochemical reduction of Nitro Blue Tetrazolium as an index of pollen v i a b i l i t y . Amer. Jour. Bot. 51: 748-52. Heinicke, D. J. 1963. The micro-climate of f r u i t trees. II. Foliage and light distribution patterns in apple trees. Proc. Amer. Soc. Hort. Sci. 83: 1-11. Heslop-Harrison, J. 1964. Cell walls, c e l l membranes, and protoplasmic connections during meiosis and pollen development, in Pollen  Physiology and Fer t i l i z a t i o n , H. F. Linskens, (Ed.), North Holland Publishing Co., Amsterdam. 1964. H i l l , D. H. and Carnahan, H. L. 1963. The evaluation of effects of two temperatures on micronuclei per quartet in Bromus inermis Leyss. Crop Sci . 3: 73-4. Hough, L. F. 1939. The pollen value of 134 apple varieties as determined by germination tests and f i e l d t r i a l s . Proc. Amer. Soc. Hort. Sci. 37: 133-40. Hruby, K. 1939. The cytology of the Duke cherries and their derivatives. Jour. Genet. 38: 125-31. Keep, Elizabeth. 1963. Personal communication. King, J. R. 1960. The peroxidase^ reaction as an indicator of pollen v i a b i l i t y . Stain Tech. 35: 225-7. Kobel, F. 1927. Zytologische untersuchungen an Prunoideen und Pomoideen. Arch. J u l . Klaus-Stift, 3_: 1-84. Lapins, K. 1963. Note on compact mutants of Lambert cherry produced by ionizing radiation. Can. Jour. PI. S c i . 43: 424-25. 1964. Personal communication. Lenander, S.E. 1962. Univalentbildning och Sankt p o l l e n f e r t i l i t e t hos Prunus avium. Jour. Royal Swed. Acad. Agric. Forest. 101: 350-83. 85. Marks, G. E. 1954. An aceto-carmine glycerol j e l l y for use in pollen f e r t i l i t y counts. Stain Techn. 28: 277-8. Morris, R. and Isikon, M. H. 1964. Cytological studies on inbred lines of maize. Can. Jour. Genet. Cytol. 6i 508-15. Pao, W. K. and L i , H. W. 1948. Desynapsis and other abnormalities induced by high temperatures. Jour. Genet. 48: 297-310 Raptopoulos, T. 1940. Pollen germination tests in cherries. Jour. Pomol. Hort. Sci. 18: 61-7. 1941. Chromosomes and f e r t i l i t y of cherries and their hybrids. Jour. Genet. 42: 91-114. Rosen, W. G. 1961. Studies on pollen tube chemotropism. Amer. Jour. Bot. 48: 889-95. Sax, K. 1937. Effect of variation in temperature on nuclear and c e l l division in Tradescantia. Amer. Jour. Bot. 24: 218-25. Snow, R. 1963. Alcoholic hydrochloric acid-carmine as a stain for chromosomes in squash preparations. Stain Techn. 38: 9-13. Sparrow, A. H., Moses, M. J. and Steele, R. 1952. A cytological and cytochemical approach to an understanding of radiation damage in dividing c e l l s . B r i t . Jour. Radiol. 25: 182-88. ,, and Sparrow, R. C. 1949 Treatment of Trilium erectum prior to and during mass production of permanent smear preparations. Stain Techn. 24: 47-,"55. 86. Steel, R. G. D. and Torrie, J. H. 1960. Principles and procedures of s t a t i s t i c s . McGraw, H i l l Book Co. Inc., New Y 0rk. Swaminathan, M. S., Magoon, M. L. and Mehra, K. L. 1954. A simple propiono-carmine smear method for plants with small chromosomes. Ind. Jour. Genet. Plant Breeding JL4: 87-8. Vasil, I. K. 1960. Studies on pollen germination of certain Cucurbitaceae. Amer. Jour. Bot. 47: 239-47. Walker, G. W. R. 1957. The effects of colchicine on microsporogenesis in cultured anthers of Tradescantia paludosa. Amer. Jour. Bot. 44: 690-6. White, M. J. D. 1961. The chromosomes. Methuen and Co. Ltd., London. Wilcox, E. C. 1955. The 1954 f r u i t tree survey in North Central Washington. Wash. Agr. Exp. Sta. Cir. 270. Zeuthen, E. 1964. Synchrony in c e l l division and growth. Inter-science Publishers, div. John Wiley and Sons Inc., New York - London - Sydney. 1964. 

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.831.1-0302507/manifest

Comment

Related Items