Marine Biological Laboratory Library
Woods Hole, Massachusetts
AN INTRODUCTION
TO THE
EMBRYOLOGY OF ANGIOSPERMS
McGRAW-HILL PUBLICATIONS IN THE BOTANICAL SCIENCES
Edmund W. Sinnott, Consulting Editor
Arnold An Introduction to Paleobotany
curtis and clark An Introduction to Plant Phvsiology
eames Morphology of the Angiosperms
eames Morphology of Vascular Plants: Lower Groups
eames and macdaniels An Introduction to Plant Anatomy
haupt An Introduction to Botany
haupt Laboratory Manual of Elementary Botany
haupt Plant Morphology
hill Economic Botanv
hill, overholts, popp, and grove Botany
johansen Plant Microtechnique
kramer Plant and Soil Water Relationships
kramer and kozlowski Physiology of Trees
lilly .and barnett Physiology of the Fungi
maheshwari An Introduction to the Embryology of the Angiosperms
miller Plant Phvsiology
pool Flowers and Flowering Plants
sharp Fundamentals of Cytology
sinnott Plant Morphogenesis
sinnott, dunn, and dobzhansky Principles of Genetics
sinnott and wilson Botany: Principles and Problems
smith Cryptogamic Botany
Vol. I. Algae and Fungi
Vol. II. Bryophytes and Pteridophvtes smith The Fresh-water Algae of the United States swingle Textbook of Systematic Botany weaver and clements Plant Ecology
There are also the related series of McGraw-Hill Publications in the Zoological Sciences, of which E. J. Boell is Consulting Editor, and in the Agricultural Sciences, of which R. A. Brink is Consulting Editor.
AN INTRODUCTION M ^^,
TO THE (fi ^
EMBRYOLOGY OF ANGIOSPERMS
BY
P. MAHESHWAR1
Professor of Botany, University of Delhi Delhi, India
McGRAW-HILL BOOK COMPANY, INC.
NEW YORK TORONTO LONDON 1950
AN INTRODUCTION TO THE EMBRYOLOGY OF ANGIOSPERMS
Copyright, 1950, by the McGraw-Hill Book Company, Inc. Printed in the United States of America. All rights reserved. This book, or parts thereof, may not be reproduced in any form without permission of the publishers.
VII 3962U
PREFACE
In these days of intense activity, when hundreds of papers are being published in every field of botany in a steadily increasing number of periodicals and in a multitude of languages, no apology is needed for an attempt to summarize the existing state of our knowledge in any branch of the subject and to point out the future possibilities in it. Since the publication of Coulter and Chamber- lain's "Morphology of Angiosperms" in 1903, no comprehensive account of this aspect of botany has appeared in the English lan- guage.
The original impetus for writing this work resulted from a course of lectures which I gave on the subject in 1930 when I was teaching at the Agra College. Several colleagues and pupils then suggested that I should produce a book on the embryology of angiosperms. This suggestion was repeated by Professor G. Tischler of the Uni- versity of Kiel, whom I visited in 1936. Teaching and adminis- trative duties and other difficulties made it impossible for me to carry on this work in India at the speed I should have liked. Soon after the war was over in 1945, therefore, I took the manuscript to the United States in order to revise it and put it in shape for publi- cation.
In a strict sense, embryology is confined to a study of the embryo, but most botanists also include under it the events which lead on to fertilization. I am in agreement with this wider comprehension of the subject and have therefore included in this volume not only an account of the embryo and endosperm, but also an account of the development of the male and female gametophytes and fertiliza- tion. To emphasize the recent trends of research in the subject, two chapters of a general nature have been added, one dealing with embryology in relation to taxonomy, and the other with experi- mental embryology. In the former, an attempt has been made to indicate the possibilities of the embryological method in the solu- tion of problems of systematic botany. In the latter, emphasis has been placed on the contacts between embryology, cytology, genetics, and plant physiology.
vi PREFACE
In compiling my materials I must acknowledge the immense help which I received from the writings of the late Professor K. Schnarf, whom I came to know rather intimately during my stay in Vienna in 1936. Without the existence of his books, entitled "Embryologie der Angiospermen" (1929), "Vergleichende Embryologie der Angio- spermen" (1931), and "Vergleichende Zytologie des Geschlecht- sapparates des Kormophyten" (1941), my task would have been appreciably greater. Mention must also be made of the numerous and very valuable publications of Professor E. C. R. Soueges (Paris), Professor K. V. O. Dahlgren (Uppsala), Dr. J. Mauritzon (Motala), Dr. F. Fagerlind (Stockholm), Dr. A. Gustafsson (Svalof), and Dr. H. Stenar (Sodertalje), upon which I drew rather freely. Pro- fessor Dahlgren, Dr. Gustafsson, and Dr. Stenar also favored me with their advice and criticisms whenever I applied to them for help. In addition, a host of teachers and students in the United States gave me every possible encouragement in the work. To record my gratitude to all of them in any complete fashion would fill several pages. I therefore content myself with naming a few who took special interest in the project. To Professors R. H. Wet more and I. W. Bailey I am heavily indebted for the free use of their facilities and their assistance in other ways during my several months' stay at Harvard. Professor A. F. Blakeslee and Mary E. Sanders, Smith College, Northampton; Professor E. W. Sinnott, Yale University; Professors A. J. Eames and L. W. Sharp, Cornell University; Drs. D. C. Cooper, R. A. Brink, and C. L. Huskins, University of Wisconsin; Dr. Th. Just, University of Notre Dame, now at the Field Museum of Natural History, Chicago; Professor J. T. Buchholz, University of Illinois; Professor A. S. Foster, Pro- fessor G. L. Stebbins, Mrs. M. S. Cave, Drs. L. Constance, Katherine Esau, and C. M. Rick, all of the University of California; Pro- fessor G. M. Smith, Stanford University; Dr. D. A. Johansen, Po- mona, and Professor A. W. Haupt, University of California at Los Angeles, gave me the benefit of their suggestions and criti- cisms. Last but not least, my colleagues and pupils, Dr. B. M. Johri, Reayat Kahn, S. Narayanaswami, and J. S. Agrawal gave me their fullest cooperation in the preparation of the bibliography and revision of the proofs.
Only a few of the illustrations are original, most of them having been borrowed from the works of other authors. Considerable care has been exercised, however, in their selection not only that tht
PREFACE vii
text may be made as clear as possible but also that the student may- acquire some familiarity with the names and contributions of the better known embryologists, both past and present. While most of the copying and redrawing was done by me personally, I am glad to acknowledge the very able assistance I received from a few friends. Figures 36, 68, 92, and 214 were drawn by Mrs. J. A. Adams of Poughkeepsie, N. Y., daughter of my former teacher, the late Dr. Winfield Dudgeon of Allahabad; Miss C. Pratt, Harvard University, drew Figures 24, 41, 59, 65, 74, 89, 104, 118, 121, and 167; Dr. B. G. L. Swamy, Bangalore, drew Figures 19, 145, 149, 150, 153, and 163; Mrs. M. S. Cave, University of Cali- fornia, drew Figure 14; Miss C. G. Nast, Wayne University, drew Figure 111; and my former research assistant, Ashraful Haque, University of Dacca, drew Figures 21, 43, 47, 54, 60, 61, 82, 85, 91, 147, 148, 152, 154, 157, 158, 159, 161, 162, 164, 165, 170, 172, 173, 175, 191, and 216. To all these I wish to tender my most grateful thanks for the willingness with which they cooperated with me.
A word about the citation of literature. No attempt has been made to give a complete list of all that has been published on angiosperm embryology, as this would make the volume too cum- bersome, but it is hoped that the references which have been cited will facilitate the task of the student who wishes to acquire fuller information.
In a work of this nature it is unavoidable that there should be some errors of judgment and also oversights and omissions. I should appreciate the suggestions and criticisms of those who use
the book.
P. Maheshwari University of Delhi, India July, 1950
CONTENTS
Preface v
1 . Historical Sketch 1
Discovery of Pollen Tube — Schleiden's Theory of Origin of Embryo — Discovery of True Relation between Pollen Tube and Embryo — Discovery
of Sexual Fusion in Lower Plants — Discovery of Nature and Development of Male and Female Gametophytes — Embryo — Discovery of Syngamy — Chalazogamy — Double Fertilization — Parthenogenesis — Twentieth Cen- tury— References.
2. The Microsporangium 28
Wall Layers — Sporogenous Tissue — Cytomixis — Cytokinesis — Microspore Tetrad — References.
3. The Megasporangium 54
Integuments — Micropyle — Nucellus — Integumentary Tapetum — Hypo- stase— Epistase — Vascular Supply of Ovule— Arch esporium — Megasporo- genesis — Functioning Megaspore — Failure of Wall Formation during Meiosis — References.
4. The Female Gametophyte 84
Monosporic embryo sacs — Polygonum Type— Oenothera Type — Bisporic embryo sacs — Allium Type — Tetrasporic embryo sacs — Peperomia Type — Penaea Type — Drusa Type — Fritillaria Type — Plumbagella Type — Adoxa Type — Plumbago Type— Aberrant and unclassified types — Limnanthes douglasii — Bahamita vulgaris — Chrysan themum cinerariaefolium — Or- ganization of mature embryo sac — Egg Apparatus — Antipodal Cells — Polar Nuclei — Embryo Sacs with Disturbed Polarity — Food Reserves in the Embryo Sac — Embryo Sac Haustoria — References.
5. The Male Gametophyte 154
Microspore — Formation of Vegetative and Generative Cells — Division of Generative Cell— Male "Cells" or "Nuclei"— Vegetative Nucleus— De- velopment of Pollen in Cyperaceae — Embryo-sac-like Pollen Grains — References.
6. Fertilization 181
Germination of Pollen — Course of Pollen Tube — Entry of Pollen Tube into Embryo Sac — Rate of Growth of Pollen Tube — Gametic Fusion — Multiple Fusions and Polyspermy — Single Fertilization — Persistence and Possible Haustorial Function of Pollen Tube — X-bodies — References.
7. The Endosperm 221
Types of Endosperm Formation — Free Nuclear Endosperm — Cellular Endosperm — Helobial Endosperm — Relationships between Different Types of Endosperm — Histology of Endosperm — Xenia — Mosaic Endo- sperm— References.
ix
x CONTENTS
8. The Embryo 268
Dicotyledons — Crucifer Type — Asterad Type — Solanad Type — Cheno- podiad Type — Caryophyllad Type — Monocotyledons — Modifications of suspensor — Unclassified and abnormal embryos — Unorganized and reduced embryos — References .
9. Apomixis 313
Nonrecurrent apomixis — Recurrent apomixis — Generative Apospory — So- matic Apospory — Unclassified Cases — Organization of Aposporic Embryo Sacs — Development of Embryo in Aposporic Embryo Sacs — Adventive embryony — References.
10. POLYEMBRYONY 343
Cleavage Polyembryony — Origin of Embryos from Cells of Embryo Sac Other than Egg — Embryos Arising from Cells Outside Embryo Sac — Embryos Originating from Other Embryo Sacs in Ovule — A Few Special Cases — Twins and Triplets — Conclusion — References.
11. Embryology in Relation to Taxonomy 357
Empetraceae — Lennoaceae — Cactaceae — Garryaceae — Onagraceae — Calli- trichaceae — Liliaceae-allioideae — Liliaceae-asphodeloideae — Liliaceae-lili- oideae — References.
12. Experimental Embryology 375
Control of Fertilization — Embryo Culture — Induced Parthenogenesis — Production of Adventive Embryos — Induced Parthenocarpy — Conclu- sion— References.
13. Theoretical Conclusions 411
Male Gametophyte — Female Gametophyte — Fertilization — Endosperm — Embryo — References.
Name Index 433
Subject and Plant Index 441
CHAPTER 1 HISTORICAL SKETCH
In tracing the history of a branch of natural science it is customary to go back to the days of Aristotle. The greater part of his technical writings is unfortunately lost to us, but it seems fairly certain that he did not recognize the presence of sex in plants. He believed instead that the male and female principles were so blended that they generated of their own accord and the offspring arose from the superfluous food in the plant.
Aristotle bequeathed his library and collections to his favorite pupil Theophrastus. In his "Enquiry into Plants," written in the third century B.C., the latter referred to the pollination of the date palm, presumably on the basis of the account of Herodotus, who had traveled in the East in the fifth century B.C. The Arabs and Assyrians, Herodotus found, used to have a special ceremony at a certain time of the year, in which a man climbed up a male tree, brought down the inflorescence, and handed it over to the high priest, who touched the female inflorescences with it, in order to ensure a good supply of dates.
Approximately three hundred years after Theophrastus, Pliny wrote an encyclopedia of natural history in which he mentioned the male palm with its erect leaves as having somewhat of a military bearing, while the females with their softer foliage and feminine ways bent toward it, to save themselves as it were from the curse of virginity or widowhood. However, Pliny did not make any ob- servations of his own. His writings and ideas were based on other people's reports and on the literature on the subject that existed in those days.
After this the problem of sexuality in plants seems to have been laid aside and forgotten for hundreds of years. Indeed, many scien- tists of the fifteenth and sixteenth centuries totally denied the occurrence of sex in plants and regarded even the mention of it as inappropriate and obscene. Some thought the stamens to be ex- cretory organs and the pollen to be a waste product.
l
2 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
It was only with the invention of the microscope that actual observation of the sexual cells took the place of conjectures. Leeuwenhoek (1677)1 discovered the sperms of some animals but mistook them at first for "Wild animalcules" arising in the seminal fluid by some sort of putrefaction.
In his "Anatomy of Plants," Grew (1682) made the first explicit mention of the stamens as the male organs of the flower. He thought that the pollen grains, by merely falling upon the stigma, transmitted to the ovary a "vivifick effluvium" which prepared it for the production of the fruit.
Rudolph Jakob Camerarius (1694), Director of the Botanical Garden at Tubingen, approached the matter more scientifically. He observed that in a female mulberry tree, which was growing without any male plants in the vicinity, the fruits contained only abortive seeds. Inspired by this discovery he next took some female plants of Mercurialis annua and kept them in pots completely isolated from the influence of male plants. Here too, he found that, although the plants grew well, not one of the fruits contained a fertile seed. This encouraged him to make further observations, which he summarized in a famous treatise called "De sexu plantarum." He carefully de- scribed the flower, anthers, pollen, and ovules. On removing the male flowers (globuli) of Ricinus before the anthers had shed and preventing the growth of the younger ones, he never obtained any perfect seed but only empty fruits which withered and fell to the ground. A similar lack of seed formation was noted in Zea mays when the stigmas had been removed from the young ear. In con- clusion he said: "In the plant kingdom, the production of seed, which is the most perfect gift of nature and the general means of maintenance of the species, does not take place unless the anthers have previously prepared the young plant contained in the ovary." To the anthers, in his opinion, was to be attributed, therefore, the role of the male sexual organs just as the ovary with its style was considered the female sexual organ.
We thus see that although Camerarius was not clear about the exact manner in which the pollen functioned, he nevertheless made a notable contribution to our knowledge by showing that some kind
1 Dates in parentheses refer to works listed in the bibliography at the end of each chapter.
HISTORICAL SKETCH
of interaction between the stamens and carpels is necessary for the production of seed-bearing fruits.
About sixty-five years later, Joseph Gottlieb Kolreuter (1761), physician and professor of natural history at Wurtemberg, published four parts of a treatise dealing with his experiments on sex in plants. He fully confirmed the work of Camerarius and gave a detailed ac- count of the importance of insects in flower pollination. He also
produced hybrids in Nicotiana,
Dianthus, Matthiola, and Hyo- scyamus and showed that if the stigma of a plant received its own pollen and that of another species at the same time, ordinarily the former alone was effective. This, he said, was the reason why hy- brids were so rare in nature, al- though they could be produced artificially.
Discovery of the Pollen Tube. After the role of the pollen began to be understood, the next step was to determine the exact man- ner in which it influenced the ovule. Accident supplied the starting point of some important discoveries. An Italian mathe- matician and astronomer named Giovanni Battista Amici (1824),
who was also a good microscope maker, found that the stigma of Portu- laca oleracea was covered with hairs which contained some granules or particles inside them. Curiosity prompted him to ascertain whether they moved in the same way as the granules he had seen in the cells of Char a. It pleased him to find that they did. While repeating the observation, he accidentally saw a pollen grain attached to the hair he had under observation. Suddenly the pollen grain split open and sent out a kind of tube or "gut" which grew along the side of the hair and entered the tissues of the stigma. For three hours he kept it under observation and watched the cytoplasmic granules circulate
Fig. 1. Giovanni Battista Amici. {Pho- tograph obtained through the courtesy oj Br. E. Battaglia.)
4 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
inside it, but eventually he lost sight of them and could not say whether they returned to the grain, entered the stigma, or dissolved away in some manner.
Amici's discovery stimulated the young French botanist Brong- niart (1827) to examine a large number of pollinated pistils with a view to understanding the interaction between the pollen and the stigma and the introduction of the fertilizing substance into the ovule. He found the formation of the pollen tubes (he called them "spermatic tubules") to be a very frequent occurrence but per- suaded himself to believe that, after penetrating the stigma, the tubes burst and discharged their granular contents, which he likened to the spermatozoids of animals and considered to be the active part of the pollen. He thought he saw these "spermatic granules" vi- brating down the whole length of the style and entering the placenta and ovule, and he drew a series of figures to illustrate the whole process. In appreciation of this work, Brongniart was awarded a prize by the Paris Academy of Sciences and recommended for ad- mission to the Academy.
Amici (1830) applied himself once again to the problem, studying Portulaca oleracea, Hibiscus syriacus, and other plants, and wrote a letter to Mirbel in which he put the following question: "Is the prolific humor passed out into the interstices of the transmitting tissue of the style, as Brongniart has seen and drawn it, to be trans- ported afterwards to the ovule, or is it that the pollen tubes elongate bit by bit and finally come in contact with the ovules, one tube for each ovule?" His observations completely ruled out the first alternative, and he definitely concluded in favor of the second.
About the same time, Robert Brown (1831, 1833) saw pollen grains on the stigmas and pollen tubes in the ovaries of certain orchids and asclepiads but was uncertain as to whether the tubes were always connected with the pollen grains. He thought instead that, at least in some cases, the tubes arose within the style itself, although pos- sibly they were stimulated to develop in consequence of the pollina- tion of the stigma.2
Schleiden's Theory of the Origin of the Embryo. Meanwhile other workers also became interested in the problem, and in 1837 Schleiden published some very detailed observations on the origin
2 It now seems that Brown was at times confusing pollen tubes with the elon- gated cells of the transmitting tissue in the style.
HISTORICAL SKETCH
and development of the ovule. He confirmed Amici's statement that the pollen tubes make their way from the stigma to the ovule, entering the latter through the micropyle. His lively imagination carried him too far, however, for he asserted that the extremity of the pollen tube pushes the membrane of the embryo sac before it and directly becomes the embryonal vesicle, which then undergoes a number of divisions to produce the embryo. The cotyledons were said to arise laterally, while the original apical point remained pr" r more or less free and formed the plumule. To him the em- bryo sac was, therefore, a sort of nidus or incubator within which the end of the pollen tube was nourished to give rise to the new plantlet. If this were really the case, there would of course be no sexuality in plants. Nevertheless, with the influence he commanded and the sharp tongue with which he denounced all opponents, Schleiden found a number of warm supporters. One of them, Schacht, sponsored this absurd idea with special en- thusiasm.
Amici boldly opposed the views of Schleiden. In a meet- ing of the Italian naturalists,
held at Padua in 1842, he tried to prove that the embryo did not arise from the tip of the pollen tube but from a portion of the ovule which was already in existence and was fertilized by the fluid in the tube.
Schleiden (1845) gave a most spirited reply to this and said that after his careful and thorough investigation of 1837 it was ridiculous on the part of novices in the field to raise such meaningless objec- tions. He described some fresh observations on Cucurbita and of- fered to demonstrate the utter falsity of Amici's observations and the complete truth of his own to anyone who visited him.
Fig. 2. Matthias Jakob Schleiden. (Photograph obtained through the cour- tesy of Prof. W. Troll.)
6
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
Discovery of the True Relation between the Pollen Tube and the Embryo. In spite of Schleiden's criticism, Amici continued further work on the subject. In 1847 he produced decisive evidence (Fig. 3) to show that in Orchis (which he found to be specially suited for such studies), a body, the germinal vesicle, was already present inside the embryo sac before the arrival of the pollen tube, and that it was this vesicle which gave rise to the embryo, stimulated no doubt by the presence of the pollen tube.
B C D E
Fig. 3. Development of ovule and embryo in Orchis. Note pollen tube in C and D and suspensor haustorium in E. {After Amici, 1847.)
Support for Amici 's views now came forward from other quarters. In a famous document, entitled "Die Entstehung des Embryo der Phanerogamen," consisting of 89 quarto pages and 14 copper plates with no fewer than 429 figures, Wilhelm Hofmeister (1849) published his observations on 38 species belonging to 19 genera and showed that in every case the embryo originated from a preexisting cell in the embryo sac and not from the pollen tube. He described his obser- vations in such a clear and dignified manner that the}' immediately carried conviction and were soon confirmed by other workers from
HISTORICAL SKETCH
England, France, and Germany. In less than two years after the publication of this memoir and in spite of his lack of a proper university training, the University of Rostock conferred upon Hofmeister the degree of Doctor of Philosophy honoris causa, thereby giving formal recognition to his high position as a scientific investi- gator. A few of Hofmeister's illustrations of the embryo sac and the relation between the pollen tube and the egg are presented in Fig. 5.
Schleiden and Schacht contin- ued to hold their previous opin- ion . Schacht brought out a large monograph in 1850, with 26 plates and a considerable number of drawings. These were beauti- fully executed, but in every case he mistook the egg cell for the tip of the pollen tube (Fig. 6) . In conclusion he said: "The tend- ency towards error is so inherent in human nature that the work of one's head, like that of his hand, is never perfect, and con- sequently I do not hold mine to be free from error and mis- conception, but I have tried to minimize these as much as pos- sible. ... In chief matter, i.e.,
the origin of the embryo from the pollen tube, no one can convince me that there has been any mistake or misconception. . . . My preparations are so conclusive on this point that I can confidently look forward to answering any criticisms that may be directed against it." The Imperial Institute of the Netherlands at Am- sterdam accepted Schacht's essay and awarded him a prize for its production.
However, the evidence against Schleiden and Schacht soon be- came so overwhelming that eventually both of them had to retract their opinions, and in 1856 Radlkofer published a comprehensive review of the question accepting Hofmeister's conclusions in toto. Schleiden soon gave up all botanical work and settled down in Dresden as a private teacher of history and philosophy.
Fig. 4. Wilhelm Hofmeister.
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
In this connection it is interesting to recall the words of the famous anatomist Hugo von Mohl, who, in 1863, at the time of Amici's death, wrote as follows: "Now that we know Schleiden's doctrine to have been an illusion, it is instructive, although sad, to look back to
D
E
B C
Ovules and embryo sacs of Monotropa hypopitys, before and after fertili-
A
Fig. 5.
zation. A,B, embryo sacs at time of fertilization. C, same, showing pollen tube about to enter micropyle. D,E, Fertilized embryo sacs, showing early stages in formation of endosperm. (After Hofmeister, 1849.)
the past and see how readily the false was accepted for the true; how some, renouncing all observation of their own, dressed up the phantom in theoretical principles; how others, with microscope in hand, but blinded by their preconceptions, believed that they saw what they could not have seen and sought to establish the correct- ness of Schleiden's notions with the aid of hundreds of figures which
HISTORICAL SKETCH 9
had anything but truth to recommend them; and how an academy by rewarding such work gave fresh proof of the well-known experi- ence that prize-essays are little adapted to contribute to the solution of a doubtful question in science."
Discovery of Sexual Fusion in Lower Plants. During this interval greater progress was being made with lower plants and animals. Thuret, in 1854, showed that in Fucus the eggs must be activated by sperms before they can germinate to give rise to new plants, and
A B C D E
Fig. 6. The so-called development of embryo from pollen tube in Martynia lutea (tp = pollen tube; em = embryo ;edp = endosperm; is = integument; se = embryo sac). A, l.s. ovule. B-D, stages in development of "pollen-tube embryo." E, older embryo, together with a few of the surrounding endosperm cells. (After Schacht, 1850.)
later he also obtained hybrids by associating the ova and sperms of different forms. In 1855 Pringsheim observed spermatozoids in the little horns (antheridia) of Vaucheria and showed that no further development occurs unless the spermatozoids enter the ovum. The decisive observation was made in 1856 in Oedogonium, where he saw the moving spermatozoid come in contact with the egg and force its way inside the latter. On the basis of these and similar dis- coveries in lower animals, the German zoologist Oscar Hertwig (1875) made a general statement that the essential feature of fertili- zation is the union of two nuclei, one furnished by the male parent and the other by the female.
In the phanerogams, where sex was supposed to be more apparent
10
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
than in cryptogams, the actual demonstration did not come until a few years later, no doubt because of the technical difficulties in making any direct observations on the embryo sac, which is sur- rounded by the opaque tissues of the nucellus and the integuments.
Fig. 7. Stages in formation of microspore tetrads in Tradescantia. {After Hof- meister, 1848; reproduced from Sharp, 1943.)
Discovery of the Nature and Development of Male and Female Gametophytes. Among early students of the development of pollen, Hofmeister (1848) presented some surprisingly good illustrations of
A B C D E F G
Fig. 8. Development of male gametophyte of Tradescantia virginica. (After Elfving, 1879.)
the process of tetrad formation (Fig. 7), and Reichenbach, Hartig and several other workers noted the presence of two nuclei in whole mounts of the mature pollen grains of several angiosperms. Stras- burger (1877) and his pupil Elfving (1879) extended these observa- tions to cover several families and demonstrated the widespread occurrence of the binucleate condition in pollen grains (Fig. 8).
HISTORICAL SKETCH
11
They further found that one of these nuclei originally lies in a small cell cut off at the periphery of the pollen grain but later be- comes free by a dissolution of the partition wall. Elfving also germinated pollen grains in artificial media, and when this was un- successful, he made preparations of pollen tubes from dissected styles. Here he was able to find the three nuclei which we now know to be the two male gametes and the tube or vegetative nucleus. Unfortunately both Strasburger _________
and Elfving made the mistake of interpreting the smaller cell in the pollen grain as vegeta- tive or prothallial and the larger as generative. They further thought that all the nuclei in the pollen tube dissolved and disappeared before fertilization. These mistakes were, however, rectified by Strasburger in a sub- sequent paper (1884), which will be referred to later.
For our knowledge of the or- ganization of the embryo sac we are indebted in the first instance to the works of Hofmeister (1847-1861). Working wholly with cleared preparations and freehand sections, he succeeded in identifying the two groups of cells at the opposite poles of the embryo sac. Those lying at the micropylar end were designated as the "germinal" or "embryonal" vesicles, all capable of giving rise to embryos and therefore to be regarded as homologous with the corpuscula (archegonia) of the gymnosperms. The cells at the chalazal end were considered to be prothallial, and the embryo sac itself was interpreted as homologous with the megaspore or female gametophyte of the heterosporous pteridophytes and the gymnosperms.
Although Hofmeister's work was important, he failed to dis- tinguish clearly between the synergids and the egg and regarded all three of them as having the same function. Further, he was unable to trace the mode of origin of the embryo sac, the general opinion
Fig. 9. Edward Strasburger. (Photo- graph obtained through the courtesy of Prof. A. W. Haupt.)
12
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
in those days being that it arose by the simple enlargement of a cell of the nucellus.
Further knowledge of the development and organization of the embryo sac became available as the result of a concentrated attack on the problem made by several botanists during the years 1877 to 1881. To be named specially in this connection are Warming,
D E F G H
Fig. 10. Development of embryo sac in Polygonum divaricatum. A, megaspore mother cell separated from nucellar epidermis by primary wall cell. B, dyad cells in division. C, tetrad of megaspores with wall cells above. D, functioning mega- spore. E-G, embryo sacs, showing two, four, and eight nuclei. H , l.s. ovule, show- ing mature embryo sac. (After Strasburger, 1879.)
Vesque, Strasburger, Fischer, Ward, Jonsson, Treub and Mellink, and Guignard. Strasburger (1879) demonstrated that at first one of the nucellar cells becomes differentiated as the megaspore mother cell (Fig. 10 A) and goes through two divisions to give rise to a row of four cells (Fig. 10B,C). Of these, the three micropylar cells soon degenerate and the chalazal alone enlarges and functions (Fig. 10D). The nucleus of this cell (the functioning megaspore) divides thrice to give rise to two groups of four nuclei, one at the micropylar end and the other at the chalazal end of the cell (Fig. 10E-G). From
HISTORICAL SKETCH
13
the former arise the egg apparatus (consisting of an egg cell and two synergids) and the upper polar nucleus; from the latter, the three antipodal cells and the lower polar nucleus. The polar nuclei were observed to fuse in the center to form a secondary nucleus, which
Fig. 11. Stages in development of embryo of Capsella bursa-pastoris (v = sus- pensor; h-h' = hypophysis ; c-c = cotyledons ; s = stem tip ;w = radicle; the shaded portions represent the dermatogen and plerome). (After Hanstein, 1870; repro- duced from Sachs, 1874-)
gave rise to the endosperm (Fig. \0H). The synergids were re- garded as modified structures assisting in the process of fertilization. Treub and Mellink (1880) confirmed these observations but also noted certain exceptions. In a few plants they found that the megaspore mother cell divides into only two daughter cells, of which either the upper (as in Agra-phis patula) or the lower (as in Narcissus
14
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
tazetta) can give rise to the embryo sac. In Lilium and Tulipa the formation of daughter cells was found to be entirely omitted so that the embryo sac arises directly from the megaspore mother cell. . The Embryo. Hanstein (1870) was the first to follow the sequence of early cell divisions in the development of the embryo. He gave a detailed description of the embryogeny in Capsella and Alisma
Fig. 12. Stages in development of embryo of Alisma plantago (v = suspensor; h = hypophysis; w = radicle; p = plumule; c = cotyledon; b = first leaf; the shaded portions represent the derma togen). (After Hanstein, 1870; reproduced rom Sachs, 187 4-)
(Figs. 11, 12). Famintzin confirmed these observations in 1879, and in the same year Treub described the embryos of several orchids with their remarkable suspensor haustoria. Two years later Guig- nard (1881) gave a full account of the extremely massive suspensors of the Leguminosae.
At about this time, detailed investigations were also made on the peculiar phenomenon of polyembryony. Long ago Leeuwenhoek (1719) had noted the occurrence of more than one embryo in certain
HISTORICAL SKETCH
15
orange seeds, and other instances of a similar nature were listed by- Alexander Braun (1859). In no case, however, had the origin of the abnormality been satisfactorily studied from the developmental point of view. Strasburger, in 1878, demonstrated for the first time that in Funkia (=Hosta) ovata, Coelebogyne ( = Alchornea) ilicifolia, Nothoscordum fragrans, and Citrus aurantium, the nucellar cells lying close to the apex of the embryo sac become richly protoplasmic and divide to form small groups of cells which project into the cavity of the embryo sac and grow into embryos (Fig. 13). Subsequent work
B D
Fig. 13. Development of ad ventive embryos in Funkia (= Hosta)ovata. A, upper part of nucellus and embryo sac. B,C, enlargement and division of some of the nucellar cells. D, more advanced stage, showing young zygotic embryo and several nucellar embryos. (After Strasburger, 1878.)
by others revealed further possibilities, such as an origin of embryos from the cells of the integument, or from those of the suspensor, or from components of the embryo sac other than the egg. In Allium odorum, Tretjakow (1895) and Hegelmaier (1897) showed that even antipodal cells could give rise to embryos.
Discovery of Syngamy. These were all notable advances, but the most important of all was Strasburger 's (1884) discovery of the actual process of syngamy, or the fusion of the male and female gametes. In a memorable paper, entitled "Neue Untersuchungen uber den Befruchtungsvorgang bei den Phanerogamen," he cor- rected some of the mistakes he had made in 1877 on the organization
16
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
of the male gametophyte. He now confessed that, owing to its similarity both in position and origin with the prothallial cells of gymnosperms, the small lenticular cell in the angiosperm pollen grain had been formerly misinterpreted by him as being the "vegeta- tive" cell. Also, in the earlier studies in his laboratory (which had been based on whole mounts stained with iodine green) the nuclei in the pollen grains were often quite indistinguishable and had there- fore been supposed to have degenerated. His improved technique (such as staining with picrocarmine and the cutting of the larger pollen grains into thin sections), devised after 1877, had shown that
B C
E F H
Fig. 14. Fertilization in Monotropa hypopitys. A, pollen grain stained in iodine green and acetic acid to show vegetative and generative nucleus. B,C, tips of pollen tubes showing the two male nuclei; in B the vegetative nucleus is also visible. D, upper part of embryo sac, showing egg apparatus. E-G, stages in union of male and female nuclei. H, syngamy completed; primary endosperm nucleus dividing. (After Strasburger, 1884-.)
this was really not the case. It was now clear that the 'first division of the microspore gives rise to two cells, the smaller being the genera- tive cell and the larger the vegetative. Further, the generative cell loosens itself from the wall of the pollen grain and divides either before or after the germination of the pollen grain, while the vegeta-
HISTORICAL SKETCH
17
tive or tube nucleus remains undivided. Thus, the pollen tube eventually shows three nuclei, one vegetative and two generative3.
On the basis of his studies on the embryo sac of Monotropa and some other plants, Strasburger further showed that the pollen tube discharges its nuclei into the sac (previous to this it was believed that fertilization occurred merely by the diffusion of the cell sap from the tube) and that one of the two male nuclei fuses with the nucleus of the egg, thus provid- ing actual proof of the nature "™ of fertilization and its import- ance in the life cycle of a plant (Fig. 14).
In the concluding part of his memoir, Strasburger made the following generalizations, which are now almost axiomatic with us: (1) the process of fertilization comprises the union of the nu- cleus of the male gamete with that of the egg; (2) the cyto- plasm of the gametes is not con- cerned in the process ; and (3) the sperm nucleus and the egg nu- cleus are true nuclei.
Chalazogamy. Strasburger's work opened the way to a more detailed study of the process of fertilization in angiosperms. Prior to 1891, it was believed that
the pollen tube always enters the ovule through the micropyle. Treub, in that year, reported that in Casuarina it enters through the chalaza (Fig. 16). This was thought to be so strange that he proposed a new classification of the angiosperms into two classes : the chala- zogams and theporogams, with Casuarina as the only representative of the former. Later investigations showed, however, that there is no uniformity in the mode of entry of the pollen tube into the embryo sac, and in Ulmus (Nawaschin, 1898a) its behavior was found to be particularly varied and irregular. The phenomenon of chalazogamy,
'These two are now called the male gametes.
Fig. 15. Melchior Treub. (Photo- graph obtained through the courtesy of Dr. F. Verdoorn)
IS
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
therefore, lost the great phylogenetic importance which had been attached to it by Treub. Today it is considered to be more of physiological than of phylogenetic significance, although it does have a certain taxonomic value in narrow circles of affinity.
Double Fertilization. The fate of the second male gam- ete discharged by the pollen tube was not known so far. In a study of Lilium marta- gon and Fritillaria tenella, S. G. Nawaschin (18986) showed that in angiosperms both male gametes are con- cerned in fertilization, one fus- ing with the egg (syngamy) and the other with the two polar nuclei (triple fusion). A few months later L. Guignard (1899) also reported the same phenomenon in Lilium and Fritillaria and presented a se- ries of beautiful drawings to illustrate it (Fig. 19). These discoveries attracted wide- spread attention and were fol- lowed by a series of similar investigations dealing with other species of angiosperms. Double fertilization was soon demonstrated in several plants and within a few years it began to be considered as of universal occurrence in angiosperms. It is interesting to note that a year earlier D. M. Mottier (1897) had seen the second male nucleus in close proximity to one of the polar nuclei, but that he had considered this proximity to be accidental and had failed to realize its true significance. Of considerable interest in this connection is Finn's (1931) report on a preparation of Scilla sibirica ( = S. cernua) made by a Russian botanist, W. Arnoldi, a number of years before Nawaschin's announcement of 1898. Finn found both
Fig. 16. Casuarina suberosa, l.s. ovule, showing chalazogamy (m = micropyle; p = pollen tube;e = embryo sac). (After Treub, 1891.)
HISTORICAL SKETCH
19
syngamy and triple fusion to be so clear in one of the sections on this slide (Fig. 20) that it is surprising that the process could have been missed at all. As it was, however, Arnoldi mistook the male gam- etes for displaced nuclei (of the nucellus?) which had in some way entered into the embryo sac during the process of sectioning, and he therefore ignored them altogether.
«.V<*8«*}
J22L £^
Fig. 17.
Sergius Nawaschin. (Photograph ob- tained through the courtesy of Dr. A. W. Haupt.)
Fig. 18. Leon Guignard.
One of the results of Nawaschin 's discovery was that it gave a plausible explanation of "xenia." This term had been coined by Focke (1881) to denote those cases in which the pollen produced a visible influence on the hereditary characters of those parts of the ovule which surround the embryo. It now became clear that just as the fertilized egg gives rise to an embryo combining the charac- ters of the two parents, so does the triple fusion nucleus give rise to a tissue containing the potentialities of both the parents.
A controversy soon started, however, on the morphological nature of the endosperm, which is neither n nor 2n but Sn. Some claimed that it was a continuation of the old gametophytic tissue, while others (Sargant, 1900) thought it to be a second embryo which took
20
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
a monstrous and highly abnormal shape because of the intrusion of the lower polar nucleus. Strasburger (1900) suggested that only the fusion of the male gamete with the egg was to be regarded as true or "generative" fertilization, while the fusion of the polar nuclei
I J K L M N
Fig. 19. Double fertilization in Lilium martagon. A, mature embryo sac. B, same, showing discharge of pollen tube. One male nucleus has entered the egg and the other is in contact with the upper polar nucleus; the nucleus of one of the synergids is in process of degeneration. C, one male nucleus in contact with the egg nucleus and the other in contact with the two polar nuclei. D, same, slightly more advanced stage. E-H , stages in fusion of egg nucleus and one male nucleus. I-N, stages in triple fusion. {After Guignard, 1899.)
HISTORICAL SKETCH
21
with the second male nucleus was in the nature of a growth stimulus and could therefore be called "vegetative" fertilization.
Parthenogenesis. About the same time that Nawaschin made his discovery of double fertilization, two Swedish botanists, H. O.
Fig. 20. Double fertilization in Scilla sibirica. A, one male nucleus in contact with egg nucleus, another in contact with the two polar nuclei. B, sperm and egg nuclei, more highly magnified. C, sperm nucleus in contact with two polar nuclei, more highly magnified. (After Finn, 1931.)
Juel (1898, 1900) and S. Murbeck (1897, 1901), were engaged in studying the mechanism of parthenogenesis in Antennaria and Al- chemilla. Some years earlier Kerner (1876) had noted that in An- tennaria alpina male plants were extremely rare in nature but that even unpollinated female plants were able to form seeds. Juel made a thorough study of the development and showed that even when staminate plants do occur, the pollen is either lacking or only feebly
22
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
developed. In the ovules the megaspore mother cell develops di- rectly into the embryo sac without any reduction in the chromosome number and the diploid egg produces an embryo without fertiliza- tion (Fig. 21). Murbeck similarly showed that some species of
^Ol
^o
01
0 O
0
0
.0,
0)
0
i
i
\o
\0
0
c
0
0
o
'0/
0
fa
o
o
'01
jy
a
o
'o
0
o
m.
-^fe®
.01
Wm
o]
&
0
(J
o
0
o.
r0
0
ol
Oy
0
0}
a
07
0
,o
0
0
! i
i
«;<€
o\
o
0
^0 K)
o
o
oi
0
0
pi
B C
Fig. 21. Apomixis in Antennaria alpina. A, mature embryo sac. B, later stage, showing enlargement of egg and increase in number of antipodal cells; polar nuclei preparing to divide. C, embryo two-celled; polar nuclei in division. (Re- drawn after Juel, 1900.)
HISTORICAL SKETCH
23
Alchemilla, belonging to the section Eualchemilla, develop partheno- genetically without any chromosome change in the life cycle.
The Twentieth Century. The year 1900 marked the beginning of a new era in angiosperm embryology. By this time most of the facts on the development of the gametophytes and embryo had been discovered, and an able summary of the literature was given by Coulter and Chamberlain (1903) in their book entitled "Morphology of Angiosperms." The stage was now set for more detailed inves- tigations of special topics to clear up previous obscurities, and for studies of a comparative nature on whole families and orders to determine what light embryo- logy could throw on problems of taxonomy.
To outline the contributions of the numerous individuals who have been engaged in such studies during recent years is out of place in this brief and introductory sketch but will be attempted in the following chapters . Mention may be made here of the names of a few whose contributions have been especially noteworthy.
Among modern students of the subject, the name of the late Karl Schnarf of Vienna stands preeminent. His two works entitled "Embryologie der Angio- spermen" (1929) and "Vergleichende Embryologie der Angiosper- men" (1931) are the most important and exhaustive treatises in this field, and still serve as valuable works of reference. E. C. R. Soueges of France has distinguished himself by his painstaking studies on the development of the embryo in several families and genera of both dicotyledons and monocotyledons; and W. W. Finn in the Ukraine has similarly engaged himself in a study of the development and structure of the male gametophyte. In Swe- den, Sv. Murbeck, 0. Rosenberg, and the late H. O. Juel in- spired a flourishing school of research on all phases of embryology
Fig. 22. Karl Schnarf.
24 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
(especially gametogenesis and apomixis), work on which is being continued at present by K. V. 0. Dahlgren, F. Fagerlind, A. Hakans- son, H. Stenar, and A. Gustafsson. Among workers from other countries may be mentioned the names of A. Chiarugi and the late E. Carano from Italy; A. Ernst from Switzerland; H. D. Wulff from Germany; and the late J. M. Coulter and D. S. Johnson from the United States. During recent years there has also been con- siderable activity in this field in India, Japan, and Australia.
Of particular interest is the origin of the new science of experi- mental embryology, dealing with problems of storage and viability of pollen, effect of environmental factors on pollen tube growth, control of fertilization, production of seedless fruits, embryo culture, and artificial induction of parthenogenesis and adventive embryony. Here embryology stands in intimate relation with physiology and genetics, and promises to offer many opportunities and openings for the future.
References
Amici, G. B. 1824. Observations microscopiques sur diverses especes de plantes.
Ann. des Sci. Nat., Bot. 2: 41-70, 211-248. . 1830. Note sur le mode d'action du pollen sur le stigmate. Extrait
d'une lettre d Amici a Mirbel. Ann. de? Sci. Nat., Bot. 21: 329-332. . 1844. Quatrieme reunion des naturahstes italiens. Padua, 1843. Flora
1:359. . 1847. Sur la fecondation des Orch'dees. Ann. des Sci. Nat., Bot.
7/8: 193-205. Aristotle. "The Works of Aristotle." Engl, transl. by J. A. Smith and W. D.
Ross. Oxford, 1913. Braun, A. 1859. tlber Polyembryonie und Keimung von Coelobogyne. Abh.
Konigl. Akad. Wiss. Berlin, phys. KL, pp. 107-263. Brongniart, A. 1827. Memoire sur la generation et le developpement de l'em-
bryon dans les vegetaux phanerogamiques. Ann. des Sci. Nat., Bot. 12:
14-53, 145-172, 225-298. Brown, R. 1831. Observations on the organs and mode of fecundation in Orchi-
deae and Asclepiadeae. Trans. Linn. Soc. London 16: 685-745. . 1833. "The Miscellaneous Botanical Works of Robert Brown." Lon- don, 1866-1868. Camerarius, R. J. 1694. "De sexu plantarum epistola." Tubingen. Coulter, J. M., and Chamberlain, C. J. 1903. "Morphology of Angiosperms."
New York. Elfving, F. 1879. Studien iiber die Pollenkorner der Angiospermen. Jenaische
Ztschr. f. Naturw. 13: 1-28.
HISTORICAL SKETCH 25
Famintzin, A. 1879. Embryologische Studien. Mem. Acad. Imp. des Sci. St.
Petersburg VII, 26(10): 1-19. Finn, W. W. 1931. Zur Geschichte der Entdeckung der doppelten Befruchtung.
Ber. deutsch. bot. Gesell. 49: 153-157. Focke, W. 0. 1881. "Die Pflanzen-Mischlinge, ein Beitrag zur Biologie der
Gewachse." Berlin. Grew, N. 1682. "The Anatomy of Plants." London. Guignard, L. 1881. Recherches d'embryogenie vegetale comparee. I. Legu-
mineuses. Ann. des Sci. Nat., Bot. 12: 5-166. . 1899. Sur les antherozoides et la double copulation sexuelle chez les
vegetaux angiospermes. Rev. Gen. de Bot. 11: 129-135. Hanstein, J. 1870. Die Entwickelung des Keimes der Monocotylen und Dicoty-
len. Bot. Abhandl. Bonn 1: 1-112. Hegelmaier, F. 1897. Zur Kenntnis der Polyembryonie von Allium odorum.
Bot. Ztg. 55: 133-140. Herodotus, circa 484-425 B.C. "Historiae." Engl, transl. by A. D. Godley.
London, 1921. Hertwig, O. 1918. "Dokumente zur Geschichte der Zeugungslehre. Eine his-
torische Studie." Bonn. Hofmeister, W. 1847. Untersuchungen des Vorgangs bei der Befruchtung der
Oenotheren. Bot. Ztg. 5: 785-792.
-. 1848. tJber die Entwicklung des Pollens. Bot. Ztg. 6 : 425-434, 649-
658, 670-674.
. 1849. "Die Entstehung des Embryo der Phanerogamen." Leipzig.
. 1859. Neue Beitrage zur Kenntnis der Embryobildung der Phanero- gamen. I. Dikotyledonen mit urspriinglich einzelligem, nur durch Zell-
theilung wachsendem Endosperm. Abh. Konigl. Sachs. Gesell. Wiss. 1859,
pp. 535-672.
1861. Neue Beitrage zur Kenntnis der Embryobildung der Phanero-
gamen. II. Monokotyledonen. Abh. Konigl. Sachs. Gesell. Wiss. 7:
629-760. Juel, H. O. 1898. Parthenogenesis bei Antennaria alpina (L) R.Br. Vorlaufige
Mittheilung. Bot. Centbl. 74: 369-372. . 1900. Vergleichende Untersuchungen iiber typische und partheno-
genetische Fortpflanzung bei der Gattung Antennaria. K. Svenska Vet.-
Akad. Handl. 33 (5) : 1-59. Kerner, A. 1876. Parthenogenesis bei einer angiospermen Pflanze. Sitzber.
Math., Nat. Kl. Akad. der Wiss. Wien 1, 74: 469. Kolreuter, J. C. 1761-1766. "Vorlaufige Nachricht von einigen das Geschlecht
der Pflanzen betreffenden Versuchen und Beobachtungen." Leeuwenhoek, A. v. 1677. "Observations de natis e semine genitali animal-
culis." . "The Secret Works of Antony van Leeuwenhoek, Containing his Micro- scopical Discoveries in Many of the Works of Nature." Engl, transl. by
Samuel Hoole. London, 1800.
26 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
Mottier, D. M. 1897. Tiber das Verhalten der Kerne bei der Entwicklung des
Embryosackes und die Vorgange bei der Befruchtung. Jahrb. f. wiss. Bot.
31: 125-158. Murbeck, S. 1897. Om vegetativ embryobildning hos flertalet Alchemillar och
den fiorklaring ofver formbestandigheten inom slagtet, som densamma innebar.
Bot. Notiser 1897, pp. 273-277. . 1901. Parthenogenetische Embryobildung in der Gattung Alchemilla.
Lunds Univ. Arsskr., Afd. II, 36(7): 1-41. Nawaschin, S. G. 1898a. tlber das Verhalten des Pollenschlauches bei der Ulme.
Bui. Acad. Imp. des Sci. St. Petersburg 8: 345-357. . 18986. Resultate einer Revision der Befruchtungs vorgange bei Lilium
martagon und Fritillaria tenella. Bui. Acad. Imp. des Sci. St. Petersburg 9 :
377-382. Pliny (Plinius Secundus, Gaius). "The Historie of the World: Commonly Called
the Naturall Historie of G. Plinius Secundus." Engl. Transl. London, 1635. Pringsheim, N. 1855. Uber die Befruchtung der Algen. Ber. Preuss. Akad. der
Wiss. Berlin 1855, pp. 133-165. — . 1856. Uber die Befruchtung und der Generationswechsel der Algen.
Monatsber. Konigl. Preuss. Akad. der Wiss. Berlin 1856, pp. 225-237. Radlkofer, L. 1856. "Die Befruchtung der Phanerogamen. Ein Beitrag zur
Entscheidung des dariiber bestehenden Streites." W. Engelmann, Leipzig. Sachs, J. 1874. "Lehrbuch der Botanik." Leipzig. Sargant, E. 1900. Recent work on the results of fertilization in angiosperms.
Ann. Bot. 22: 121-186. Schacht, H. 1850. "Entwicklungsgeschichte der Pflanzenembryo." Amsterdam. Schleiden, M. J. 1837. Einige Blicke auf die Entwicklungsgeschichte des vege-
tablischen Organismus bei den Phanerogamen. Arch. Bwl. Naturgeschichte
III, 1 : 289-320.
— . 1845. tlber Amicis letzten Beitrag zur Lehre von der Befruchtung der
Pflanzen. Flora: 593-600. Schnarf, K. 1929. "Embryologie der Angiospermen." Berlin.
— . 1931. "Vergleichende Embryologie der Angiospermen." Berlin. Sharp, L. W. 1943. "Fundamentals of Cytology." McGraw-Hill Book Com- pany. Strasburger, E. 1877. Uber Befruchtung und Zelltheilung. Jenaische Ztschr. f .
Naturw. 11: 435-536. -. 1878. Uber Polyembryonie. Jenaische Ztschr. f. Naturw. 12 : 647-670.
. 1879. "Die Angiospermen und die Gymnospermen." Jena.
— . 1884. "Neue Untersuchungen liber den Befruchtungsvorgang bei den Phanerogamen." Jena. . 1900. Einige Bemerkungen zur Frage nach der "doppelten Befruchtung"
bei Angiospermen. Bot. Ztg. II, 58: 293-316. Thuret, G. 1854. Recherches sur la fecondation des Fucacees, suivies d'observa-
tions sur les antheridies des algues. Ann. des Sci. Nat., Bot. 2: 196-214. Tretjakow, S. 1895. Die Beteilung der Antipoden in Fallen der Polyembryonie
bei Allium odorum. Ber. deutsch bot. Gesell. 13: 13-17.
HISTORICAL SKETCH 27
Tieub, M. 1879. Notes sur l'embryogenie de quelques Orchidees. Natuurk.
Verh. Koninkl. Akad. Amsterdam 19: 1-50. . 1891. Sur les Casuarinees et leur place dans le systeme naturel. Ann.
Jard. Bot. Buitenzorg 10: 145-231.
and Mellink, J. 1880. Notice sur le developpement du sac embryonnaire
dans quelques Angiospermes. Arch. Neerland. 15: 452-457. Von Mohl, H. 1863. Giambattista Amici. Bot. Ztg. 21 (Beilage 34) : l-\
CHAPTER 2
THE MICROSPORANGIUM
In considering the course of events leading to the origin of the embryo, we must first deal with the development of the micro- and megasporangia. It is the microsporangium which produces the mi- crospores and eventually the male gametophyte. Similarly, the megasporangium, or ovule, is the place of formation of the mega- spores and the female gametophyte. The latter, after fertilization,
Fig. 23. T.s. anther of Lilium philadelphicum, showing dissolution of cells sepa- rating the two microsporangia on each side. Note the fibrous endothecium and stomium s. The minute punctate markings lining the inner wall of the anther probably represent remnants of tapetum. (After Coulter and Chamberlain, 1903.)
produces the embryo and endosperm, while the entire megasporan- gium with its enclosed structures becomes the seed and the progen- itor of the next generation.
A typical anther comprises four elongated microsporangia, but at maturity the two sporangia of each side become confluent owing to the breaking down of the partition between them (Fig. 23). A cross section of a very young anther shows a mass of homogeneous meristematic cells surrounded by the epidermis (Fig. 24 A, B). It
28
THE MICROSPORANGIUM
29
soon becomes slightly four-lobed, and rows of hypodermal cells become differentiated in each lobe by their larger size, radial elon- gation, and more conspicuous nuclei. These form the archespor- ium. The extent of the archesporial tissue varies considerably both lengthwise and breadthwise. Either a single archesporial cell may be seen in each lobe in a cross section of the anther, as in Sanse- vieria (Guerin, 1927), Dionaea (Smith, 1929), and Boerhaavia (Ma- heshwari, 1929), or a plate of such cells, as in Ophiopogon (Mahesh- wari, 1934), Urginea (Capoor, 1937a), and most other plants. In longitudinal section also the row may comprise only one cell as in
Fig. 24. A-E, differentiation of parietal and sporogenous tissue in anthers of Chrysanthemum leucanthemum (e = epidermis; end = endothecium ; m = middle layer; t = tapetum; sp = sporogenous cell). (After Wanning, 1873.)
Enalus (Kausik, 1941), or two cells as in Boerhaavia (Maheshwari, 1929), or several cells as in Urginea (Capoor, 1937a).
Figure 24C-E shows the stages leading to the origin of the sporog- enous tissue. The archesporial cells divide to form a primary parietal layer toward the outside and a primary sporogenous layer toward the inside. The cells of the former divide by periclinal and anticlinal walls to give rise to a series of concentric layers, usually three to five, composing the wall of the anther. The primary sporog- enous cells either function directly as the spore mother cells or undergo further divisions to form a larger number of cells.
30 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
In a few plants a hypodermal archesporium has not been clearly distinguished and more deep-seated cells are said to give rise to the sporogenous tissue. In Doryanthes (Newman, 1928), Pholisma (Copeland, 1935), and Holoptelea (Capoor, 19376) it is stated that there is no definite system of periclinal divisions separating the parietal tissue from the archesporium and that the sporogenous function is gradually taken over by a group of cells about three or four layers below the epidermis. It is probable, however, that such appearances are due to the difficulty of obtaining a sharp differen- tiation between the cells during the early stages of development of the anther, and further studies may reveal the hypodermal origin of the archesporium in these plants also.
The Wall Layers. The epidermis, which is the outermost layer of the anther, undergoes only anticlinal divisions. Its cells become greatly stretched and flattened in order to keep pace with the en- largement of the anther, and in many plants, especially those of dry habitats, they eventually lose contact with each other so that only their withering remains can be seen at maturity.
The layer of cells lying immediately beneath the epidermis is the endothecium. Its maximum development is attained at the time when the pollen grains are about to be shed (Fig. 23). The cells become radially elongated, and from their inner tangential walls fibrous bands run upward, ending near the outer wall of each cell. In aquatics with aerial flowers like Utricularia (Kausik, 1938) and even such reduced forms as Wolffia (Gupta, 1935) the fibrous thick- enings occur as usual, but in several members of the Hydrocharita- ceae (Ernst-Schwarzenbach, 1945; Maheshwari and Johri, 1950), and in some cleistogamous forms whose flowers never open, they fail to develop and there is no special mode of dehiscence. In those plants, also, whose anthers open by apical pores, the endothecium may not develop any fibrous thickenings and dehiscence takes place here by the dissolution of certain cells at the apex of the anther. In Erica, which is an example of this kind, there is a further pecu- liarity in that the "apical" pores are in fact basal. Figure 25 shows some stages in the curvature of the anther which bring about this inversion (Matthews and Taylor, 1926).
Among other exceptions may be cited Musa (Juliano and Alcala, 1933), Sesamum (Nohara, 1934), Anona (Juliano, 1935a), Ipomoea, (Juliano, 19356), Aeginetia (Juliano, 1935c), and Melastoma (Subra-
THE MICROSPORANGIUM
31
manyam, 1948) in which the fibrous thickening are absent but the walls of the epidermal cells undergo a general cutinization and lignification over the entire surface. Oryza (Juliano and Aldama, 1937), Ditepalanthus (Fagerlind, 1938), and Balanophora (Fager- lind, 1945) are peculiar in that the parietal layers, one or two in number, become crushed and disorganized during the development of the anther so that a fibrous layer is absent and the epidermis abuts directly on the tapetum.1
A
Fig. 25. Development of anther of Erica hirtiflora. A, B, l.s. young stamens, showing gradual inversion of anther. C, l.s. stamen at spore mother cell stage, showing almost complete inversion of anther, so that its lower end comes to lie toward the upper side. (After Matthews and Taylor, 1926.)
Next to the endothecium there are usually one to three "middle" layers. As a rule, all of them become flattened and crushed at the time of the meiotic divisions in the microspore mother cells, but there are a few exceptions. In Holoptelea (Capoor, 19376) there are three to four middle layers, of which the outermost persists for a long time. In Ranunculus (Singh, 1936) there are two middle layers, of which the inner soon disappears but the outer persists;
1 In Styphelia (Brough, 1924), Arceuthobium (Pisek, 1924) and some members of the Ericales it is the epidermis which is said to develop fibrous thickenings and function as an endothecium, but this deserves confirmation.
32 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
occasionally its cells become densely protoplasmic and simulate those of the tapetum. In Lilium there are several middle layers, of which those lying adjacent to the endothecium persist for a long time (Fig. 23), and in Gloriosa (Eunus, 1949) the outermost middle layer develops fibrous thickenings similar to these of the endo- thecium.
Rarely, a middle layer may be absent as in the anthers of Wolffia (Gupta, 1935) and Vallisneria (Witmer, 1937), but some previous reports of the absence of a middle layer have been shown to be mis-
A B
Fig. 26. Anthers, showing microspore mother cells and tapetum. A, Bougain- villea, t.s. portion of anther, showing mitotic divisions in tapetal cells. (After Cooper, 1931.) B, Salvia mellifera, t.s. portion of anther lobe. The tapetal cells lying toward the connective are considerably larger than those on the outer side. (After Carlson and Stuart, 1936.)
interpretations caused by its ephemeral nature and early disappear- ance. Johri (1934) has demonstrated the presence of a middle layer in Cuscuta where it was formerly reported to be absent (Peters, 1908).
The innermost wall layer or tapetum is of considerable physio- logical significance, for all the food materials entering into the sporog- enous cells must pass through it.2 Its cells are full of dense cyto- plasm, and at the beginning of meiosis the tapetal nuclei may also undergo some divisions (Fig. 26). 3 Because of these similarities of
2 Typically the tapetum is a single layer of cells but in Nicolaia and Costus (Boehm, 1931) it is composed of several layers.
3 Rarely, tapetal nuclei may even pass through a condition resembling the prophase of a meiotic division. Gates and Rees (1921) figure some tapetal
THE MICROSPORANGIUM
33
appearance and behavior between the cells of the tapetum and the micrcsporogenous tissue, earlier botanists supposed that the former is derived by a sterilization of the outer sporogenous cells. Develop- mental studies of a precise nature have, however, nearly always confirmed its parietal origin.4
The nuclear divisions in the tapetum were formerly believed to be amitotic, but recent studies (Bonnet, 1912; Cooper, 1933; Wit- kus, 1945) have shown that this is incorrect and that appearances suggesting amitosis are really caused by mitotic irregularities and
G H I J K
Fig. 27. Nuclear divisions in tapetal cells of Zea mays (A-F), Lilium canadense (G-H), and Podophyllum peltatum (I-K). (After Cooper, 1933.)
nuclear fusions. According to present conceptions, the nucleus of a tapetal cell may divide in any of the following ways:5
1. By normal mitosis. The division takes place in the ordinary
nuclei of Lactuca in the synizesis stage, and Moissl (1941) reports a similar condi- tion in some members of the Caprifoliaceae.
4 Recently, Capoor (19376) has reported that in Holoptelea the tapetal cells are almost indistinguishable from the adjacent cells of the sporogenous tissue. He cautiously adds, however, that this fact alone is insufficient to justify any inference regarding the sporogenous origin of the tapetum.
6 It is to be noted that in a few families and orders, viz., Mimosaceae, Cras- sulaceae, Gentianaceae, Boraginaceae, Hydrophyllaceae, Juncaceae, Orchidaceae, and Helobiales, the tapetal cells usually remain uninucleate from the time of their formation to their eventual disintegration.
84
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
way, but no cell plate is laid down. The two daughter nuclei, which are diploid, remain inside the cell (Fig. 27 A-F).
2. By a "sticky" type of division. Here the chromosomes behave normally up to the early anaphase stage. After this, one or more of them fail to separate, forming chromosome bridges which persist during the telophase as well as the resting stage. As a result a single dumbbell-shaped tetraploid nucleus is formed whose middle portion may be broad or narrow depending on the number of chromosome bridges present (Fig. 27 G-K).
3. By endomitosis.6 Here the nucleolus and the nuclear mem- brane remain intact and there is no spindle formation. The chro-
Fig. 28. Diagrams showing "endomitosis" in tapetal cells of Spinacia oleracea. A, endoprophase. B, endometaphase. C, endo-anaphase. D, endotelo phase. (Drawing supplied by Dr. E. R. Witkus.)
mosomes contract and split longitudinally, but all of them remain within the same nucleus, which becomes tetraploid (Fig. 28).
The first nuclear division in a tapetal cell is often followed by further divisions. Some of the divisions may be accompanied by nuclear fusions, resulting in one or more large polyploid nuclei. The latter may, however, divide again and give rise to smaller nu- clei. Since this type of behavior is very frequent in tapetal cells, it is unnecessary to give specific instances.
An interesting condition has been reported in certain haploid and
6 This type of division was first postulated by Meyer (1925). In the tapetal cells of Leontodon he found diploid nuclei in younger stages and polyploid nuclei in older stages. Since no spindle fibers were observed, he concluded that there was an "internal division" of the chromosomes without any nuclear division. See also Brown (1949) who has recently given a detailed account of endomitosis in the tapetal cells of tomato.
THE MICROSPORANGIUM 35
diploid plants of Oenothera rubricalyx (Gates and Goodwin, 1930). In the former the tapetal cells are uninucleate and in the latter they are binucleate — a fact which is no doubt related to the general reduction of tissues in haploid individuals. More difficult to ex- plain is the marked difference in shape and structure of the tapetal cells belonging to the same anther. In Lathraea (Gates and Latter, 1927), Salvia (Carlson and Stuart, 1936) (Fig. 265), and Moringa (Puri, 1941) the tapetal cells on the inner side of the loculus show a marked radial elongation and are much larger than those on the outer side. Further, in Lathraea the cells on the outer side are uni- nucleate while those adjacent to the connective are binucleate. In Lactuca sativa (Gates and Rees, 1921) the tapetal cells lying on one side of the loculus may be quadrinucleate while those on the other are binucleate. The binucleate cells are nearly always shorter and broader than the quadrinucleate. Possibly these differences are related to the varying amounts of nutritive materials passing into the cells.
Toward the close of the meiotic divisions in the microspore mother cells, the tapetal cells begin to lose contact with each other. Large vacuoles appear in the cytoplasm and the nuclei begin to show signs of degeneration.7 Finally the cells are entirely absorbed at the time when the microspores begin to separate from one another. This type of tapetum, in which the cells remain in situ, is called the glandular or secretory tapetum and is of common occurrence in angiosperms. However, there are several genera and families (see Juel, 1915; Tischler, 1915; Mascre, 1919 a, b) in which the walls of the tapetal cells break down but the protoplasts, which remain intact, protrude and "wander" inside the loculus, where they may coalesce to form a continuous mass called the tapetal periplasmodium (Fig. 29). Clausen (1927), who has reviewed the previous literature in this connection, classifies this kind of tapetum (often called the "amoeboid" tapetum) into four subtypes:
1. Sagittaria type. The tapetal cells lose their walls by the time the microspore tetrads have been formed, and their protoplasts begin to project inward as soon as the microspores have separated. Later the periplasmodium becomes continuous. Examples: Sagit- taria, Alisma, Limnocharis, Hydrocharis.
7 At this stage the anther loculi frequently show a densely staining jelly-like or mucilaginous fluid which disappears at maturity. As suggested by Nietsch (1941), this is probably a secretion from the tapetal cells.
36
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
2. Butomus type. In this case the formation of the periplasmo- dium occurs a little earlier, when the microspores are still grouped in tetrads. Examples: Butomus, Stratiotes, and Ouvirandra.
3. Sparganium type. Here also the fusion of protoplasts begins at the tetrad stage but the tapetal cells are multinucleate. Ex- amples: Sparganium, Typha, Tradescantia.
4. Triglochin type. In a few plants the tapetum begins its activ- ity while the microspore mother cells are still undergoing the meio- tic divisions. The tapetal protoplasts and nuclei protrude into the
\.:-v..i.
Fig. 29. Tapetal Plasmodium in Symphoricarpos racemosm (A) and Lonicera pyrenaica (B). (After Moissl, 1941.)
spaces between the mother cells so that the periplasmodium is formed at a very early stage. Examples: Triglochin, Potamogeton, and several members of the Araceae.
Like the glandular or secretory tapetum, the amoeboid tapetum also serves for the nutrition of the spores, and is probably more effective for this purpose. As some authors (see Mezzetti-Bamba- cioni, 1941) have suggested, it seems probable that the periplas- modium contributes to the formation of the exine, but this point deserves further study. A curious feature which has been observed in several plants (see Ubisch, 1927; Kosmath, 1927; Kajale, 1940; Puri, 1941; Singh, 1950) is the appearance of small granular mark-
THE MICROSPORANGIUM
37
ings on the inner surface of the tapetum (Fig. 30) and later on the inner surface of the middle layers or the endothecium. They give the same staining reactions as the exine of the pollen grains and prob- ably contribute to the development of the latter. This seems to be supported by Gorczynski's (1934) observations on Cardamine, ac- cording to which the exine first begins to develop on that side of the pollen grains which lies towards the tapetum.
Fig. 30. Tapetal cells, showing cutinization of inner walls. A, Magnolia youlan, tapetal cell, showing prominent thickenings on inner surface. B, the thickenings as seen in surface view. C, Lilium tigrinum, thickenings on inner walls of tapetal cells. D, same in surface view. E, more highly magnified than D. (After Kos- math, 1927.)
Sporogenous Tissue. The primary sporogenous cells give rise to the microspore mother cells. In some plants the sporogenous cells undergo several divisions, in others only a few divisions, and rarely there are no divisions at all, so that the primary sporogenous cells function directly as the microspore mother cells. Alangium, Sansevieria, Knautia, and some members of the Malvaceae and Cucurbitaceae are examples of the third kind, showing a single row of microspore mother cells in each anther lobe.
38 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
A peculiar feature met with in some members of the Mimosaceae is the development of transversely placed sterile septa in the anther lobes (Fig. 31). In some members of the Loranthaceae also, viz., Dendrophthoe (Rauch, 1936), Elytranthe, and Amyema (Schaeppi
Fig. 31. Structure of anther in some members of the Mimosaceae. A, Parkia, l.s. anther showing two rows of pollinia. B,C, pollinia dissected out from anther. D, Dichrostachys, l.s. anther, showing pollinia; note stalked gland gl at apex of anther. (After Engler, 1876.)
and Steindl, 1942), the microsporangia become vertically parti- tioned by the formation of sterile septa, and in Viscum (Schaeppi and Steindl, 1945) such partitions arise not only in the vertical plane but also in the horizontal one so that each anther has as many as 50 loculi.
THE MICROSPORANGIUM 39
Formation of sterile septa is also known in a few other plants. Caldwell (1899) reports that in Lemna the archesporial tissue orig- inally comprises a single mass of cells. After the usual wall layers have been cut off, a plate of sterile cells divides this mass into two and then into four. In Limnophyton (Johri, 1935) and Ranunculus (Singh, 1936) a cross section of the young anther shows an oval or somewhat dumbbell-shaped outline with a plate of archesporial cells on each side. Both of these become partitioned by the ap- pearance of a sterile septum resulting in the usual tetralocular con- dition. In Quamoclit (Fedortschuk, 1932) there is a single row of sporogenous cells in each lobe of the anther but one or two of these fail to keep pace with the others and become nonfunctional. These give rise to sterile partitions separating the loculus into two or three parts.
In some plants there are fewer than four groups of sporogenous cells. In the family Malvaceae (Stenar, 1925) the anthers are uni- formly bisporangiate and the two loculi eventually fuse to form a single loculus. In Elodea (Wylie, 1904), Styphelia (Brough, 1924), Circaeaster (Junell, 1931), Phoradendron (Billings, 1932), Wolffia (Gupta, 1935), and Moringa (Puri, 1941) also, there are two micro- sporangia which may later become confluent by the breaking down of the intervening cell layers. The anthers of Naias (Campbell, 1897) are said to be unilocular, but the developmental stages have not been traced satisfactorily. In Vallisneria (Witmer, 1937) there are all gradations from a unilocular to a tetralocular condition. Typically two loculi are formed, owing to the appearance of a sterile septum in the sporogenous tissue, but sometimes the septum is incomplete, resulting in a unilocular condition, and frequently each of the two loculi becomes bisected so as to form four loculi.
The stamens of Piper betle (Johnson, 1910) are peculiar in that the number of microsporangia in an anther may be four, three, two, or one, and it remains constant from the time of initiation of the sporangia to the maturation of the anther. There is no secondary fusion of the sporogenous tissue.
In Korthalsella (Stevenson, 1934; Rutishauser, 1935) there are three stamens, each of which consists of two microsporangia, but since all the anthers fuse to form a synandrium, a cross section of the flower shows six microsporangia arranged in a ring. At matur-
40 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
ity the partitions between the sporangia break down and the loculi become continuous.
The genus Arceuthobium is unique in having a single annular pollen sac forming a continuous ring around a central column of sterile cells called "columella" (Stadtler, 1923; Fisek, 1924; Thoday and Johnson, 1930; Dowding, 1931). Regarding the origin of this condition there is, however, some difference of opinion. Stadtler (1923) thinks that the anther is at first multilocular but the par- titions break down at maturity. Pisek (1924), on the other hand, contends that it is unilocular from the commencement, and this is supported by Thoday and Johnson (1930) who state that even in the youngest anthers there is a ring-shaped archesporium surround- ing the central columella. Dowding (1931) agrees regarding the continuity of the archesporium but finds that the columella exhibits a considerable amount of variation. It frequently forms a sort of flange dividing the anther into two halves; sometimes the first flange tends to disappear, and a new one arises at right angles to it. Rarely, the flanges give out branches extending outwards to the anther wall. In Dowding 's opinion these flanges of the columella are to be regarded as remnants of the septa which once separated four distinct archesporia.
Although all the sporogenous cells in the anther are potentially capable of giving rise to microspores, some of them frequently de- generate and become absorbed by the remaining cells. In Ophiopo- gon (Maheshwari, 1934) and Holoptelea (Capoor, 19376) some of the sporogenous cells do not reach even the mother cell stage and prob- ably serve to nourish the remaining cells. In Zoster a (Rosenberg, 1901) most of the sporogenous cells divide longitudinally to form the numerous long microspore mother cells but others interspersed be- tween them undergo transverse divisions and give rise to sterile cells which are later crushed and used up by the functioning cells. In certain members of the Gentianaceae (Guerin, 1926) which are devoid of any well-formed tapetum, the nutritive function is taken over by some of the sporogenous cells themselves. These become sterile and do not go through the reduction divisions (Fig. 32). In Kigelia (Venkatasubban, 1945) degeneration takes place at a later stage; some of the microspores in a tretrad fail to develop fur- ther and become functionless.
Cytomixis. While making a study of Oenothera gigas and 0.
THE MICROSPORANGIUM
41
biennis, Gates (1911) observed a frequent migration of chromatic material from one microspore mother cell into another and called it cytomixis. Since then it has been reported in several other plants, and while it is most frequent between the synizesis and diakinesis stages, it may sometimes occur even during the interkinesis stage, i.e., after the first meiotic division has been completed. In Lathraea (Gates and Latter, 1927), which is an instance of this kind, the microspore mother cells do not round up but remain in close contact with one another. During interkinesis the nuclei of the two dyad cells occupy an eccentric position near the cell wall so that the
A B
Fig. 32. Sterilization of part of sporogenous tissue in anthers of Sivertia perennis. A, anther lobe at microspore mother cell stage. B, same, at microspore tetrad stage. {After Guerin, 1926.)
chances of cytomixis are increased. In Coreopsis tripteris (Gelin, 1934) cytomixis may also occur at the close of the meiotic divisions but the multinucleate cells formed in this way again break up into smaller units consisting of one or two nuclei.
In some plants individual chromosomes, or groups of chromo- somes, or even whole spindles are said to be carried from one cell into another. It is believed, however, that it is a pathological phenomenon, or that such appearances are caused by faulty fixa- tion. Woodworth (1931), who used smear preparations of anthers, states that cytomixis was common when a little extra pressure was used in squeezing out the microspore mother cells. Further, such abnormalities were found to be more frequent in hybrids than in other plants, and he attributes this to an "innate unbalance" in the
42 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
heterozygous cytoplasm which makes it more susceptible to pressures and other similar treatments.
Mention may also be made here of fusions of entire cells of the sporogenous tissue. Matsura (1935) reported that in Phacellan- thus the separating walls between adjacent microspore mother cells sometimes dissolve and fuse in pairs to form giant cells, which may either give rise to polyploid gametes or degenerate without com- pleting the meiotic divisions. In two haploid plants of Phleum pratense, Levan (1941) observed the fusion of as many as 30 micro- spore mother cells, giving rise to large plasmodia or "syncytes." A similar behavior has also been reported by Stern (1946) in sugar suspensions of the microspore mother cells of Trillium erectum. Here the extent of the fusions appeared to be unlimited, although the maximum number of nuclei actually observed in a cell was 32.
Cytokinesis. The divisions of the microspore mother cells may be of the successive or the simultaneous type.8 In the former a cell plate is laid down immediately after the first meiotic division and another in each of the two daughter cells after the second meio- tic division. In the simultaneous type, on the other hand, no wall is laid down after the first division and the mother cell becomes separated all at once into four parts after both the meiotic divisions are over.
The investigations of C. H. Farr (1916) and others have shown that there is also another difference in the mechanism of cytokine- sis. In the successive type the cell plate is laid down in the center and then extends centrifugally on both sides, dividing the cell into two equal halves. In the simultaneous type, on the other hand, the division usually occurs by centripetally advancing constriction fur- rows, which meet in the center and divide the mother cell into four parts.
Farr (1916) studied Nicotiana tdbacum in special detail. At first there is an enlargement of the nucleus of the microspore mother cell, accompanied by a thickening of the mother cell wall. No cell plate is laid down after Meiosis I, and the spindle fibers of this divi- sion disappear during the metaphases of Meiosis II. After the four daughter nuclei have become organized, they assume a tetrahedral arrangement and a spindle is re-formed between every two nuclei,
8 For an account of the nuclear changes in meiosis, see Sharp (1943) and other vvorks on cytology
THE MICROSPORANGIUM 43
making a total of six spindles. However, these spindles have noth- ing to do with the quadripartition of the mother cell, and there is no laying down of centrifugally growing cell plates such as are characteristic of other dividing cells. Instead, constriction furrows now start at the periphery and proceed inward until they meet at the center, so that there is a simultaneous division of the protoplast into four cells, i.e., the microspores.
In Melilotus alba (Castetter, 1925) vacuoles seem to play a con- spicuous part in cytokinesis (Fig. 33). After Meiosis II, hyaline areas develop between the four nuclei, apparently as the result of a migration of the denser cytoplasm toward the nuclei and an ex- trusion of sap into the regions between them. The small vacuoles arising in this manner soon fuse to form larger ones which virtually split the cytoplasm into four masses. Furrows originating at the surface now grow inward and soon meet the vacuoles. Meanwhile, the mother cell rounds up and secretes a thick layer of callose or some other gelatinous material, which extends inward with the cleavage furrows and eventually completes the division of the cell into the four microspores.9
Zea mays (Reeves, 1928) may be taken as an example of the suc- cessive type of microspore formation (Fig. 34). At the end of Meiosis I, thickenings are formed on the spindle fibers at the equatorial region of the cell. They gradually increase in size, coming in con- tact with each other and fusing to form the cell plate. Additional spindle fibers continue to appear just beyond the periphery of the plate so as to increase the diameter of the spindle. At the same time the cell plate extends centrifugally and joins the wall of the mother cell, so as to complete the division of the protoplast into two halves. Now the second meiotic division follows, and a new partition wall develops in each cell in the same way as after Meiosis I, resulting in a tetrad showing the bilateral arrangement of microspores.
The question as to which of the two modes of tetrad formation is primitive and which is the more advanced is difficult to decide. It seems, however, that since a division by furrowing is common
9 The mode of origin of this gelatinous layer has been a subject of much discus- sion. Beer (1906), Gates (1925), and Castetter (1925) have expressed the view that it is secreted by the cytoplasm of the mother cell, while Farr (1922), Bowers (1931), and Capoor (1937a) believe that it is the result of a swelling of the secondary lamellae of the cell wall.
44
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
Fig. 33. Cytokinesis in microspore mother cells of Melilotus alba. A, telophase of Meiosis I. B, metaphase of Meiosis II. C, end of Meiosis II, showing three of the four microspore nuclei. Note large extranuclear "centrosome-like bodies" seen here and in A. D, microspore nuclei in resting stage. Spindles have almost disappeared and protoplast has begun to invaginate at the periphery, at points equidistant from the nuclei. Note origin of special wall, shown in black. E,F, formation cf vacuoles in portions of cytoplasm lying between nuclei. G, special wall entering furrows. //, special walls have met in center, forming partitions be- tween microspore nuclei. I, fully formed microspores. (After Castetter, 1925.)
THE MICROSPORANGIUM
45
in the Thallophytes and other lower plants, the simultaneous type is the more ancient and the successive type the derived. In gen- eral, the former is prevalent in the majority of dicotyledons and the latter in the majority of monocotyledons. There is no hard and fast rule, however, and exceptions are frequent. Thus the suc- cessive type is found in a few dicotyledonous families like the Asclepiadaceae, Podostemonaceae, and Apocynaceae, and the simul- taneous type in a few monocotyledonous families, viz., the Iridaceae, Taccaceae, Juncaceae, and Dioscoreaceae, and in several genera of the Liliaceae, Palmaceae, and Orchidaceae.
In Magnolia (Farr, 1918) there are isobilateral tetrads formed by furrowing instead of by cell plates. A cleavage furrow starts after
A B C D E
Fig. 34. Cytokinesis in microspore mother cells of Zea mays. A, anaphase of
Meiosis I. B,C, laying down of partition wall after Meiosis I. D, telophase of Meiosis II. E, isobilateral tetrad. {After Reeves, 1928.)
Meiosis I, but its development is arrested during the second meiotic division. It resumes growth at the end of Meiosis II and forms a partition through the equatorial region of the mother cell. At the same time additional furrows originate at the periphery, and the two dyad cells now become subdivided to give rise to the four microspores. A similar condition occurs in Anona (Juliano, 1935a) and Asimina (Locke, 1936).
The Microspore Tetrad. As mentioned above, the microspores are usually arranged in a tetrahedral (Fig. 35 A) or isobilateral (Fig. 35B) fashion, but there are exceptions (Fig. 35C-E). A decussate arrangement of the cells has been recorded in Magnolia (Farr, 1918), Atriplex (Billings, 1934), Comas (D'Amato, 1946), and many other plants. In some genera of the Asclepiadaceae (Gager, 1902) and in the genus Halophila of the Hydrocharitaceae (Kausik and Rao, 1942) the mother cells divide transversely so as to give rise to
46
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
linear tetrads (Fig. 36A-D). T-shaped tetrads also occur some- times as in Aristolochia (Samuelsson, 1914) and Butomopsis (Johri, 1936). In Zostera (Rosenberg, 1901) the elongated microspore mother cells, measuring 5 by 60 microns at the time of meiosis, divide in a plane parallel to the longitudinal axis of the cell, result- ing in a group of four filiform cells which undergo further elongation and become approximately 2000 microns long when mature.10 Of considerable interest are Musa (Juliano and Alcala, 1933), Neottia (Goebel, 1933), Agave (Vignoli, 1936, 1937), Nicolaia (Boehm, 1931), Habenaria (Swamy, 1946), Laurus (Battaglia, 1947), and Ottelia (Islam, 1950) in which two or three types of dispositions may be found in one and the same species.
Occasionally there are either fewer than four spores resulting from the divisions of the microspore mother cell, or more than four.
A 'B " — ^C ^-^D ^^E
Fig. 35. Diagram showing different types of microspore tetrads. A, tetrahedral. B, isobilateral. C, decussate. Z),T-shaped. ^linear. (B-E, ajter\ Boehm, 1931.)
The former condition originates as the result of a failure of one division, or the formation of a "restitution nucleus" after the first division, or an irregular wall formation giving rise to one binucleate and two uninucleate spores. The latter condition, i.e., the forma- tion of more than four spores (polyspory), usually results from the occurrence of lagging chromosomes which organize into micronuclei. In general, however, such abnormalities in the number of micro- spores are found only in hybrids characterized by a high degree of sterility and the pollen grains arising in this way are nonfunctional. Usually the microspores soon separate from one another but in some plants they adhere in tetrads to form the so-called "com- pound" pollen grains.11 As examples may be cited Drimys, Anona,
10 Filiform pollen grains also occur in Phyllospadix and Cymodocea, but the method by which they arise does not seem to have been studied so far.
11 For detailed information on such variations of external form, see Wodehouse (1936) and Erdtman (1943, 1945).
THE MICROSPORANGIUM
47
E F G H I
Fig. 36. Development of microspores and male gametophyte of Halophila ovata. A, l.s. of young staminate flower. B, microspore mother cells with a few tapetal cells t. C, chains of microspores; note vacant spaces x separating individual tetrads. D, single microspore. E, microspore, showing tube and generative cells. F, older stage, showing spindle-shaped generative cell lying inside vegeta- tive cytoplasm. G, pollen grain, showing division of generative cell. H , same, more advanced stage. Note formation of constriction furrow across generative cell. I,J, formation of sperm cells completed. K,L, division of generative cell, showing formation of transitory cell plate. (After Kausik and Rao, 194%.)
48 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
Drosera, Elodea, Typha, Furcraea, and several members of the Eri- caceae, Apocynaceae, Asclepiadaceae, Juncaceae, and Orchidaceae. In the Mimosaceae there are larger units composed of 8 to 64 cells, and in a number of genera belonging to the Asclepiadaceae all the microspores in a sporangium remain together to form a single mass called the pollinium. The family Orchidaceae is especially interest- ing in this connection (see Swamy, 1948). In some genera, such as Cypripedium and Vanilla, the microspores separate from one another and become free. In Pogonia the four cells of a tetrad adhere and form a compound pollen grain. In the tribes Ophrydeae and Neot- tieae this tendency is carried further and the compound grains are themselves held together in small units known as massulae. Fi- nally, in Coelogyne and Pholidota all the microspore mother cells and their derivatives remain together and continue their development as a single unit.
References
Battaglia, E. 1947. Meiosi anormale nella microsporogenesi di Laurus nobilis L.
Atti d. Societa Toscana Sci. Nat. 54: 1-22. Beer, R. 1906. On the development of the pollen grain and anther of some
Onagraceae. Beihefte bot. Centbl. 19A: 286-313. Billings, F. H. 1932. Microsporogenesis in Phoradendron. Ann. Bot. 46: 979-
992. . 1934. Male gametophyte of Atriplex hymenelytra. Bot. Gaz. 95:
477-484. Boehm, K. 1931. Embryologische Untersuchungen an Zingiberaceen. Planta
14: 411-440. Bonnet, F. 1912. Recherches sur revolution des cellules nourricieres du pollen
chez les Angiospermes. Arch. f. Zellforsch. 7: 605-722. Bowers, C. G. 1931. The development of pollen and viscin strands in Rhododen- dron catawbiense. Bui. Torrey Bot. Club 57: 285-314. Brough, P. 1924. Studies in the Epacridaceae. I. The life history of Styphelia
longifolia. Proc. Linn. Soc. N. S. Wales 49: 162-178. Brown, S. W. 1949. Endomitosis in the tapetum of tomato. Amer. Jour. Bot.
36: 703-716. Caldwell, 0. W. 1899. On the life history of Lemna minor. Bot. Gaz. 27:
37-66. Campbell, D. H. 1897. A morphological study of Naias and Zannichellia. Proc.
Calif. Acad. Sci. III. Bot, 1: 1-71. Capoor, S. P. 1937a. Contributions to the morphology of some Indian Liliaceae.
II. The gametophytes of Urginea indica Kunth. Beihefte bot. Centbl. 57A:
156-170.
THE MICROSPORANGIUM 49
Capoor, S. P. 19376. The life history of Holoptelea integrifolia Planch. Beihefte
bot. Centlbl. 57A: 233-249. Carlson, E. M., and Stuart, B. C. 1936. Development of spores and gameto-
phytes in certain New World species of Salvia. New Phytol. 35: 68-91. Castetter, E. F. 1925. Studies on the comparative cytology of the annual and
biennial varieties of Melilotus alba. Amer. Jour. Bot. 12: 270-286. Claussen, P. 1927. Uber das Verhalten des Antherentapetums bei einigen Mono-
cotylen und Ranales. Bot. Arch. 18 : 1-27. Cooper, D. C. 1931. Microsporogenesis in Bougainvillaea glabra. Amer. Jour.
Bot. 18: 337-358. . 1933. Nuclear divisions in the tapetal cells of certain angiosperms.
Amer. Jour. Bot. 20: 358-364. Copeland, H. F. 1935. The structure of the flower of Pholisma arenarium.
Amer. Jour. Bot. 22: 366-383. Coulter, J. M., and Chamberlain, C. J. 1903. "Morphology of Angiosperms."
New York. D'Amato, F. 1946. Osservazioni cito-embryologiche su Cornus mas L. con parti-
colare riguardo alia sterilita di un biotipo triploide. Nuovo Gior. Bot. Ital.
N.S. 53: 170-210. Dowding, E. S. 1931. Floral morphology of Arceuthobium americanum. Bot.
Gaz. 91 : 42-54. Engler, A. 1876. Beitrage zur Kenntnis der Antherenbildung der Metaspermen.
Jahrb. f. wiss. Bot. 10: 275-316. Erdtman, G. 1943. "An Introduction to Pollen Analysis." Chronica Botanica
Company. . 1945. Pollen morphology and plant taxonomy. V. On the occurrence
of tetrads and dyads. Svensk Bot. Tidskr. 39:286-297. Ernst-Schwarzenbach, M. 1945. Zur Blutenbiologie einiger Hydrocharitaceen.
Ber. schweiz. bot. Gesell. 55: 33-69. Eunus, A. M. 1949. Contributions to the embryology of the Liliaceae (Gloriosa
superba). Proc. 36th Indian Sci. Cong. Allahabad, Bot. Sect. No. 25. Fagerlind, F. 1938. Ditepalanthus, eine neue Balanophoraceen Gattung aus
Madagaskar. Arkiv for Bot. 29A: 1-15. . 1945. Blute und Blutenstand der Gattung Balanophora. Bot. Notiser
1945, pp. 330-350. Farr, C. H. 1916. Cytokinesis of the pollen mother cells of certain dicotyledons.
Mem. N. Y. Bot. Gard. 6: 253-317. . 1918. Cell division by furrowing in Magnolia. Amer. Jour. Bot. 5:
379-395. . 1922. Quadripartition by furrowing in Sisyrinchium. Bui. Torrey Bot.
Club 49: 51-61. Fedortschuk, W. 1932. Entwicklung und Bau des mannlichen Gametophyten
bei den Artcn der Convolvulaceen-Gattung Qua?noclit. Planta 16: 554-574. Gager, C. S. 1902. The development of the pollinium and sperm cells in
Asclepias cornuti Decne. Ann. Bot. 16: 123-148.
50 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
Gates, R. R. 1911. Pollen formation in Oenothera gigas. Ann. Bot. 25 : 909-940 . 1925. Pollen tetrad wall formation in Lathraea. Cellule 35 : 47-60.
and Goodwin, K. M. 1930. A new haploid Oenothera, with some con- siderations on haploidy in plants and animals. Jour. Genet. 23: 123-156.
and Latter, J. 1927. Observations on the pollen development of two
species of Lathraea. Jour. Roy. Micros. Soc. 1927, pp. 209-224.
and Rees, E. M. 1921. A cytological study of pollen development in
Lactuca. Ann. Bot. 16: 123-148. Gelin, O. E. V. 1934. Embryologische und cytologische Studien in Heliantheae-
Coreopsidinae. Acta Horti Bergiani 11: 99-123. Goebel, K. 1933. "Organographie der Pflanzen. III. Samenpflanzen." 3d ed.
Jena. Gorczyhski, T. 1934. Zytologische Analyse einiger Pollenentwicklungs vorgange
bei der Apfelsorte "Schoner von Boskoop." Acta Soc. Bot. Poloniae 11: 103-
118. Guerin, P. 1926. Le developpement de l'anthere chez les Gentianacees. Bui.
Soc. Bot. de France 73: 5-18. . 1927. Le developpement de l'anthere et du pollen chez les Liliacees
(Sansevieria, Ophiopogon, Peliosanthes) . Bui. Soc. Bot. de France 74:
102-107. Gupta, B. L. 1935. Studies in the development of the pollen grain and embryo
sac of Wolffia arrhiza. Current Sci. [India] 4: 104-105. Islam, A. S. 1950. The embryology of Ottelia alismoides Pers. Jour. Indian
Bot. Soc. 29: 79-91. Johnson, D. S. 1910. Studies in the development of Piperaceae. I. The sup- pression and extension of sporogenous tissue in the flower of Piper betle L. var.
monoicum C. DC. Jour. Expt. Zool. 9: 715-749. Johri, B. M. 1934. The development of the male and female gametophytes in
Cuscuta rejlexa Roxb. Proc. Indian Acad. Sci. Sect. B. 1 : 2S3-239. . 1935. Studies in the family Alismaceae. I. Limnophyton obtusifolium
Miq. Jour. Indian Bot. Soc. 14: 49-66.
1936. The life history of Butomopsis lanceolata Kunth. Proc. Indian
Acad. Sci. Sect. B. 4: 139-162. Juel, H. O. 1915. Untersuchung liber die Auflosung der Tapetenzellen in den
Pollensacken der Angiospermen. Jahrb. f. wiss. Bot. 58: 337-364. Juliano, J. B. 1935a. Morphological contribution on the genus Anona. Philip- pine Agr. 24: 528-541. . 19356. Morphology of the sweet potato, Ipomoea batatus (Linn.) Poir.
Philippine Agr. 23: 833-858. . 1935c. Anatomy and morphology of the Bunga Aeginetia indica L.
Philippine Jour. Sci. 56: 405-451. and Alcala, P. E. 1933. Floral morphology of Musa errans (Blanco)
Teodoro var. Botoan Teodoro. Philippine Agr. 22: 91-126.
— and Aldama, M. J. 1937. Morphology of Oryza saliva L. Philippine
Agr. 26: 1-134.
THE MICROSPORANGIUM 51
Junell, S. 1931. Die Entwicklungsgeschichte von Circaeaster agrestis. Svensk
Bot. Tidskr. 25:238-270. Kajale, L. B. 1940. A contribution to the embryology of the Amaranthaceae.
Proc. Natl. Inst. Sci. India 6: 597-625. Kausik, S. B. 1938. Pollen development and seed formation in Utricularia
coerulea L. Beihefte bot, Centbl. 58A : 365-378. . 1941. Structure and development of the staminate flower and male
gametophyte of Enalus acoroides (L. fil) Steud. Proc. Indian Acad. Sci.
Sect. B. 14: 1-16. and Rao, P. V. K. 1942. The male gametophyte of Halophila ovata
Gaudich. Jour. Mysore Univ. Sect. B. 3: 43-49. Kosmath, L. 1927. Studien liber das Antherentapetum. Osterr. bot. Ztschr.
76: 235-241. Levan, A. 1941. Syncyte formation in the pollen mother cells of haploid Phleum
pratense. Hereditas 27: 243-252. Locke, J. F. 1936. Microsporogenesis and cytokinesis in Asimina triloba. Bot.
Gaz. 98: 159-168. Maheshwari, P. 1929. Contributions to the morphology of Boerhaavia diffusa.
I. Jour. Indian Bot. Soc. 8: 219-234. . 1934. Contributions to the morphology of some Indian Liliaceae. I.
The gametophytes of Ophiopogon wallichianus Hook. f. Proc. Indian Acad.
Sci. Sect. B. 1: 197-204.
and Johri, B. M. 1950. The embryology of Hydrilla verticillata (in press).
Mascre, M. 1919a. Sur le role de l'assise nourriciere du pollen. Compt. Rend.
Acad, des Sci. Paris 168: 1120-1122. . 19196. Nouvelles remarques sur le role de l'assise nourriciere du pollen.
Compt. Rend. Acad, des Sci. Paris 168: 1214-1216. Matthews, J. R., and Taylor, G. 1926. The structure and development of
the stamen in Erica hirtiUora. Trans, and Proc. Bot. Soc. Edinb. 29:
235-242. Meyer, K. 1925. tlber die Entwicklung des Pollens bei Leontodon autumnalis
L. Ber. deutsch. bot. Gesell. 43: 108-114. Mezzetti-Bambacioni, V. 1941. Ricerche morfologiche sulle Lauraceae. Embri-
ologia della Umbellularia calif ornica Nutt. Ann. di Bot. 22: 99-116. Moissl, E. 1941. Vergleichende embryologische Studien liber die Familie der
Caprifoliaceae. Osterr. bot. Ztschr. 90: 153-212. Newman, I. V. 1928. The life history of Doryanthes excelsa. I. Some ecological
and vegetative features and spore production. Proc. Linn. Soc. N. S. Wales
53: 499-538. Nietsch, H. 1941. Zur systematischen Stellung von Cyanastrum. Osterr. bot.
Ztschr. 90: 31-52. Nohara, S. 1934. Gametogenesis and embryogeny of Sesamum indicum L.
Jour. Col. Agr. Tokyo Imp. Univ. 13: 19-25. Peters, K. 1903. "Vergleichende Untersuchungen fiber die Ausbildung der
sexuellen Reproduktionsorgane bei Convolvulus und Cmcuta." Diss. Zurich.
52 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
Pisek, A. 1924. Antherenentwicklung und meiotische Teilung bei der Wachol- dermistel (Arceuthobium oxycedri [D.C.] M.B.); Antherenbau und Chromo- somen zahlen von Loranthus europaeus Jacq. Sitzber. math., nat. Kl. Akad. der Wiss. Wien 1 133: 1-15.
Puri, V. 1941. Life history of Moringa oleifera Lamk. Jour. Indian Bot. Soc. 20: 263-284.
Rauch, K. V. 1936. Cytologische-embryologische Untersuchungen an Scurrula atropurpurea Dans, und Dendrophthoe pentandra Miq. Ber. schweiz. bot. Gesell. 45:5-61.
Reeves, R. G. 1928. Partition wall formation in the pollen mother cells of Zea mays. Amer. Jour. Bot. 15: 114-122.
Rosenberg, O. 1901. tJber die Embryologie von Zostera marina L. Bihang Till K. Svenska Vet.-Akad. Handl, HI, 27(6) : 1-24.
Rutishauser, A. 1935. Entwicklungsgeschichtliche und zytologische Untersuch- ungen an KorthaJsella dacrydii (Ridl.) Danser. Ber. schweiz. bot. Gesell. 44: 389-436.
Samuelsson, G. 1914. tlber die Pollenentwicklung von Anona und Aristolochia und ihre systematische Bedeutung. Svensk Bot. Tidskr. 8: 181-189.
Schaeppi, H., and Steindl, F. 1942. Bliitenmorphologische und embryologische Untersuchungen an Loranthoideen. Vrtljschr. naturf. Gesell. Zurich 87:
301-372.
and . 1945. Bliitenmorphologische und embryologische Unter- suchungen an einigen Viscoideen. Vrtljschr. naturf. Gesell. Zurich 90: 1-46.
Sharp, L. W. 1943. "Fundamentals of Cytology." McGraw-Hill Book Company.
Singh, Bahadur. 1936. The life history of Ranuncidus sceleratus Linn. Proc. Indian Acad. Sci. Sect. B. 4: 75-91.
. 1950. The embryology of Dendrophthoe falcata (Linn, fil.) Ettingshausen.
Proc. 37th Indian Sci. Cong. Sect. Bot.
Smith, C. M. 1929. Development of Dionaea muscipula. I. Flower and seed. Bot. Gaz. 87: 507-530.
Stadtler, G. 1923. tlber Reduktionserscheinungen im Bau der Antherenwand von Angiospermen. Flora 16: 85-108.
Stenar, H. 1925. "Embryologische Studien. I. Zur Embryologie einiger Columniferen. II. Die Embryologie der Amaryllidaceen." Diss. Uppsala.
Stern, H. 1946. The formation of polynucleated pollen mother cells. Jour. Hered. 37: 47-50.
Stevenson, G. B. 1934. The life-history of the New Zealand species of the para- sitic genus Korthahella. Trans, and Proc. Roy. Soc. New Zeal. 64: 175-190.
Subramanyam, K. 1948. An embryological study of Melastoma malabathricum L. Jour. Indian Bot. Soc. 27: 1-9.
Swamy, B. G. L. 1946. Embryology of Habenaria. Proc. Natl. Inst. Sci. India
12: 413-426. . 1948. Embryological studies in Orchidaceae. I. Gametophytes.
Amer. Midland Nat. 41: 184-201.
THE MICROSPORANGIUM 53
Thoday, D., and Johnson, E. T. 1930. On Arceuthobium pusillum Peck. II.
Flowers and fruit. Ann. Bot. 44: 813-824. Tischler, G. 1915. Die Periplasmodienbildung in den Antheren der Commelina-
ceen und Ausblicke auf das Verhalten der Tapetenzellen bei den ubrigen Mono-
cotylen. Jahrb. f. wiss. Bot. 55: 52-90. Ubisch, G. V. 1927. Zur Entwicklungsgeschichte der Antheren. Planta 3:
490-495. Venkatasubban, K. R. 1945. Cytological studies in Bignoniaceae. The cytology
of Dolichandrone rheedii Seem, and allied genera. Proc. Indian Acad. Sci.
Sect. B. 21: 77-92. Vignoli, L. 1936. Cariologia del genre Agave. I. Lavori, R. 1st. Bot. Palermo 7.
. 1937. Cariologia del genre Agave. II. Lavori f. R. 1st. Bot. Palermo 8.
Warming, E. 1873. Untersuchungen iiber Pollenbildende Phyllome und Kaul-
ome. Hanstein's Bot. Abhandl. 2: 1-90. Witkus, E. R. 1945. Endomitotic tapetal cell divisions in Spinacia. Amer.
Jour. Bot. 32: 326-330. Witmer, S. W. 1937. Morphology and cytology of Vallisneria. Amer. Midland
Nat. 18: 309-327. Wodehouse, R. P. 1936. "Pollen Grains." McGraw-Hill Book Company. Woodworth, R. H. 1931. Cytomixis. Jour. Arnold Arboretum 12: 23-25. Wylie, R. B. 1904. The morphology of Elodea canadensis. Bot. Gaz. 37: 1-22.
CHAPTER 3
THE MEGASPORANGIUM
The megasporangium or ovule consists of the nucellus and one or two integuments. It may have various forms, which sometimes intergrade into one another, and very often the same ovule changes its form during the course of its development. Mature ovules are usually classed under five types. In the orthotropous or atropous type the micropyle lies directly in line with the hilum and above it (Fig. 37 A) as in Polygonaceae, Urticaceae, Cistaceae and Pipera- ceae. In the anatropous type the body of the ovule becomes com- pletely inverted so that the micropyle and hilum come to lie very close to each other (Fig. 375). This form is universal in almost all members of the Sympetalae and is also found in several other families belonging to both dicotyledons and monocotyledons. When the ovule is curved, as in some of the Resedaceae and Leguminosae, it is called campylotropous (Fig. 37C); when the curvature is more pronounced and also affects the embryo sac, so that the latter be- comes bent like a horseshoe, as in the Alismaceae, Butomaceae, and Centrospermales, the ovule is called amphitropous (Fig. 37 E); and when the nucellus and integuments lie more or less at right angles to the funiculus as in Ranunculus, Nothoscordum, and Tulbag- hia, it is called hemianatropous or hemitropous (Fig. 37 D). Ovules may also be designated as epitropous, apotropous, or pleurotropous, according as the inversion or bending is directed towards the top, bottom, or sides of the ovary.
A very peculiar type of ovule is seen in some members of the Plumbaginaceae (Fig. 38). Here the nucellar protuberance is at first in the same line as the axis, but the rapid growth on one side causes it to become anatropous. The curvature does not stop but continues until the ovule has turned over completely so that the micropylar end again points upwards. It has been suggested that this kind of ovule, also seen in Opuntia (Fig. 39), is distinctive enough to merit a separate name, circinotropous (Archibald, 1939).
54
THE MEGASPORANGIUM
55
Integuments. Ordinarily the ovule has either one or two integu- ments. The number is constant in most families, and only in rare cases do unitegmic and bitegmic ovules occur in the same family. In the Sympetalae a single massive integument is almost universal,
ABC D E
Fig. 37. Types of ovules as seen in vertical longitudinal section. A, atropous or orthotropous. B, anatropous. C, campy] otropous. D, hemianatropous. E, am- phitropous. (After Prantl.)
the Plumbaginales and Primulales being the only important excep- tions. In the Archichlamydeae and the monocotyledons most gen- era have two integuments but a few have only one. There is evidence that in many cases the single integument has originated
C ~ v D E F
Fig. 38. Development of ovule of Plumbago capensis. (After Haupt, 1934-)
by a fusion of two separate primordia. Transitional types have been observed in some members of the Ranunculaceae, Rosaceae, Connaraceae, and Icacinaceae.
The unitegmic condition may also arise by an elimination of one
56
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
of the two integuments. In Cytinus, a member of the Ramesia- ceae, the outer integument is arrested in its development. In the Salicaceae (see Schnarf, 1929) Populus tremula has a single integu- ment, while P. canadensis and P. candicans also possess a weakly developed inner integument which is apparently on its way to ex- tinction. In the Icacinaceae (Fagerlind, 1945c) Gomphandra and Gonocaryum are unitegmic but Phytocrene shows two primordia. In some plants there is also a third integument or aril.1 In Ulmus
Fig. 39. Development of ovule of Opuntia aurantiaca. A, front view of young ovule. B-D, longitudinal sections of progressively older ovules (nu = nucellus; it = inner integument; oi = outer integument; / = funiculus). (After Archibald, 1939.)
(Shattuck, 1905) it is said to originate by the splitting of the outer integument, but in most other cases it is a new structure arising from the base of the ovule. Good examples of this kind are seen in Asphodelus (Fig. 40 A, B) and Trianthema (Fig. 40C). Of a differ- ent origin is the "caruncle," found in several members of the Euphor- biaceae, which arises by a proliferation of the integumentary cells at the micropylar region (Landes, 1946). Sometimes this prolifera-
1 In Canangium, Mezzettia, and Xylopia Corner (1949) records the presence of a "middle integument" arising between the outer and inner integuments. In Canangium and Xylopia, which also have an aril, the middle integument becomes the fourth integument of the seed.
THE MEGASPORANGIUM
57
tion becomes more pronounced and takes the form of a backwardly directed process (Fig. 40D-G) which resembles an aril in later stages. A very peculiar condition occurs in Opuntia (Archibald, 1939), where the extremely long funiculus completely surrounds the ovule and looks like a third integument (Fig. 39).
D E F G
Fig. 40. Diagrams of ovules showing origin of aril or third integument (A-C) and caruncle (D-G). A,B, Asphodelus fistulosus. (After Stenar, 1928.) C, Trian- thema monogyna. (After Bhargava, 1935.) D, Brachychilum horsfieldii. (After Mauritzon, 1936.) E, Burbidgea scMzocheila. (After Mauritzon, 1936.) F,G, Careya arborea. (After Mauritzon, 1939.)
Whatever may be the condition of the integuments in the younger stages, they often present a very different and a more complicated aspect in the mature seed. Frequently several layers of cells are
58 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
completely absorbed and do not take any part in the formation of the seed coat. In the Umbelliferae only two or three of the outer layers persist at maturity; in the Compositae most of the cells dis- appear, leaving only a thin layer of crushed and disorganized tissue ; and in Circaeaster (Junell, 1931), Thesium (Rutishauser, 1937), and Zea (Randolph, 1936) practically nothing remains of the seed coat. In Symplocarpus (Rosendahl, 1909) both integuments and endo- sperm are consumed so that the embryo lies naked inside the ovary wall. So variable is the nature of the cell layers surrounding the embryo that only a thorough study of the developmental stages can reveal their true nature.
Mention must be made of a few records of the occurrence of chlorophyll in the integuments. Hofmeister (1861) observed this in Brunsvigia minor and Amaryllis belladonna, and Treub (1879) in Sobralia micrantha. Later, Berg (1898) and Puri (1941) reported the presence of chlorophyll in the outer integument and a portion of the chalaza in Gladiolus communis, Lilium martagon, and Mor- inga oleifera. Schlimbach (1924) observed the presence of stomata on the outer integument of Nerine curvifolia, and Flint and More- land (1943) have described the occurrence of an elaborate chloro- phyllous tissue with stomata in Hymenocallis occidentalis. Stomata have also been found on the outer integument of Gossypium, but they are believed to be concerned with respiration rather than transpiration or photosynthesis (Seshadri Ayyangar, 1948).u
Micropyle. When two integuments are present, the micropyle may be formed either by the inner integument as in the Centro- spermales and Plumbaginales (Fig. 38) or by both inner and outer integuments as in the Pontederiaceae (Fig. 142). Less frequently, as in the Podostemonaceae, Rhamnaceae, and Euphorbiaceae, it may be formed by the outer integument alone (Fig. 67 A). When both the integuments take part in the formation of the micropyle, the passage formed by the outer integument (exostome) may not be in line with that formed by the inner integument (endostome) so that the micropylar canal has a somewhat zigzag outline. Good examples of this kind are seen in the Resedaceae (Oksijuk, 1937) and in some members of the Melastomaceae (Subramanyam, 1948). In Leitneria (Pfeiffer, 1912) and Malpighia (Subba Rao, 1941) there
laSee Boursnell (1950) on the occurrence of a fungus in the funiculus and outer integument of Helianthemum chamaecistus.
THE MEGASPORANGIUM 59
is an excessive development of the upper portion of the integuments so that the micropylar canal lies in folds over the nucellus. Rarely, as in Ficus (Condit, 1932), Fouquieria (Khan, 1943), and Cyno- morium (Steindl, 1945), the integumentary cells come in such in- timate contact with each other that the micropylar canal is ex- tremely narrow and imperceptible.
Nucellus.16 Depending on the extent of development of the nucel- lus, ovules are called crassinucellate or tenuinucellate.2 In the first type, there is a well-developed parietal tissue and the megaspore mother cell is separated from the nucellar epidermis by one or sev- eral layers of cells. In the second type, parietal cells are absent and the megaspore mother cell lies directly below the nucellar epidermis.3
In the crassinucellate forms the nucellus may enlarge either by an increase in the number of the parietal cells or by periclinal divisions of the nucellar epidermis. In some plants like Zizyphus (Kajale, 1944) and Quisqualis (Fagerlind, 1941) (Fig. 41) both these processes take place simultaneously.
Several members of the Salicaceae, Nyctaginaceae, Euphorbia- ceae, Polygonaceae, and Cucurbitaceae are characterized by having a beak-shaped nucellus which reaches out into the micropyle. In one species, Polygonum persicaria (Soueges, 1919), the beak forms a very conspicuous structure protruding upward to the base of the style (Fig. 42).
The tenuinucellate forms are of two kinds: (1) those in which the nucellus is short and the primordia of the integument or integu-
16 For more detailed information on the nucellus, see Dahlgren (1927).
2 It should be noted that the above distinction between crassinucellate and tenuinucellate ovules, although convenient and useful, is not always sharp and clear-cut and there are various intergradations between them. Further, both types may sometimes occur in one and the same species. To mention only two examples, in Butomus (Holmgren, 1913) and Ophiopogon (Maheshwari, 1934) in some ovules the megaspore mother cell is situated directly below the nucellar epi- dermis while in others it is separated from the latter by a wall cell.
3 Even in those plants in which the ovules are usually tenuinucellate and devoid of parietal cells, some of the cells of the nucellar epidermis may undergo one or two periclinal divisions. Svensson (1925) and Dahlgren (1927) have figured this in Helioptropium and Cobaea. Here the epidermal cells just above the megaspore tetrad undergo a radial elongation followed by a periclinal division which may give the false impression of the cutting off of parietal cells.
60 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
ments arise near its apex (Fig. 43£), and (2) those in which the nucel- lus is elongated and the integuments arise near its base (Fig. 43 A). The Asclepiadaceae, Orobanchaceae, and Rubiaceae are good ex- amples of the first condition, and the Orchidaceae of the second. As the embryo sac matures, the nucellar cells gradually become
C
Fig. 41. L.s. ovules of Quisqualis indica showing progressively increasing amount of parietal tissue, arising partly by divisions of the wall layers and partly by divi- sions of cells of nucellar epidermis. A, B, megaspore mother cell stage. C, func- tioning megaspore stage. In B and C, note enlarging cells of obturator. (After Fagerlind, 19/, J.)
THE MEGASPORANGIUM
61
used up.4 In the tenuinucellate forms this takes place at such an early stage (even before fertilization) that some workers have mis- interpreted the integument as the nucellus. Schleiden (1837) wrote long ago that in the Rubiaceae the ovules are naked. Lloyd (1902) demonstrated the presence of an integument in all the genera studied by him excepting Houstonia. Owing to its narrow and incon-
A
B
Fig. 42. Formation of nucellar beak in Polygonum persicaria. A, young nucellus, showing megaspore mother cell and four wall cells; note periclinal division of a cell of the nucellar epidermis. B, older stage, showing megaspore tetrad, wall cells, and nucellar beak. C, mature embryo sac with part of nucellar beak; wall cells have degenerated and disappeared. (After Soueges, 1919.)
spicuous micropyle, Schleiden mistook the integument for the nucel- lus, while the latter escaped his notice altogether. More recently, Fagerlind (1937) has shown that even in Houstonia an integument is present as usual and it is really the nucellus which is on its way to extinction. He presents a series of stages to show how this con- dition has been derived (Fig. 44). In Phyllis, which is at the begin-
4 It is only in a few families like the Piperaceae and Scitamineae that the nucellus persists in the seed; it is then known as the perisperm.
62 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
ning of the series, the nucellus comprises a single layer of cells (the epidermis) surrounding the archesporium (Fig. 4:4 A). This is in accordance with the general condition in the Sympetalae. In Bou- vardia and Vaillantia, which represent the next stage, the nucellar epidermis is reduced to a few cells lying immediately above the sporogenous tissue (Fig. 44B,C). In Rubia olivieri there is further
A B
Fig. 43. Young ovules of Orchis maculatus (A) and Aeginetia indica (B). Note that in Orchis the integuments arise near base of megaspore mother cell, while in Aeginetia the single integument arises near apical end of nucellus. (A, after Hage- rup, 1944,' B, afar Juliano, 1935.)
A B C D E F
Fig. 44. Diagram illustrating different types of nucelli found in the Rubiaceae. A, Phyllis. B, Bouvardia. C, Vaillantia. D, Rubia. E, Oldenlandia. F, Hous- tonia. {After Fagerlind, 1937.)
THE MEGASPORANGIUM 63
reduction in their number, although this is accompanied by a pro- nounced radial elongation of the walls (Fig. 44Z)). In Oldenlandia the micella r epidermis is represented by one or two cells only (Fig. 4AE), and in Houstonia, which is the last member of the series, there is no distinguishable epidermis and the ovule consists of only the sporogenous cells and the integument (Fig. 44i^).
Fagerlind's series is so clear and convincing that there is no longer any doubt about the true relationships of the nucellus and integu- ment in the Kubiaceae. Houk's (1938) statement that in Coffea there is no distinction between the tissues of the integument and nucellus is therefore incorrect (see also Mendes, 1941).
Woodcock's (1943) report that in Ipotnoea the ovule has no dis- tinct integument and the micropyle is formed by an "invagination" is also due to a misinterpretation. As in other members of the Convolvulaceae (see Maheshwari, 1944), an integument is present and it is the nucellus which soon disappears. The micropyle is not an invagination but a continuous passage, which begins to be more or less occluded in postfertilization stages and is therefore difficult to demonstrate in nonmedian sections.
Formerly the Olacaceae were also believed to have naked ovules. A recent study by Fagerlind (1947) has shown that an integument is present as usual but the nucellus is extremely reduced and ephem- eral and is represented by only a few epidermal cells lying just above the megaspore mother cell.
A complete absence of the integuments is known only in some members of the Loranthaceae and Balanophoraceae, but it seems probable that this is a derived condition. Fagerlind (1945c?) has given a series of illustrations showing the stages by which this may have been brought about (Fig. 45). The case of Crinum (Amarylli- daceae), in which the nucellus is ephemeral and the integuments are said to be absent (Tomita, 1931), deserves further study.
Integumentary Tapetum. In those plants in which the nucellus is soon disorganized, the embryo sac comes in direct contact with the inner layer of the seed coat. The cells of this layer frequently become specially differentiated from the rest by their form and con- tents (Fig. 46). They show a pronounced radial elongation and sometimes become binucleate. Owing to these similarities with the cells of the anther tapetum, this layer of cells is known as the integu- mentary tapetum or endothelium.
64
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
There seems to be no doubt that the endothelium is a nutritive layer whose chief function is to serve as an intermediary for the transport of food materials from the integument to the embryo sac.
E F G H I
Fig. 45. Diagram illustrating derivation of the female flower and ovule of Bala- nophora. A, l.s. hypothetical ovary showing two ovules. B, ovary of Thesium. C, ovary, as in Osyris, Santalum and Myzodendron. D, as in Arceuthobium and Helosis. E, as in Korthalsella. F, as in Viscum and Dendrophthoe. G,H, as in Scurrula. I, as in Balanophora. {After Fagerlind, 1945d.)
Some writers also claim that it contains diastase and other enzymes which convert the food into a suitable form for the use of the embryo sac. In later stages, when the embryo is approaching maturity,
THE MEGASPORANGIUM
65
the inner surface of the endothelium becomes cutinized and this
layer seems to take up a protective instead of a nutritive function.
Hypostase.6 Just at the level of origin of the two integuments
and directly below the embryo sac, there is often a well-defined but
Fig. 46. Stages in the formation of integumentary tapetum in Lobelia trigona. A, two-nucleate embryo sac with remains of degenerating megaspores; nucellar epidermis still intact. B, Four-nucleate embryo sac, showing degeneration of nucellar epidermis and formation of integumentary tapetum from inner layer of integument. C, mature embryo sac bounded by cells of integumentary tapetum. (After Kausik, 1935.)
irregularly outlined group of nucellar cells which are usually poor in cytoplasmic contents but have partially lignified or suberized walls composed of a highly refractive material. Van Tieghem (1901),
6 Dahlgren (1940) has reviewed the literature on the occurrence of the hypostase in angiosperms and also recommended some changes of terminology. Reference should be made to this paper for fuller information on the subject.
66
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
who first called attention to this patch of cells, gave it the name hypostase, and believed that it formed a sort of barrier or boundary for the growing em- bryo sac and prevented it from pushing into the base of the ovule. Goebel (1933) says, however, that the peculiar po- sition of this tissue — directly above the termination of the vascular supply of the ovule — is indicative of its relation to the water economy of the em- bryo sac. While the function of the hypostase is still in doubt, morphologically it is a very characteristic feature of certain families and genera. Zostera (Dahlgren, 1939) offers an especially good instance of a well-developed hypostase (Fig. 47). The hypostase may not always consist of thick- walled cells. In Knautia (La- vialle, 1925) it comprises a group of small thin-walled cells having a number of schizoge- nous cavities which branch and anastomose and become filled with a yellowish sub- stance, which also spreads into the antipodal cells and other adjacent tissue. In Dionaea (Smith, 1929) some of the thin- walled cells in the chalaza be- come disorganized and replaced Jig. 47. Zosteia marina, l.s. young seed, °
showing prominent hypostase and well-de- by airspaces . In A Ilium odorum veloped embryo. (After Dahlgren, 1939.) (Haberlandt, 1923) the cells
THE MEGASPORANGIUM 67
become richly protoplasmic and the hypostase has an appearance similar to that of the epithem of many hydathodes. Haberlandt considers it to be a sort of glandular tissue secreting some hor- mone or enzyme required for the growth of the embryo sac.
Epistase. Van Tieghem also reported the occasional presence of a similar well-marked tissue in the micropylar part of the ovule and called it the epistase. Usually it originates from the apical cells of the nucellar epidermis, which show a marked radial elonga- tion and become somewhat thickened or suberized. Occasionally the cells undergo one or more periclinal divisions to form the so- called nucellar cap, which persists as a hood over the apex of the embryo sac even after the cells at the sides have disorganized and disappeared.6 In Castalia (Cook, 1906) the epidermal cells lying at the apex of the embryo sac show "a very pronounced sclerifica- tion," and in Costus (Boehm, 1931) the inner tangential walls of these cells become conspicuously thickened. In Nicolaia (Boehm, 1931) the walls surrounding the megaspore tetrad become cutinized and form a firm covering, which becomes ruptured and separated into two parts only with the continued enlargement of the embryo sac. The thickenings at the micropylar end disappear but are seen once again at the time of organization of the mature embryo sac.
In some plants the apical cells of the integuments give rise to a proliferation usually called the "operculum." To mention a few examples, in Lemna (Caldwell, 1899) the cells forming the micro- pylar portion of the two integuments enlarge and divide to form a compact tissue lying just above the nucellus (Fig. 48). In Dionaea (Smith, 1929) a similar tissue is formed by the cells of the inner integument. In Acorus (Buell, 1935) the cells become elongated and coiled around one another, so as to form a plug in the lower part of the micropyle.
Vascular Supply of Ovule. As a rule the vascular bundle entering the ovule terminates at the chalaza but in some plants it gives out branches, a few of which enter the integument. If two integuments are present, the branches may enter only the outer integument or both the outer and the inner integuments. Since integumentary vascular bundles are common in gymnosperms, their presence is usually considered to be a primitive feature and the loss of the con-
6 Dahlgren (1940) designates a persistent nucellar cap of this kind by the name "petasus."
68
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
ducting tissue to be an advanced one. There is, however, no defi- nite evidence in favor of this view. Integumentary vascular bun- dles are now known to occur in a number of families, both primitive (Betulaceae, Euphorbiaceae, Ranunculaceae, and Berberidaceae) and specialized (Moringaceae, Leguminosae, Punicaceae, Rhamna- ceae, Convolvulaceae, Cuscutaceae, Boraginaceae, Caprifoliaceae, Compositae, and Cyanastraceae). In Zizyphus (Kajale, 1944) the vascular strands extend far up into the tip of the outer integument. In the large succulent seeds of Hymenocallis occidentalis (Whitehead
emb
ABC
Fig. 48. Development of "operculum" in Lemna minor. A, ovule showing two integuments, ox and xi, nucellar cap nc, and young embryo sac. B, upper part of ovule showing portion of embryo emb, nucellar cap, and two integuments; note enlargement of cells of outer integument. C, embryo, nucellus, and thickened tips of integuments which form the so-called "operculum." (After Caldwell, 1899.)
and Brown, 1940) four bundles enter the ovule and during their upward course they freely branch and anastomose so that a cross section of a large seed shows from 14 to 18 bundles in the seed coat. The inner integument of Croton (Landes, 1946) shows a network of tracheids which remain conspicuous even after the other cells of the integument have become flattened and crushed.
The occurrence of vascular elements in the nucellus is much rarer. Benson (1894), Frye (1902), and Benson, Sanday, and Berridge (1906) identified some nucellar tracheids in Castanea, Asclepias, and Carpinus respectively, but they showed no connection with the vascular bundle of the funiculus. Guerin (1915) described the oc- currence of connecting nucellar tracheids in some genera of the Thymelaeaceae,7 and Orr (1921a, b) reported the same in a few mem- bers of the Capparidaceae and Resedaceae.
7 According to Mauritzon (1939) all statements of the occurrence of xylem ele-
THE MEGASPORANGIUM 69
Among recent records, in Agave (Grove, 1941) and Strombosia (Fagerlind, 1947), the vascular strand of the ovule is said to pene- trate into the nucellus up to the base of the embryo sac, and in Magnolia (Earle, 1938) it gives out short branches in the chalaza, one of which is directed towards the embryo sac. In Acalypha (Landes, 1946) the main bundle of the ovule proceeds up to the hy- postase and forms a number of short branches whose ultimate rami- fications extend into the nucellus up to a distance about one-fifth of the length of the ovule. More striking still is the condition recently reported in Casuarina (Swamy, 1948), where the funicular strand extends up to the base of the sporogenous tissue, some of whose cells elongate and themselves assume a conducting function instead of giving rise to embryo sacs (Fig. 51).
The occurrence of vascular elements in the nucellus is of con- siderable theoretical importance, as such a condition has been consid- ered by some authors to be a relic of the highly developed "trachei- dal envelope" found in some fossil gymnosperms. A few years ago integumentary vascular bundles were considered to be very un- common in angiosperms, but now they are known to occur in several families. Possibly the occurrence of xylem elements in the nucellus may also be found to be more frequent than the few reports just mentioned may seem to indicate.
Archesporium. The archesporial tissue is of hypodermal origin. In general, one cell of the nucellus, situated directly below the epi- dermis, becomes more conspicuous than the others owing to its larger size, denser cytoplasm, and more prominent nucleus. This is the primary archesporial cell. Frequently the cells situated below it lie in a row so that the archesporial cell appears as the terminal member of a series of nucellar cells (Fig. 43 A).
The archesporial cell may divide to form a primary parietal cell and a primary sporogenous cell (Fig. 49A-B), or it may func- tion directly as the megaspore mother cell (Fig. 50H). The primary parietal cell may remain undivided or it may undergo periclinal and anticlinal divisions to form a variable number of wall layers. The
ments in the nucellus or inner integument of the Thymelaeaceae are due to mis- interpretations. In his opinion these tracheids really belong to the chalazal tissue which, by "vigorous growth," extends around the endosperm and thus forms a part of the seed coat. Fuchs (1938) and Kausik (1940) also failed to observe any nucellar tracheids in the species studied by them.
H G
/ig. 49. Hydrilla verticil lain, formation of megaspores. .4, hypodermal arche- sporial cell. B, cutting off of primary parietal cell. C, anticlinal division of pri- mary parietal cell. D, ovule at megaspore mother cell stage. E, megaspore mother cell in prophase of Meiosis I ; two other cells of the nucellus lying below it simulate sporogenous cells. F,G, first division of megaspore mother cell resulting in formation of dyad cells. H, tetrad of megaspores.
70
THE MEGASPORANGIUM 71
primary sporogenous cell usually functions as the megaspore mother cell without undergoing any further divisions.
The outline presented above is subject to many variations. In some plants the archesporial cell is said to originate from the third layer of cells in the nucellus, but this is probably a misinterpretation caused by the difficulty in distinguishing the archesporial cell at an earlier stage of development. Sometimes, as in the Onagraceae (Khan, 1942), the archesporium may comprise a small group of half a dozen cells or more (Fig. 5022). Of these usually the central cell alone is functional, but frequently one or two of the other cells also reach the megaspore mother cell stage. In the Malvaceae (Stenar, 1925) the primary sporogenous cell divides to form a few accessory cells in addition to the functional megaspore mother cell. In some members of the Rubiaceae and Compositae there are sev- eral sporogenous cells, all of which may go through the meiotic divisions (Fig. 50B-C). In Scurrula (Rauch, 1936) and Dendroph- thoe (Singh, 1950), which have a very massive archesporium, the sporogenous cells undergo further division to give rise to a still larger number of cells. These begin to elongate very actively and become so closely interlocked that the whole tissue gives an appear- ance suggestive of the hy menial layer of an ascomycete.
In Hydrilla there are sometimes two or three archesporial cells in a single row (Fig. 50A). In Ruppia (Murbeck, 1902), Butomus (Holmgren, 1913), and Urginea (Capoor, 1937) the primary parie- tal cell may also assume a sporogenous function so that two mega- spore tetrads are formed in the same row (Fig. 52D). In Oncidium praetextum (Afzelius, 1916) a cell of the nucellar epidermis may func- tion as a megaspore mother cell (Fig. 50F), and in Solanum (Bha- duri, 1932) and Limnanthes (Fagerlind, 1939) some of the integu- mentary cells may behave similarly (Fig. 50D,G).
In the Sympetalae a parietal cell is absent, the only important exceptions being the Plumbaginales and some members of the fam- ily Convolvulaceae. Since this is also the condition in some other advanced families like the Umbelliferae and Orchidaceae, the pres- ence of a massive parietal tissue is regarded as a primitive feature and its absence as advanced. An objection to this view is that parietal cells are often absent even in some admittedly primitive families like the Ranunculaceae (Hafliger, 1943).
In the Casuarinaceae (Fig. 51), and some other families (see
72 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
Fig. 50. Variations in origin and extent of sporogenous tissue in ovule. A, Hydrilla verticillata, young ovule, showing a row of three sporogenous cells. B, Achillea millefolium, nucellus, showing a number of megaspore mother cells in
THE MEGASPORANGIUM 73
Schnarf, 1929) there is a multicellular archesporium and also an extensive parietal tissue. In some members of the Rosaceae the cells of the nucellar epidermis also divide periclinally and thus add to the wall tissue (see also page 60).
Megasporogenesis. The megaspore mother cell undergoes the usual meiotic divisions to form a tetrad of four cells. The first division is always transverse and gives rise to two dyad cells (Fig. 4QF-G). Typically the second division is also transverse and re- sults in a linear tetrad of four megaspores (Fig. 52 A). Frequently the micropylar dyad cell divides in a plane at right angles to that of the chalazal dyad cell. This results in a T-shaped tetrad in which the two outer megaspores lie in contact with the third mega- spore which separates them both from the chalazal megaspore (Fig. 52B). Since both linear and T-shaped tetrads may occur in ovules of one and the same ovary, it is unnecessary to give specific ex- amples. Tetrads of an intermediate type (Fig. 4QH), in which the wall separating the two micropylar megaspores lies at an angle of approximately 45° with respect to the chalazal megaspores, are also not infrequent.
Rarely, the two upper megaspores of a tetrad lie in a line parallel to the long axis of the ovule but the lower two lie at right angles to it. Such 1-shaped tetrads are sometimes found in the Onagra- ceae and have been reported in Zauschneria (Johansen, 1931a), Anogra (Johansen, 19316), and Ludwigia (Maheshwari and Gupta, 1934). 8 Among other examples of a similar kind may be cited Drimiopsis (Baranow, 1926), Tacca (Paetow, 1931), Styrax (Cope-
8 The occurrence of 1-shaped tetrads in the Onagraceae is probably related to the fact that here the micropylar megaspore gives rise to the embryo sac, while the three chalazal megaspores are nonfunctional (see Chap. 4).
prophase. (After Dahlgren, 1927.) C, Chrysanthemum corymbosum, nucellus, show- ing megaspore mother cells each with four megaspore nuclei; one cell at the bottom has lagged behind. (After Dahlgren, 1927.) D, Solanum melongena, ovule show- ing hypodermal megaspore mother cell and two other such cells in the tissues of the integument. (After Bhaduri, 1932.) E, Jussieua repens, nucellus, showing multicellular archesporium. (After Khan, 1942.) F, Oncidium praetextum, ovule, showing supernumerary archesporial cell arising from the nucellar epidermis. (After Afzelius, 1916.) G, Limnanthes douglasii, normal archesporial cells in nucellus and supernumerary archesporial cell in integument. (After Fagerlind, 1939.) H, Machaerocarpus californicus, megaspore mother cell in prophase. (Af- ter Maheshwari and Singh, 1943.)
74 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
land, 1938), Cyathula (Kajale, 1940), Costus (Banerji, 1940), and Desmodium (Pantulu, 1941).
An isobilateral or a tetrahedral arrangement of megaspores is very rare and has been reported only as an abnormality (Fig. 52C).
Fig. 51. Part of ovule of Casuarina montana, showing multiple archesporium. Some of the sporogenous cells are in prophase; others have gone through meiotic divisions to form megaspore tetrads; and a few have formed two- and four-nucleate embryo sacs. Note that in tetrad at upper end, all megaspores are binucleate. {After Swamy, 1948.)
THE MEGASPORANGIUM
75
The genus Musa is of special interest, for here tetrads of four dif- ferent kinds — linear, T-shaped, 1-shaped and isobilateral — may oc- cur in the same species (Dodds, 1945). Tetrads of very variable appearances have also been described in Poa alpina (Hakansson, 1943).
Frequently a row of only three cells is seen in place of the usual four. This is due to an omission of the second meiotic division in one of the two dyad cells, usually the upper.9 All intermediate stages leading towards this condition have been seen. In some plants the division in the upper dyad cell merely lags behind that
D
Fig. 52. Megaspore tetrads in Urginea indica. A, linear tetrad. 5,T-shaped tetrad. C, tetrad showing decussate arrangement of megaspores. D, two tetrads lying in same row. (After Capoor, 1937.)
in the lower dyad cell, and all four cells are formed as usual; in others the nucleus divides normally, but a separating wall is not laid down; in still others the division is abortive and merely gives rise to two degenerating clumps of chromosomes; and in a few cases the nucleus degenerates without undergoing any division.
Functioning Megaspore. Normally it is the chalazal megaspore of the tetrad which functions and gives rise to the embryo sac, while the remaining three megaspores degenerate and disappear. But
9 It is to be noted that sometimes one gets a false impression of the occurrence of a row of three cells either because of the plane of the section or because of the orientation of the wall between the two upper megaspores so that one of the cells lies superposed over the other. In such cases more careful focusing, or a study of the adjacent section, reveals the presence of the fourth megaspore (see Graves, 1908).
K L
Fig. 53. Formation of megaspore tetrads in various angiosperms. A, Balano- phora elongata, megaspore mother cell. B, dyad stage; lower dyad cell is much smaller than upper. C,D, both dyad cells dividing. E,F, tetrad stage; note that uppermost megaspore functions, while the other three degenerate and disappear. (After Fagerlind, 1945b.) G,H, Gloriosa virescens, megaspore tetrads in which every cell is binucleate. /, one of the megaspores has developed to four-nucleate stage and another to two-nucleate. (After Afzelius, 1916.) J, Aristotelia racemosa, two tetrads with third megaspore functioning. (After Mauritzon, 1934.) K, Rosa, two tetrads with micropylar megaspores functioning. L, same, megaspore tetrad with both micropylar and submicropylar megaspores enlarging. (After Hurst, 1931.) M, Senecio abrotanifolius, megaspore tetrad, showing two middle mega- spores lying side by side. (After Afzelius, 1924) N, Culcitium reflexum, mega- spore tetrad, showing micropylar as well as chalazal megaspore enlarging. (After Afzelius, 1924)
76
THE MEGASPORANGIUM
77
in Elytranthe (Schaeppi and Steindl, 1942), Langsdorffia (Fagerlind, 1945a), and Balanophora (Fagerlind, 19456) (Fig. 53 A-F) the micro- pylar megaspore gives rise to the embryo sac and the other three soon degenerate. A similar condition occurs in the Onagraceae and in a few members of the Compositae, although here it is not unusual to find both the terminal megaspores, micropylar and chala- zal, growing concurrently (Fig. 53ilf,iV). In Rosa (Hurst, 1931)
Fig. 54. Formation of megaspore haustoria in Galium lucidum (A-C), Sedum sempervivoides (D), and Rosularia pallida (E). (A-C after Fagerlind, 1987; D-E after Mauritzon, 1933.)
it is usually the micropylar megaspore which functions (Fig. 532£) but sometimes it is the second (Fig. 53L). In Aristotelia (Maurit- zon, 1934), belonging to the Elaeocarpaceae, it is the third mega- spore from the micropylar end which gives rise to the embryo sac (Fig. 53 J"). Rarely, as in Gloriosa (Afzelius, 1918; Eunus, 1949) (Fig. 53G-I), Ostrya (Finn, 1936), Poa (Hakansson, 1943), and Casuarina (Swamy, 1948) (Fig. 51) all or any of the four megaspores may begin to enlarge and divide. A peculiar condition occurs in Rosularia, Sedum (Mauritzon, 1933), Laurus (Bambacioni-Mezzetti,
78 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
1935), Potentilla (Rutishauser, 1945), and some members of the Ru- biaceae (Fagerlind, 1937), in which the megaspores give out lateral tubes which subsequently begin to grow upward and are in a state of competition with one another (Fig. 54). In Putoria and Galium the tubes enter the tissues of the integument. In Rosularia two to three megaspore tetrads may be formed, and as every megaspore can give rise to a haustorium, the upper part of the nucellus often shows quite a tangle of haustorial processes competing with one another.
Failure of Wall Formation during Meiosis. The functioning of only one megaspore out of four is the commonest condition in angiosperms, but in several plants only the first of the two meiotic divisions is accompanied by wall formation, so that after the meiotic divisions are over each of the dyad cells is binucleate. In others wall formation fails altogether, or if walls are laid down they soon disappear, so that all the four megaspore nuclei lie within the same cell. Such differences in the mode of origin of the megaspores form the basis for a classification of the types of embryo sacs in angiosperms.
References
Afzelius, K. 1916. Zur Embryosackentwicklung der Orchideen. Svensk Bot.
Tidskr. 10: 183-227. . 1918. Zur Entwicklungsgeschichte der Gattung Gloriosa. Acta Horti
Bergiani 6(3): 3-12. . 1924. Embryologische und zytologische Studien in Senecio und ver-
wandten Gattungen. Acta Horti Bergiani 8: 123-219. Archibald, E. E. A. 1939. The development of the ovule and seed of jointed
cactus (Opuntia aurantiaca Lindley). South African Jour. Sci. 36: 195-211. Bambacioni-Mezzetti, V. 1935. Ricerche morfologiche sulle Lauraceae. Lo
sviluppo dell'ovulo e dei sacchi pollinici nel Laurus nobilis L. Ann. di Bot.
21(1): 1-19. Banerji, I. 1940. A contribution to the life history of Costm speciosus Smith.
Jour. Indian Bot. Soc. 19: 181-196. Baranow, P. 1926. Zytologische und embryologische Untersuchungen an Drimi-
opsis maculata Lindl. Ztschr. f. Zellforsch. und Mikros. Anat. 3: 131-148. Benson, M. 1894. Contribution to the embryology of the Amentiferae. I.
Trans. Linn. Soc. London, II. Bot. 3 : 409-424. , Sanday, E., and Berridge, E. 1906. Contribution to the embryology of
the Amentiferae. II. Carpinus betulus. Trans. Linn. Soc. London, II. Bot.
7: 37-44. Berg, 0. 1898. "Beitrag zur Kenntnis der Entwicklung des Embryosackes der
Angiospermen." Diss. Erlangen.
THE MEGASPORANGIUM 79
Bhargava, H. R. 1935. The life history of Trianthema monogyna L. Proc.
Indian Acad. Sci. Sect. B. 2: 49-58. Boehm, K. 1931. Embryologische Untersuchungen an Zingiberaceen. Planta
14: 411-440. Bhaduri, P. N. 1932. The development of ovule and embryo sac in Solanum
melongena L. Jour. Indian Bot. Soc. 11 : 202-224. Boursnell, J. G. 1950. The symbiotic seed-borne fungus in the Cistaceae. Ann.
Bot. 14: 217-243. Buell, M. F. 1935. Seed and seedling of Acorus calamus. Bot. Gaz. 96: 758-765. Caldwell, 0. W. 1899. On the life history of Lemna minor. Bot. Gaz. 27:
37-66. Capoor, S. P. 1937. Contributions to the morphology of some Indian Liliaceae.
II. The gametophytes of Urginea indica Kunth. Beihefte bot. Centbl.
56A: 156-170. Condit, I. J. 1932. The structure and development of flowers in Ficus carica
L. Hilgardia 6: 443-481. Cook, M. T. 1906. The embryology of some Cuban Nymphaeaceae. Bot. Gaz.
42: 376-392. Copeland, H. F. 1938. The Styrax of Northern California and the relationships
of the Styracaceae. Amer. Jour. Bot. 25: 771-780. Corner, E. J. H. 1949. The annonaceous seed and its four integuments. New
Phytol. 48: 332-364. Dahlgren, K. V. O. 1927. Die Morphologie des Nuzellus mit besonderer Beruck-
sichtigung der deckzellosen Typen. Jahrb. f. wiss. Bot. 67: 347-426. . 1939. Endosperm und Embryobildung bei Zostera marina. Bot. Notiser
1939, pp. 607-615. — . 1940. Postamentbildungen in den Embryosacken der Angiospermen.
Bot. Notiser 1940, pp. 347-369. Dodds, K. S. 1945. Genetical and cytological studies of Musa. VI. The
development of female cells of certain edible diploids. Jour. Genet. 46: 161—
179. Earle, T. T. 1938. Embryology of certain Ranales. Bot. Gaz. 100: 257-275. Eunus, A. M. 1949. Contributions to the embryology of the Liliaceae (Gloriosa
superba). Proc. 36th Indian Sci. Cong. Allahabad, Bot. Sect. No. 25. Fagerlind, F. 1937. Embryologische, zytologische und Bestaubungs-experimen-
telle Studien in der Familie Rubiaceae nebst Bemerkungen liber einige Pol-
ploiditats-probleme. Acta Horti Bergiani 11: 195-470. . 1939. Kritische und revidierende Untersuchungen iiber das Vorkommen
des Adoxa ("Lilium")-Typus. Acta Horti Bergiani 13: 1-49. . 1941. Der Bau der Samenanlage und des Makrogametophyten bei
Quisqualis indica. Bot. Notiser 1941, pp. 217-222. ■ . 1945a. Bau der floralen Organe bei der Gattung Langsdorffia. Svensk
Bot. Tidskr. 39 : 197-210. 19456. Bildung und Entwicklung des Embryosacks bei sexuellen und
agamospermischen Balanophora-Arten. Svensk Bot. Tidskr. 39: 65-82.
80 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
Fagerlind, F. 1945c. Bau des Gynoceums, der Samenanlage und de? Embryo- sackes bei einigen Reprasentaten der Familie Icacinaceae. Svensk Bot. Tidskr. 39:346-364.
. 1945<i. Bliite und Bliitenstand der Gattung Balanophora. Bot. Notiser
1945, pp. 330-350.
1947. Gynoceummorphologische und embryologische Studien in der
Familie Olacaceae. Bot. Notiser 1947, pp. 207-230. Finn, W. W. 1936. Zur Entwicklungsgeschichte der Chalazogamen, Ostrya
carpinifolia Scop. Jour. Inst. Bot. Acad. Sci. Ukraine 8: 15-25. Flint, L. H., and Moreland, C. G. 1943. Notes on photosynthetic activity in
seeds of the spider lily. Amer. Jour. Bot. 30: 315-317. Frye, T. C. 1902. A morphological study of certain Ascelpiadaceae. Bot.
Gaz. 34: 389-413. Fuchs, A. 1938. Beitrage zur Embryologie der Thymelaeaceae. Osterr. bot.
Ztschr. 87: 1-41. Goebel, K. 1933. "Organographie der Pflanzen. III. Samenpflanzen." 3d ed.
Jena. Graves, A. H. 1908. The morphology of Ruppia maritima. Trans. Conn. Acad.
Arts and Sci. 14:59-170. Grove, A. R. 1941. Morphological study of Agave lechuguilla. Bot. Gaz. 103:
354-365. Gue>in, P. 1915. Reliquiae Treubinae. I. Recherches sur la structure ana-
tomique de l'ovule et de la graine des Thymeleacees. Ann. Jard. Bot. Buiten-
zorg 14: 1-35. Haberlandt, G. 1923. Zur Embryologie von Allium odorum. Ber. deutsch. bot.
Gesell. 41: 174-179. Hafliger, E. 1943. Zytologisch-embryologische Untersuchungen pseudogamer
Ranunkeln der Auricomus-Gruppe. Ber. schweiz. bot. Gesell. 53: 317-382. Hagerup, O. 1944. On fertilisation, polyploidy and haploidy in Orchis maculatus
L. sens. lat. Dansk Bot. Arkiv 11(5): 1-26. Hakansson, A. 1943. Die Entwicklung des Embryosackes und die Befruchtung
bei Poa alpina. Hereditas 28: 25-61. Haupt. A. W. 1934. Ovule and embryo sac of Plumbago capensis. Bot. Gaz.
95: 649-659. Hofmeister, W. 1861. Neue Beitrage zur Kenntnis der Embryobildung der
Phanerogamen. II. Monokotyledonen. Abh. Konigl. Sachs. Gesell. Wiss.
7: 629-760. Holmgren, J. 1913. Zur Entwicklungsgeschichte von Butomus umbellatus L.
Svensk Bot. Tidskr. 7: 58-77. Houk, W. G. 1938. Endosperm and perisperm of coffee with notes on the mor- phology of the ovule and seed development. Amer. Jour. Bot. 25 : 56-61. Hurst, C. C. 1931. Embryo sac formation in diploid and polyploid species of
Roseae. Proc. Roy. Soc. London Ser. B. 109: 126-148. Johansen, D. A. 1931a. Studies on the morphology of the Onagraceae. V.
Zauschneria latifolia, typical of a genus characterized by irregular embryology.
Ann. N. Y. Acad. Sci. 33: 1-26. . 19316. Studies on the morphology of the Onagraceae. VI. Anogra
vallida. Amer. Jour. Bot. 18: 854-863.
THE MEGASPORANGIUM 81
Juliano, J. B. 1935. Anatomy and morphology of the Bunga, Aeginetia indica
L. Philippine Jour. Sci. 56: 405-451. Junell, S. 1931. Die Entwicklungsgeschichte von Circaeaster agrestis. Svensk
Bot. Tidskr. 25: 238-270. Kajale, L. B. 1940. A contribution to the embryology of the Amarantaceae
Proc. Nat. Inst. Sci. India 6: 597-625. . 1944. A contribution to the life history of Zizyphus jujuba Lamk. Proc
Nat. Inst. Sci. India 10: 387-391. Kausik, S. B. 1935. The life history of Lobelia trigona Roxb. with special refer- ence to the nutrition of the embryo sac. Proc. Indian Acad. Sci. Sect. B.
2: 410-418. . 1940. Structure and development of the ovule and embryo sac of
Lasiosiphon eriocephalus Dene. Proc. Natl. Inst. Sci. India 6: 117-132. Khan, R. 1942. A contribution to the embryology of Jussieua repens L. Jour.
Indian Bot. Soc. 21: 267-282. . 1943. The ovule and embryo sac of Fouquieria. Proc. Natl. Inst. Sci.
India 9: 253-256. Landes, M. 1946. Seed development in Acalypha rhomboidea and some other
Euphorbiaceae. Amer. Jour. Bot. 33: 562-568. Lavialle, P. 1925. Sur la nutrition du sac embryonnaire chez Knautia arvensis.
Compt. Rend. Acad, des Sci. Paris 180: 2055-2056. Lloyd, F. E. 1902. The comparative morphology of the Rubiaceae. Mem.
TorreyBot. Club 8: 1-112. Maheshwari, P. 1934. Contributions to the morphology of some Indian Liliaceae.
I. The gametophytes of Ophiopogon wallichianus Hook f. Proc. Indian
Acad. Sci. Sect. B. 1: 197-204. . 1944. The seed structure of Ipomoea, a criticism. Sci. and Culture 9 :
557. and Gupta, B. L. 1934. The development of the female gametophyte
of Ludivigia parviflora Roxb. and Jussieua repens Linn. Current Sci. [India]
3: 107-108.
and Singh, Balwant. 1943. Studies in the family Alismaceae. V. The
embryology of Machaerocarpus californicus (Torr.) Small. Proc. Natl. Inst.
Sci. India 9: 311-322. Mauritzon, J. 1933. "Studien iiber die Embryologie der Familien Crassulaceae
und Saxifragaceae." Diss. Lund. . 1934. Zur Embryologie der Elaeocarpaceae. Arkiv for Bot. 26A:
1-8. . 1936. Samenbau und Embroyologie einiger Scitameen. Lunds Univ.
Arsskr. N. F. Avd. II, 31(9): 1-31. '
1939. Contributions to the embryology of the orders Rosales and
Myrtales. Lunds Univ. Arsskr. N.F. Avd. II, 35(2): 1-120. Mendes, A. J. T. 1941. Cytological observations in Coffea. VI. Embryo and
endosperm development in Coffea arabica L. Amer. Jour. Bot. 28: 784-789. Murbeck, S. 1902. tJber die Embryologie von Ruppia rostellata. K. Svenska
Vet.-Akad. Handl. 36(5): 1-21. Okeijuk, P. 1937. Vergleichende zytologisch-embryologische Untersuchung der
82 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
Familie Resedaceae. I. Reseda und Astrocarpus. Jour. Inst. Bot. Acad.
Sci. Ukraine 12: 3-81. Orr, M. Y. 1921a. The occurrence of tracheids in the nucellus of Steriphoma
cleomides Spreng. Notes Roy. Bot. Gard. Edinb. 12: 241-242. . 19216. The occurrence of a tracheal tissue enveloping the embryo in
certain Capparidaceae. Notes Roy. Bot. Gard. Edinb. 12: 249-257. Paetow, W. 1931. Embryologische Untersuchungen an Taccaceen, Meliaceen
und Dilleniaceen. Planta 14: 441-470. Pantulu, J. V. 1941. Some unusual megaspore tetrads in the Leguminosae.
Current Sci. [India] 10: 175-176. Pfeiffer, W. M. 1912. The morphology of Leitneria floridana. Bot. Gaz. 53:
189-203. Puri, V. 1941. Life history of Moringa oleifera Lamk. Jour. Indian Bot. Soc.
20: 263-284. Randolph, L. F. 1936. Developmental morphology of the caryopsis in maize.
Jour. Agr. Res. 53: 881-916. Rauch, K. V. 1936. Cytologisch-embryologische Untersuchungen an Scurrula
atropurpurea Dans, und Dendrophthoe pentandra Miq. Ber. schweiz. bot.
Gesell. 45:5-61. Rosendahl, C. O. 1909. Embryo sac development and embryology of Symplo-
carpus foetidus. Minn. Bot. Studies 4: 1-9. Rutishauser, A. 1937. Entwicklungsgeschichtliche Untersuchungen an Thesium
rostratum. Mitt, naturf. Gesell. Schaffhausen 13: 25-47. . 1945. Zur Embryologie amphimiktischer Potentillen. Ber. schweiz.
bot. Gesell 55: 19-32. Schaeppi, H., and Steindl, F. 1942. Bliitenmorphologische und embryologische
Untersuchungen an Loranthoideen. Vtljschr. naturf. Gesell. Zurich 87:
301-372. Schleiden, M. J. 1837. Einige Blicke auf die Entwicklungsgeschichte der vege-
tablischen Organismus bei den Phanerogamen. Wiegmanns Archiv 3:
289-320. Schlimbach, H. 1924. Beitrage zur Kenntnis der Samenanlagen und Samen der
Amaryllidaceen mit Beriicksichtigung des Wassergehaltes der Samen. Flora
117:41-54. Schnarf, K. 1929 "Embryologie der Angiospermen." Berlin. Seshadri Ayyangar, G. 1948. Some observtions on stomata found on cotton
ovules. Indian Cotton Growing Rev. 2: 187-192. Shattuck, C. H. 1905. A morphological study of Ulmus americana. Bot. Gaz.
40: 209-223. Singh, Bahadur. 1950. The embryology of Dendrophthoe falcata (Linn, fil.)
Ettingshausen. Proc. 37th Indian Sci. Cong. Sect. Bot. Smith, C. M. 1929. Development of Dionaea muscipula. I. Flower and seed.
Bot. Gaz. 87:508-530. Soueges, E. C. R. 1919. Recherches sur l'embryogeme des Polygonacees. Bui.
Soc. Bot de France 66: 168-199; 67: 1-11, 75-85.
THE MEGASPORANGIUM 83
Steindl, F. 1945. Beitrag zur Pollen- und Embryobildung bei Cynomorium
coccinewn L. Arch. Julius Klaus-Stift. f. Vererbungsforsch. 20: 342-355. Stenar, H. 1925. "Embryologische Studien. I. Zur Embryologie der Colum-
niferen. II. Die Embryologie der Amaryllideen." Diss. Uppsala. . 1928. Zur Embryologie der Asphodeline-Gruppe. Ein Beitrag zur syste-
matischen Stellung der Gattungen Bulbine und Paradisia. Svensk. Bot.
Tidskr. 22: 145-159. Subba Rao, A. M. 1941. Studies in the Malpighiaceae. II. Structure and de- velopment of the ovules and embryo sacs of Malpighia coccifera Linn, and
Tristellateia australis. Proc. Natl. Inst. Sci. India 7: 393-404. Subramanyam, K. 1948. An embryological study of Melastoma malabathricum
L. Jour. Indian Bot. Soc. 27: 1-9. Svensson, H. G. 1925. "Zur Embryologie der Hydrophyllaceen, Borraginaceen
und Heliotropiaceen." Diss. Uppsala. Swamy, B. G. L. 1948. A contribution to the life history of Casuarina. Proc.
Amer. Acad. Arts and Sci. 77: 1-32. Tomita, K. 1931. Uber die Entwicklung des nackten Embryos von Crinum
latifolium L. Sci. Rpt. Tohoku Imp. Univ. 6: 163-169. Treub, M. 1879. Notes sur l'embryogenie de quelques Orchidees. Natuurk.
Verb. Koninkl. Akad. Amsterdam 19: 1-50. Van Tieghem, Ph. 1901. L'hypostase, sa structure et son role constantes, sa
position et sa forme variables. Bui. Mus. Hist. Nat. 7: 412-418. Whitehead, M. R., and Brown, C. A. 1940. The seed of the spider lily, Hymeno-
callis occidentalis. Amer. Jour. Bot. 27: 199-203. Woodcock, E. F. 1943. Seed development in morning-glory (Ipomoea rubro
caerulea Hook). Papers Mich. Acad. Sci. Arts and Letters 28: 209-212.
CHAPTER 4 THE FEMALE GAMETOPHYTE1
Depending on the number of megaspore nuclei taking part in the development, the female gametophytes of angiosperms may be classified into three main types: monosporic, bisporic, and tetra- sporic. In the first, only one of the four megaspores takes part in the development of the gametophyte. In the second, two mega- spore nuclei take part in its formation; and in the third, all four of them. A further subdivision is based on the number of nuclear divisions intervening between the time of megaspore formation and the time of differentiation of the egg, and the total number of nuclei present in the gametophyte at the moment when such differentiation takes place. A secondary increase in their number, which sometimes takes place at a later stage, is not taken into account in this classi- fication.
The monosporic female gametophytes or embryo sacs fall under two types: 8-nucleate and 4-nucleate. In the development of the 8-nucleate embryo sacs, the first division of the functioning mega- spore gives rise to 2 nuclei : the primary micropylar and the primary chalazal. The second division produces one pair of nuclei at the micropylar end and one at the chalazal, and the third results in two groups of 4 nuclei lying at the opposite poles of the elongated embryo sac. The micropylar quartet differentiates into a three-celled egg apparatus and the upper polar nucleus, and the chalazal quartet into a group of three antipodal cells (or nuclei) and the lower polar nucleus. The 2 polar nuclei fuse to give rise to a secondary nucleus.
This type of embryo sac is the most common and is, therefore, commonly designated as the "Normal type." However, since the others are by no means so infrequent as was once supposed, it will be designated here as the "Polygonum type," for it was in Polygo- num divaricatum that Strasburger (1879) gave the first clear and well-
1 In writing this chapter the author has drawn freely upon some of his review articles (Maheshwari, 1937, 1941, 1946a,6; 1947, 1948), to which reference may be made for fuller information.
84
THE FEMALE GAMETOPHYTE 85
illustrated account of the development of a monosporic 8-nucleate embryo sac.
In certain other monosporic embryo sacs, the megaspore nucleus undergoes only two divisions and a micropylar quartet alone is formed. This quartet gives rise to a normal egg apparatus and a single polar nucleus. The lower polar nucleus and antipodal nuclei are absent. This type of development is known as the "Oenothera type" and has so far been reported only in the family Onagraceae.
The bisporic embryo sacs are typically 8-nucleate ("Allium type") and arise from one of the two dyad cells formed after Meiosis I. Since no wall is laid down after Meiosis II and both the megaspore nuclei formed in the functional dyad cell take part in the develop- ment of the embryo sac, only two further divisions are necessary to give rise to the 8-nucleate stage. A doubtful 4-nucleate type ("Po- dostemon type") has been reported in a few members of the Podo- stemonaceae but this is questionable and will not receive detailed consideration.
The tctrasporic embryo sacs present a great deal of variation. In several cases 16 nuclei are formed as the result of two divisions fol- lowing megasporogenesis. These are classified under the following types, depending on the polarity and organization of the nuclei in the sac: "Peperomia type," "Penaea type," "Drusa type."
In some plants, owing to a crowding of 3 of the megaspore nuclei into the chalazal end of the cell (1 + 3 arrangement), there is a fusion of their spindles in the next division, resulting in a secondary 4-nucleate stage with 2 haploicl nuclei at the micropylar end and 2 triploid ones at the chalazal. The next division results in 8 nuclei, 4 of which are haploid and 4 triploid. This mode of development is known as the "Fritillaria type."
The "Plumbagella type," reported only in Plumbagella micrantha, is similar to the Fritillaria type, except that here the development stops at the secondary 4-nucleate stage, which is at once followed by the organization of the embryo sac.
Finally, there are the "Adoxa" and "Plumbago" types, in both of which the 4 megaspore nuclei divide once to give rise to 8 nuclei. In Adoxa, however, the organization is bipolar and in Plumbago it is tetrapolar.
All the variations of embryo sac development described above are shown diagrammatically in Fig. 55.
86
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
Type
Megasporogenesis
Megaspore mother cell
Division I
Division II
Megagametogenesis
Division III
Division IV
Division V
Mature embryo sac
Monosporic
8 -nucleate
Polygonum type
Monosporic
4 -nucleate
Oenothera type
Bisporic 8 -nucleate Allium type
Tetrasporic
16 -nucleate
Peperomia type
Tetrasporic 16 -nucleate Penaea type
Tetrasporic 16 -nucleate Drusa type
Tetrasporic
8 -nucleate
Fritillaria type
Tetrasporic
8 -nucleate
Plumbagella type
Tetrasporic
8 -nucleate
Plumbago type
Tetrasporic 8 -nucleate Adoxa type
^
^
©0\ /©© ©«,
© 01 I© ©
8
§
nfetj
w
Fig. 55. Diagram showing important types of embryo sacs in angiosperms.
THE FEMALE GAMETOPHYTE 87
MONOSPORIC EMBRYO SACS
Polygonum type. The monosporic 8-nucleate embryo sac, formed by three divisions of the functioning megaspore, occurs in at least 70 per cent of the angiosperms now known. The enlargement of the megaspore is always accompanied by increased vacuolation, one large vacuole usually appearing on either side of the nucleus in the direction of the long axis of the cell (Fig. 56 A). After the first division has taken place, the two daughter nuclei move apart to opposite poles. Most of the cytoplasm is aggregated around them and the rest forms a thin peripheral layer, the center being occupied by a large vacuole (Fig. 565). The next division gives rise to a 4-nucleate stage (Fig. 56C) which is followed by the 8-nucleate stage comprising a micropylar and a chalazal quartet.
Of the 8 nuclei arising in this manner, 3 at the micropylar end give rise to the egg and two synergids; 3 at the chalazal end give rise to antipodal cells;2 and the remaining 2, one from each pole, fuse in the center to form a secondary nucleus (Fig. 56/)).
Occasionally embryo sacs are found with less than the normal quota of 8 nuclei. This is usually because of an early degeneration of the antipodals, which obscures the true nature of the embryo sac. Even when the antipodals are present, they are sometimes overlooked because of their being situated in the narrow chalazal end of the embryo sac, which is seen only in median sections (Puri, 1939, 1941).
In some cases there is a genuine reduction in the number of nuclei. In certain species of Phajus, Corallorhiza, Broughtonia (Sharp, 1912), Chamaeorchis, Oncidium (Afzelius, 1916), Elatine (Frisendahl, 1927) (Fig. 57 A), Thesium (Rutishauser, 1937a), Calypso (Stenar, 1940), and Bulbophyllum and Geodorum (Swamy, 1949a) the embryo sacs are 6-nucleate owing to a suppression of division of the two chalazal nuclei of the 4-nucleate stage. In Orchis morio (Afzelius, 1916) the primary chalazal nucleus of the 2-nucleate stage may degenerate without undergoing any division, so as to result in a 5-nucleate embryo sac.
A reduction in the number of nuclei may also be brought about in a different way. In Epipactis pubescens (Brown and Sharp, 1911)
8 In several plants, like Thesiiwi rostratum (Rutishauser, 1937a), cell formation does not occur at the chalazal end and the antipodal nuclei remain free.
D C
Fig. 56. Development of embryo sac in Hydrilla verticillat a. A, tetrad of mega- spores with chalazal cell functioning. B,C, two-nucleate and four-nucleate embryo sacs. D, mature embryo sac; synergids have degenerated.
88
THE FEMALE GAMETOPHYTE
89
and Paphiopedilum insigne (Afzelius, 1916) it has been noted that sometimes the two chalazal spindles of the last division come to lie more or less parallel and very close to each other and eventually coalesce to form a single large spindle which produces 2 diploid nuclei instead of the 4 haploid ones which would have been formed
B C
Fig. 57. Embryo sacs with fewer or more than eight nuclei. A, Elatine triandra, six-nucleate embryo sac which has arisen by omission of last division at chalazal end; the two black masses represent degenerated synergids. B, E. hydropiper, fusion of two eight-nucleate embryo sacs, resulting in 16-nucleate compound embryo sac. (After Frisendahl, 1927.) C, Sandoricum koetjape, embryo sac containing cytoplasmic vesicle with several nuclei. (After Juliano, 1934-)
in the ordinary way. This results in a 6-nucleate embryo sac with a haploid micropylar quartet, a diploid lower polar nucleus, and a single diploid antipodal cell (Fig. 58).
The reverse condition, i.e., the occurrence of more than 8 nuclei in the embryo sac, is less frequent and may arise in three ways: (1) fusion of two embryo sacs (2) migration of the nuclei of nucellar cells into the embryo sac ; and (3) occurrence of secondary divisions of some of the first-formed 8 nuclei.
90
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
In plants having a multicellular archesporium, several megaspore tetrads are formed and a number of megaspores may begin to en- large. As examples may be mentioned the Casuarinaceae, Loran- thaceae, and Rosaceae, and some members of the Rhamnaceae, Rubiaceae, and Compositae.3 Commonly most of the embryo sacs become arrested in their development at a comparatively early stage and only a few reach maturity. Rarely the separating walls between the sacs may dissolve so that the contents become included
A B C D E
Fig. 58. Some stages in development of embryo sac of Epipactis pubescens. A, telophase of last division in embryo sac; note that two chalazal spindles show tend- ency to lie parallel to each other. B, metaphase of similar division; two chalazal spindles have coalesced to form single large spindle. C,D, later stages of division. E, mature embryo sac, showing egg apparatus, two polar nuclei, and single antip- odal cell. (After Brown and Sharp, 1911.)
in a common cavity. A very good example of this kind has been figured in Elatine hydropiper (Frisendahl, 1927), showing an embryo sac with two egg apparatuses, two pairs of polar nuclei, and two groups of three antipodal cells each (Fig. 57 B). This must clearly have originated by a fusion of two normally growing sacs. Similar "compound" sacs have been noted by Oksijuk (1937) in Reseda alba and R. inodora. Sometimes he found less than 16 nuclei, which is
3 In Potentilla, heptaphylla (Rutishauser, 1945) as many as nine embryo sacs were seen in one ovule.
THE FEMALE GAMETOPHYTE 91
quite possible if one of the fusing gametophytes is at a younger stage of development than the other.
In a Musa variety known as "I.R. 53," Dodds (1945) has re- cently described one compound embryo sac with three egg appa- ratuses and two pairs of polar nuclei; and another with two egg apparatuses, one pair of polar nuclei, and an additional group of 7 large "polar-like" nuclei at the chalazal end. Juliano (1934) has figured a peculiar embryo sac in a fallen flower of Sandoricum koetjape with a normal egg apparatus, two polar nuclei, and a large cytoplasmic vesicle extending from the chalazal end of the sac to its middle and containing more than a dozen nuclei (Fig. 57C). Since the antipodals are very ephemeral in this species, it is con- sidered probable that the embryo sac proper was formed from the third megaspore and that the multinucleate vesicle arose as a result of some free nuclear divisions in the fourth megaspore.4
In some plants there is a migration of the nucellar nuclei into the embryo sac. This migration is due to the fact that during the growth and enlargement of the latter, the adjacent cells of the nucellus become flattened and crushed. Their walls, which are very thin and delicate, get ruptured, and the contents — both cyto- plasm and nuclei, or only the latter — may "wander" into the em- bryo sac and become incorporated in it.5 Two instances of this nature deserve special mention. In Hedychium gardnerianum (Madge, 1934) the nuclei of the nucellar cells lying just below the hypostase migrate "from cell to cell" through a small hole in the walls until they reach the hypostase. Here their progress is stopped for a time and groups of 20 or 30 nuclei collect together, surrounded by the ragged cell walls of the ruptured cells. Some of the nuclei now make their way around the hypostase into the cavity of the embryo sac, where they are believed to serve a nutritive function. In Pandanus (Fagerlind, 1940), which has no thick-walled hypos- tase, the nucellar cells lying directly beneath and on the sides of the young embryo sac show a marked tendency to enlarge. Their nuclei become swollen and the plasma assumes an appearance simi-
4 Another possibility, not mentioned by Juliano, is that the embryo sac proper arose normally from the chalazal megaspore and the vesicle was of aposporic origin.
5 This is comparable to the condition in many gymnosperms in which the nuclei of the jacket cells often make their way inside the egg.
92 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
lar to that of the embryo sac (Fig. 59A). The enlarged nuclei eoon approach the embryo sac wall, which becomes perforated at such points. Gradually the pores become wider and finally the entire separating wall is absorbed. The embryo sac now encroaches upon these areas and soon incorporates them, coming in contact with newer cells which may also meet the same fate (Fig. 595,(7) .6 Even at the 4-nucleate stage, as many as 10 or more nucellar cells may become included inside the embryo sac in this fashion. Their nuclei divide synchronously with the sac nuclei, resulting in the formation of 8 haploid and a variable number of diploid nuclei (Fig. 59 D). The secondary nucleus attains varying degrees of poly- ploidy, depending not only on the number of the nuclei which take part in the fusion but also on their chromosome content. In the mature embryo sac (Fig. 59F), the egg apparatus contains haploid nuclei only; some of the antipodal cells contain haploid nuclei, others diploid; most of the lateral cells (Fig. 59E) are diploid.
The third possibility, i.e., an increase in the number of nuclei caused by further divisions of the original nuclei of the sac, is rare except with regard to the antipodal nuclei or cells, for which see page 134. To mention some examples from recent literature, in Crassula schmidtii and Umbilicus intermedins (Mauritzon, 1933) it is reported that occasionally there is a fourth division in the embryo sac, resulting in the formation of 16 nuclei, which organize to form four synergids, two eggs, six antipodal cells, and four polar nuclei. In Crepis capillaris (Gerassimova, 1933) some supernu- merary egg cells were occasionally seen in addition to the other and usual components of the embryo sac, but their origin could not be traced and eventually they were found to degenerate and disappear. In Nicotiana, Goodspeed (1947) has recently reported some embryo sacs having 9 to 16 nuclei, — "obviously the result of division of from one to all of the normal eight nuclei." Here 3 to 5 nuclei were found to take part in polar fusion.
Special mention may be made of the development of the embryo sac in Balanophora and Langsdorffia (Fagerlind, 1945a,6). In both cases the micropylar megaspore functions and the three chalazal mega spores degenerate at a very early stage, although their re-
6 Harling (1946) reports that in Carludovica the nucellar cells enlarge and push against the wall of the embryo sac, but in this case their contents do not actually enter the sac.
D E F
Fig. 59. Development of embryo sac in Fandanus. A, P. ornatus, four-nucleate stage; note enlargement of nucellar cells in chalazal region. B, same, with some nucellar cells incorporated inside embryo sac. C, P. dubius, four-nucleate stage, showing nucellar nuclei entering into embryo sac. D, P. oleiocephalus, embryo sac, showing several nuclei some of which are apparently derived from nucellus. E, same, mature embryo sac, showing two lateral cells and supernumerary polar nuclei. F, P. ornatus, mature embryo sac, showing large number of antipodal cells. (After Fagerlind, 1940.)
93
94
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
mains can be recognized for a considerable time (Fig. 60B,C; Fig. 61A-C). In Balanophora even the dyad cells show a marked dif- ference in size, the micropylar cell being much larger than the chalazal (Fig. GO A). At the 2- or 4-nucleate stage a tubular out-
C ' — ' F v G
Fig. 60. Development of embryo sac in Balanophora elongata. A, dyad cells undergoing second meiotic division. B, functioning megaspore, with remains of three degenerating megaspores at its base. C,D, two-nucleate embryo sacs; note lateral outgrowth from basal part of embryo sac in D. E, four-nucleate embryo sac. F, eight-nucleate embryo sac; note that egg apparatus has organized in morphologically lower end of embryo sac. G, older stage of same, showing fusion of four nuclei at antipodal end. (After Fagerlind, 1945a.)
growth arises from the embryo sac and then grows upward. In Balanophora it originates near the basal end of the sac (Fig. 60D) and in Langsdorffia near its apical end (Fig. 61 A). In both cases it grows very quickly and soon comes to lie at a higher level than the originally upper end of the embryo sac (Fig. QOE, Fig. 61B,C). The 4 nuclei of the sac now undergo the last division to form the
THE FEMALE GAMETOPHYTE
95
usual 8 nuclei, of which those belonging to the morphologically basal end give rise to the egg apparatus and one polar nucleus (Fig. 60 G) and those belonging to the upper end fuse to form an irregularly lobed nucleus which usually degenerates in situ. A sim- ilar fusion takes place in Langsdorffia (Fig. 61F-7) except that some-
^-'-•■■''•v.:-".Oa:\; 3 •v*.*:.r 5
V
B ^ C \2») D
Fig. 61. Development of embryo sac in Langsdorffia hypogaea. A, four-nucleate stage; note three degenerating megaspores at lower end and formation of lateral protuberance near upper end. B, older stage, showing entry of two basal nuclei of sac into lateral arm. C, more advanced stage, showing pronounced upward growth of lateral arm, which is now situated at a higher level than morphologically upper end of sac; note three degenerating megaspores at lower end. D, two upper nuclei of the sac, dividing. E, mature embryo sac. F-I, stages in fusion of four nuclei belonging to antipodal end of embryo sac. (After Fagerlind, 1945b.)
96 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
times the last division fails to occur at this end of the embryo sac (Fig. 6 ID), so that only 6 nuclei are formed (Fig. 61 E).
Oenothera Type. About a hundred years ago, Hofmeister (1847, 1849) published some remarkably accurate figures of the embryo sac of a few members of the Onagraceae, but because of the crude technique of those days he was unable to give a full account of the development. Geerts, in 1908, found that in Oenothera lamarckiana the embryo sac is usually formed by the micropylar megaspore of the tetrad, which undergoes only two nuclear divisions instead of the usual three occurring in the Polygonum type of embryo sac. In this way, 4 nuclei are produced which organize into the two synergids, the egg, and a single polar nucleus. Since the third division is omitted and all the nuclei are situated in the micropylar part of the developing embryo sac, there is neither a lower polar nucleus nor any antipodal cells. Modilewski (1909) independently studied species of Oenothera, Epilobium, and Circaea and confirmed the observations of Geerts in all essential respects. These two in- vestigations were soon followed by several others and this mode of development, known as the Oenothera type, has been found to be a characteristic and constant feature of the entire family Onagraceae, having been demonstrated in more than 16 genera. The only ex- ception is Trapa, which has an 8-nucleate embryo sac of the Polygonum type, but this genus, as most systematists now agree, is best assigned to a separate family, the Hydrocaryaceae or Trapaceae.
A noteworthy feature in the development of the Oenothera type of embryo sac is the concurrent growth of more than one cell of the tetrad. Eventually it is the micropylar megaspore which func- tions, but sometimes it may be the chalazal and occasionally both grow simultaneously forming "twin" embryo sacs (Fig. 62).
Rarely, more than 4 nuclei may be seen in an embryo sac. Usu- ally this condition results from the incorporation of an adjacent megaspore and its contents, but it appears that sometimes there may be further division or divisions of the nuclei of the embryo sac. In Anogra pallida7 Johansen (1931a) reported repeated ami-
7 This plant is a native of the arid regions of southern Arizona and California. It shows little or no seed production, and propagation occurs by means of offshoots at the ends of subterranean stolons.
THE FEMALE GAMETOPHYTE
97
totic divisions of the polar nucleus, and in one embryo sac as many as 140 nuclei were formed by this method. In a few instances he found a synergid containing about 20 nuclei. In Zauschneria lati- folia (Johansen, 19316) the nuclei of the nucellar cells are said to
B C D E F
Fig. 62. Development of embryo sac in Oenothera suaveolens. A , tetrad of mega- spores; both micropylar and chalazal megaspores are enlarging. B, embryo sac formed from micropylar megaspore; the three chalazal megaspores in process of degeneration. C, embryo sac formed from micropylar megaspore; one of chalazal megaspores has also developed up to two-nucleate stage. D-F, twin embryo sacs formed by concurrent growth of two megaspores. (After Hoeppener and Renner, 1929.)
migrate into the embryo sac to form a variable number of bodies looking like micronuclei of different sizes.
Embryo sacs with fewer than 4 nuclei are rare, but in Hartmannia tetraptera (Johansen, 1929) and Jussieua repens, Khan (1942) saw two 3-nucleate embryo sacs having a single synergid, an egg, and a polar nucleus. Their origin is probably to be explained by a lack of division of the primary synergid nucleus of the 2-nucleate stage.
98
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
BISPORIC EMBRYO SAC
Allium Type. A bisporic embryo sac was first described in Al- lium fistulosum (Strasburger, 1879) and has since been confirmed in several species of this genus (Weber, 1929; Messeri, 1931; Jones and Emsweller, 1936; and others). The megaspore mother cell (Fig. 63 A) divides to form two dyad cells, of which the upper is much smaller and soon degenerates (Fig. 63B). The nucleus of the lower divides to form 2 (Fig. 63C), 4 (Fig. 63D) and then 8 nuclei, which give rise to an embryo sac with the usual organization.
Fig. 63. Early stages in development of embryo sac of Allium cepa. A, mega- spore mother cell. B, dyad cells, upper degenerating. C, two-nucleate embryo sac. D, four-nucleate embryo sac. {After Jones and Emsweller, 1936.)
Treub and Mellink (1880) independently described the same type of development in Agraphis patula (= Scilla hispanica), and this has also been found to be true of other species of Scilla (see Hoare, 1934). The chief difference between Allium and Scilla lies in the fact that while in Allium it is the lower dyad cell which gives rise to the embryo sac, in Scilla it is usually the upper. The lower does not degenerate at once, however, but often develops up to the 4-nucleate stage forming the so-called "antigone," which probably serves for the nutrition of the functional embryo sac.
During the last seven decades the Allium type has been reported in several plants belonging to diverse groups and it appears to be quite characteristic of certain families, viz., Podostemonaceae, Bu-
THE FEMALE GAMETOPHYTE 99
tomaceae (except Butomus), Alismaceae, and the tribe Viscoideae of the Loranthaceae. It is also found in several members of the Balanophoraceae, Liliaceae, Amaryllidaceae, and Orchidaceae, but in other groups its occurrence is more or less sporadic.
The chief variation in development is a tendency toward reduc- tion in the number of nuclei at the chalazal end. This has been very clearly demonstrated in the Alismaceae (Dahlgren, 19286, 1934; Johri, 1935a,6,c, 1936a; Maheshwari and Singh, 1943), Butoma- ceae (Johri, 19366, 1938a, 6), Podostemonaceae (Went, 1910, 1912, 1926), and some members of the Orchidaceae (see Swamy, 1949a). In the Alismaeae, of which Machaerocarpus calif ornicus (Maheshwari and Singh, 1943) may be cited as an example (Fig. 64), the development usually proceeds normally up to the 4-nucleate stage. After this only the 2 micropylar nuclei divide again, re- sulting in a 6-nucleate stage comprising an egg apparatus, two polar nuclei, and a single antipodal nucleus. In those plants in which reduction has gone still further, only 5 nuclei are formed, four at the upper end and the undivided primary chalazal nucleus at the lower. The mature embryo sac therefore comprises an egg apparatus, an upper polar nucleus, and a single antipodal nucleus; a lower polar nucleus is absent.
Special mention may be made of a few plants following the Allium type of development.
In the tribe Viscoideae, belonging to the Loranthaceae, this mode of development seems to be of general occurrence and has recently been described in some detail in Ginalloa (Rutishauser, 19376), Korthalsella (Rutishauser, 1935, 19376 ; Schaeppi and Steindl, 1945), and Viscum (Steindl, 1935; Schaeppi and Steindl, 1945). In all these genera the central ovarian papilla has two or more arche- sporial cells, each of which divides to form two dyad cells (Fig. Q5A-C). Of these, the upper dyad cell is the larger and functions, while the lower soon degenerates (Fig. 65 D). A peculiar feature is that after the 4-nucleate stage there is a slow but steady curva- ture of the embryo sac, which causes its lower end to bend out of the papilla and proceed upward into the carpellary tissue (Fig. 65£"). Meanwhile, the 4 nuclei divide to form 8, one quartet being situated at each pole of the embryo sac. The egg apparatus dif- ferentiates in the originally lower pole, which is, however, now situated at a higher level than the upper (Fig. 65F).
100
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
In Convallaria majalis (Stenar, 1941) the first division of the megaspore mother cell results in the formation of the usual dyad cells (Fig. QQA,B). Walls are also laid down after the second division (Fig. 66C), but these soon break down so that the dyad cells are again restored although each of them is now binucleate
B
D
Fig. 64. Development of embryo sac in Machaerocarpus calif ornicus. A, mega- spore mother cell. B, dyad stage; nucleus of lower dyad cell dividing. C, two- nucleate embryo sac with remains of degenerated upper dyad cell. D, four- nucleate embryo sac. E, six-nucleate embryo sac; lower two nuclei of four-nucleate stage have remained undivided. F, mature embryo sac, showing two synergids, egg, two polar nuclei, and single antipodal nucleus. (After Maheshwari and Singh, 1948.)
THE FEMALE GAMETOPHYTE
101
(Fig. 66D). The micropylar dyad cell is at first the larger and more vacuolated (Fig. QQE) but gradually the chalazal dyad cell increases in size and plays the more dominant role (Fig. 66F). The
E
Fig. 65. Development of embryo sac in Korthahella dacrydii. A, Is. central papilla. B, portion of older papilla, showing a megaspore mother cell. C, telo- phase of Meiosis I. D, two-nucleate embryo sac formed from upper dyad cell; note degenerating lower dyad cell. E, central papilla showing two four-nucleate embryo sacs; note beginning of curvature in embryo sac on right. F, mature embryo sac; egg apparatus has differentiated in originally basal end, which has now penetrated upward into tissues of carpel. (After Rutishauser, 1935.)
2 nuclei of this cell divide to form 4 (Fig. ffiG-H) and then the 8 nuclei of the mature stage (Fig. 667). The interesting point in the development is that it starts like that of a monosporic form
102 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
but is actually bisporic as a result of an early dissolution of the cell walls laid down after the second meiotic division.
In 1907, Pace published an interesting paper on the develop- ment of the embryo sac in four species of Cypripedium. According
G H
Fig. 66. Development of embryo sac in Convallaria majalis. A, l.s. nucellus, show- ing megaspore mother cell. B, dyad stage. C, T-shaped tetrad. D, wall sepa- rating the two megaspores of each dyad cell has disappeared. E, the two dyad cells, each binucleate. F, upper dyad cell degenerating; lower enlarging. G, four-nucleate embryo sac formed from lower dyad cell. H, same, more advanced stage. /, mature embryo sac. (After Stenar, 1941.)
THE FEMALE GAMETOPHYTE 103
to her account, the megaspore mother cell divides to form two dyad cells, of which the lower develops normally up to the 4-nu- cleate stage. One of the micropylar nuclei is now said to form the egg and the other a synergid; the second synergid is formed by one of the chalazal nuclei which migrates upward; and the remaining nucleus functions as the single polar. At the time of fertilization, one of the synergid nuclei is said to become displaced by the in- coming pollen tube and forced, as it were, to take part in triple fusion. Owing to its unique and distinctive nature, this mode of development was designated as the "Cypripedium type."
The reinvestigations made by Prosina (1930), Francini (1931), Carlson (1945), and Swamy (1945) have, however, shown that the development does not end at the 4-nucleate stage but continues further. Occasionally all 8 nuclei may be formed, but in any case at least the 2 micropylar nuclei go through the next division, so that the embryo sacs are 6 nucleate.
The ovules and embryo sacs of the Podostemonaceae show several interesting features to which a brief reference may be made here, using Podostemon ceratophyllum (Hammond, 1937) as an example. The outer integument appears first and forms the micropyle (Fig. 67 A). The megaspore mother cell (Fig. 67B,C), which is situated directly below the epidermis, divides to form the two dyad cells (Fig. 67D), of which the micropylar soon aborts although its nucleus may occasionally divide (Fig. &7E). The nucleus of the chalazal dyad cell divides to form 2 nuclei (Fig. 67F-H), of which the lower promptly degenerates and disappears (Fig. 67 G). The remaining nucleus undergoes two divisions, resulting in 4 nuclei (Fig. 677), which organize to form two synergids, an egg, and a polar nucleus. Occasionally the primary chalazal nucleus persists up to this stage so that the 5-nucleate nature of the embryo sac is easily recognized. More commonly, however, only 4 nuclei are seen and the fifth is no longer recognizable at this stage (see also Razi, 1949).
The following members of the Podostemonaceae are reported to have tetranucleate embryo sacs: Podostemon subulatus, Hydrobium (= Zeylanidium) olivaceum, Farmeria metzgerioides (Magnus, 1913), and Weddelina squamulosa (Chiarugi, 1933). Here the lower dyad cell is said to undergo only two divisions, resulting in 4 nuclei which organize into the egg apparatus and a single polar nucleus (Fig. Q7J-K, QSA-F). This type of development, sometimes called
E F I J
Fig. 67. Development of embryo sac in Podostemon ceratophyllum (A-I) and Weddelina squamulosa (J,K). A, Podostemon, l.s. ovule, diagrammatic. B, l.s. young ovule, showing archesporial cell. C, older stage, showing formation of psuedo embryo sac by disintegration of nucellar cells lying just below the mega- spore mother cell. D, formation of dyad cells. E, degeneration of upper dyad cell. F, two-nucleate embryo sac; note enlarging pseudo embryo sac. G, two- nucleate embryo sac; primary chalazal nucleus disorganizing. H, primary micro- pylar nucleus divided into two daughter nuclei; primary chalazal nucleus has disap- peared. J, five-nucleate stage in which the primary chalazal nucleus has degenerated and disappeared. (After Hammond, 1937.) J,K, Weddelina, stages corresponding to I. (After Chiarugi, 1983.)
104
I J K
Fig. 68. Development of embryo sac in Podostemon subidatus (A-F) and Dicraea elongata (G-K). A, Podostemon, megaspore mother cell. B, dyad cells. C, lower dyad cell enlarging; upper in course of degeneration. D, two-nucleate embryo sac formed from lower dyad cell. E, four-nucleate stage with accompanying wall formation. F, mature embryo sac showing two synergids, egg, and a polar nucleus. G, Dicraea, upper dyad cell degenerating; lower divided into two cells. H-J, upper dyad cell crushed and disorganized; of the other two cells, upper has divided trans- versely and lower has divided vertically. K, embryo sac after fertilization, show- ing degeneration of all the cells except the zygote. (After Magnus, 1913.)
105
106 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
the "Podostemon type", has, however, always been considered doubtful, and the four plants named above deserve to be re- investigated.
An even more doubtful case is that of Dicraea elongata (Magnus, 1913), in which the chalazal dyad cell is said to divide transversely to form two cells (Fig. 68G). Of these the upper, which is larger, again divides in the same plane (Fig. 68//,/) to produce one syn- ergid and an egg cell, and the lower divides anticlinally to form two antipodal cells (Fig. 68/). According to this interpretation the polar nuclei are absent, and all the cells except the zygote degenerate after fertilization (Fig. 68/v). These observations need to be confirmed before they can be accepted.
TETRASPORIC EMBRYO SACS8
Peperomia Type. Campbell (1899«,o; 1901) and Johnson (1900) reported that in Peperomia pellucida each of the 4 megaspore nuclei divides twice, resulting in a total of 16 nuclei which become more or less uniformly distributed in the rather thick layer of cytoplasm lying at the periphery of the embryo sac. According to Johnson, 2 nuclei at the micropylar end now become organized to form the egg and a synergid, 8 fuse to form the secondary nucleus, and the remaining 6 are cut off at the periphery of the embryo sac. Ac- cording to Campbell, on the other hand, 1 to 3 nuclei in the vicinity of the egg show a more or less evident aggregation of cytoplasm around them and are to be regarded as the equivalents of syn- ergids; approximately 8 nuclei enter into the formation of the sec- ondary nucleus; and the remaining 4 to 6 nuclei are cut off as antipodal cells.
Subsequent studies, made by others on several species of Peperomia, have confirmed Johnson's account. The chief varia- tions concern the number of nuclei which fuse to form the secondary nucleus, and the number left over to form the antipodals. In every case only one synergid was observed.
A recent study of Peperomia pellucida (Fagerlind, 1939a) has shown that after the meiotic divisions are over (Fig. 69 A-C), the coenomegaspore9 may either retain its more or less spherical form or become slightly pear-shaped with a little protuberance at the
8 See Fagerlind (1944) for fuller information on tetrasporic embryo sacs.
9 This term is used to denote the cell containing the four free megaspore nuclei.
THE FEMALE GAMETOPHYTE
107
micropylar end. The 4 mega spore nuclei are usually arranged tet- rahedrally, but in the embryo sacs of the pear-shaped type one nucleus projects rather conspicuously towards the upper papillate
Fig.
I K L
69. Development of embryo sac in Peperomia pellucida.
A, megaspore
mother cell. B, two-nucleate stage. C, four-nucleate stage. D,E, eight-nucleate stage as seen in spherical and pear-shaped tj^pes of embryo sacs respectively. F,G, 16-nucleate stage. H, mature embryo sac of spherical type, showing a single synergid, six lateral cells, and eight polar nuclei (all nuclei of sac are not seen in this section). 7, same, showing one lateral cell in close proximity to egg and there- fore simulating a second synergid. J-M, successive sections through a pear-shaped embryo sac, showing a three-celled egg apparatus. (After Fagerlind, 1939a.)
end. In the next stage, the 8 nuclei are either distributed more or less symmetrically around the periphery of the embryo sac (Fig. 69D), or 2 nuclei may lie somewhat closer to each other at its upper end (Fig. 692?) . The fourth and the last division now gives rise to
108 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
16 nuclei, which may form either eight groups of 2 nuclei each (in the spherical embryo sacs) (Fig. 69F) or six groups of 2 and a micro- pylar group of 4 nuclei (in the pear-shaped embryo sacs) (Fig. QQG). In the former case the egg apparatus is usually two-celled (egg and one synergid) ; 8 nuclei fuse in the center to form the secondary nucleus; and 6 nuclei are cut off at the periphery (Fig. 69H). Only occasionally, because of slight displacements and the small size of the embryo sac, one may find another peripheral cell lying so close to the egg that the egg apparatus may be said to comprise three cells (Fig. 697). In the pear-shaped embryo sacs, however, a three-celled egg apparatus is the rule, the fourth nucleus from the micropylar end and one member from each of the six peripheral pairs form the seven polars, and 6 nuclei are cut off to form the lateral cells (Fig. 69K-M).
Fagerlind's observations help us to understand the slight diver- gence between the account of Johnson (1900) and that of Campbell (1899a,6; 1901). The former saw only one synergid, while the latter believed that there were more than one. Now it appears that both these conditions are possible, depending on the form which the embryo sac takes during its growth and development. In the pear-shaped type there are invariably two synergids; in the spherical type there is usually only one synergid unless another peripheral cell accidentally happens to lie so close to the egg as to look like a second synergid.
Johnson (1914) discovered a different type, however, in P. his- pidula. At the 8-nucleate stage, 2 nuclei are seen at the micro- pylar and 6 at the chalazal end (Fig. 70A); at the 16-nucleate stage, 4 lie at the micropylar end and 12 at the chalazal (Fig. 70B). Two nuclei of the micropylar group now form the egg and single synergid, as in other species, but the remaining 2 nuclei of this group and all the remaining 12 nuclei meet near the center and fuse to form a single large secondary nucleus (Fig. 70C,D).
The embryo sac of Gunnera (Haloragidaceae) is essentially similar to that of Peperomia pcllucida. Two species have been studied: G. macrophylla (Ernst, 1908; Samuels, 1912) and G. chilensis (Modilewski, 1908). After the 16-nucleate stage, 3 of the micro- pylar nuclei form the egg apparatus, the fourth descends and fuses with 6 other nuclei tu form a large secondary nucleus, and the re- maining 6 are cut off as antipodal cells. It is possible that if the
THE FEMALE GAMETOPHYTE
109
C D
Fig. 70. Some stages in development of embryo sac of Peperomia hispidula. A, eight-nucleate embryo sac; nuclei shown in dotted outline have been included from adjacent sections of sac. B, l.s. ovule, showing 16-nucleate embryo sac. C, embryo sac, showing egg, one synergid, and 12 polar nuclei. D, embryo sac with egg, single synergid, and large lobed secondary nucleus. (After Johnson, 1914-)
110 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
embryo sac of Gunnera were to be studied again and a sufficient quantity of material examined, it would show a range of variation similar to that in Peperomia.
Penaea Type. Stephens (1909) described an interesting mode of development in three genera of the Penaeaceae, viz., Penaea, Br achy 'siphon, and Sarcocolla (Fig. 71). Here the 16 nuclei lie in four distinct quarters which are arranged crosswise, one at each end of the embryo sac and two at the sides. Now 3 nuclei of each quartet become cut off as cells, while the fourth remains free and
^i&te
L^rMPi
Fig. 71. Development of embryo sac in Penaeaceae. A, Sarcocolla minor, four megaspore nuclei at close of the second meiotic division. B, S. formosa, eight- nucleate stage. C, S. squamosa, 16-nucleate stage, showing four groups of four nuclei each. D, Penaea mucronata, mature embryo sac. {After Stephens, 1909.)
moves to the center. There are thus four "triads" and four polar nuclei. As a rule, the egg cell of the micropylar "triad" alone is functional, although the others often look very similar.
Embryo sacs of this type have since been described in several members of the Malpighiaceae (see Stenar, 1937; Subba Rao, 1940, 1941) and Euphorbiaceae (Modilewski, 1910, 1911; Arnoldi, 1912; Tateishi, 1927; and others) and in a few scattered genera belonging to other families.
Special mention may be made of the embryo sac of Acalypha indica (Maheshwari and Johri, 1941), which, although similar, does not entirely fit into the type described above. Up to the 16-nu-
THE FEMALE GAMETOPHYTE
111
cleate stage (Fig. 72A-D) the development corresponds with that of the Penaeaceae and other species of Acalypha, but the organiza- tion of the mature embryo sac presents a great variation. The commonest condition found was that 2 nuclei of each quartet re- main free and migrate to the center of the embryo sac, while the other two organize into cells. Thus there are four groups of two cells each at the periphery and 8 free nuclei in the center (Fig. 72E).
Fig. 72. Development of embryo sac in Acalypha indica. A, megaspore mother cell with four megaspore nuclei. B, megaspore nuclei in division. C, eight- nucleate stage. D, sixteen-nucleate stage. E, mature embryo sac, showing four peripheral pairs of cells and eight polar nuclei. (After Maheshwari and Johri,
mi.)
This was not the only kind of organization, however. Some ovules showed a micropylar group of three cells and three other groups of two cells each, leaving only 7 nuclei (instead of the usual 8) to fuse in the center. In one embryo sac, the chalazal group had three cells and all the rest had two cells each. Another em- bryo sac showed three two-celled groups, one lateral cell, and 9 free nuclei meeting in the center. In a third and very peculiar embryo sac, one lateral group was entirely missing, the second had only one cell, the micropylar had three cells, and the chalazal had two cells, leaving 10 free nuclei to take part in polar fusion.
A few cases were noted in which it seemed that fewer than 16 nuclei had been formed, and others with slightly more than this
112 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
number. These counts could not be regarded as certain, however, since such embryo sacs ran into three or four sections and their exact reconstruction was a matter of doubt.
Abnormalities of a somewhat similar nature have also been re- corded in Combretum (Mauritzon, 1939), but it has to be seen how far these are related to differences of environment.
Drusa Type. A 16-nucleate embryo sac of a different nature was recorded by Hakansson (1923) in Drusa oppositifolia, a member of the family Umbelliferae (Fig. 73 A). After the meiotic divisions are over, three of the megaspore nuclei pass down to the basal end of the embryo sac, and only one remains at the micropylar end. This 1+3 arrangement is followed by a 2+6 and then a 4 + 12 stage. The four micropylar nuclei give rise to the egg apparatus and upper polar nucleus, and the 12 chalazal nuclei to a lower polar nucleus and 11 antipodal cells.
During recent years this type of development has been recorded in Mallotus japonicus (Ventura, 1934), Maianthemum bifolium and M. canadense (Stenar, 1934; Swamy, 194969a), Crucianella latifolia, Rubia olivieri (Fagerlind, 1937), Tanacetum vulgare, Chrysanthemum parthenium (Fagerlind, 1941), Ulmus (Ekdahl, 1941; Walker, 1950), and a few other plants. A few of these deserve special mention and are briefly discussed below.
Shattuck (1905) reported an 8-nucleate embryo sac of the Adoxa type in Ulmus americana, but he observed that frequently there seemed to be a further nuclear division. Several embryo sacs were found to contain as many as 12 or more nuclei, rather evenly dis-
9a In M. canadense, according to Swamy (19496), in about 13 per cent of the ovules the chalazal spindles of the last division fuse in pairs so that the mature embryo sac comes to possess 4 haploid nuclei at the micropylar end and 6 diploid nuclei at the chalazal end.
Pig. 73. Development of embryo sac in Drusa oppositifolia (A), Chrysanthemum parthenium (B-II), and Crucianella latifolia (I-M). A, Drusa, 16-nucleate embryo sac, showing four nuclei at micropylar end and twelve at chalazal. (After Hakans- son, 1928.) B,C, Chrysanthemum, young embryo sacs showing varying arrange- ments of the four megaspore nuclei. D-F, eight-nucleate stage; note degeneration of basal nucleus in E. G, last division in embryo sac, basal nucleus degenerating. //, mature embryo sac, showing 12 nuclei. (After Fagerlind, 1941.) I, Crucianella, megaspore nuclei. J, same, in division. K, fourth division in embryo sac; some of nuclei at chalazal end have failed to divide. L, embryo sac, showing 15 nuclei. M, mature embryo sac. (After Fagerlind, 1937.)
THE FEMALE GAMETOPHYTE
113
& !
■> ■'
ff
M^g
«3?^
K\
'(.
ty
ii
iG
1/
I
H
a
a
■!■;-;;>'
1
m
114 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
tributed and very similar to one another in appearance. Some other workers also obtained similar results, and noted that the mature embryo sacs occasionally showed more than three antip- odals. D'Amato (1940a), Ekdahl (1941), and Walker (1950), work- ing on several species of Ulmus, have clarified the position by show- ing that as a rule four divisions intervene between the megaspore mother cell stage and the organization of the embryo sac, and not three. The coenomegaspore shows a 1 + 3 arrangement of the megaspore nuclei, each of which undergoes two further divisions, resulting in the formation of 4 nuclei at the micropylar end and 12 at the chalazal end. Frequently, however, some of the chalazal nuclei fail to undergo the fourth division, resulting in a total of 14, 12, or only 10 nuclei, of which 4 are at the micropylar end and the rest at the chalazal. Several of the latter degenerate soon after their formation, so that there is a further decrease in the number of nuclei, and eventually only two to four antipodal cells may be differ- entiated. Also, in certain cases the 4 megaspore nuclei divide only once, so as to give rise to an 8-nucleate embryo sac of the Adoxa type.
The embryo sac of Chrysanthemum parthenium presents a range of variation which seems to indicate that there are several races of this plant which behave somewhat differently from one another, although possibly the differences are related to environmental con- ditions. According to Palm (1915), who gave the first detailed account of the embryo sac of this species, each of the 4 megaspore nuclei divides twice. The 16 nuclei arising in this way organize to form a three-celled egg apparatus, two polar nuclei, and eight antipodal cells of which the basal cell is four-nucleate.
Fagerlind (1941) studied two specimens of the same species. In specimen 1 the 4 megaspore nuclei were observed to take up the most variable positions, and frequently the 3 basal nuclei were seen to lie in close contact (Fig. 735). With the subsequent elon- gation of the sac the nuclei became separated from one another by vacuoles, the micropylar nucleus being larger than the rest (Fig. 73C). All the nuclei now divided simultaneously, resulting in 8 nuclei, of which the 2 basal were the smallest and soon began to degenerate (Fig. 73D-F). When the next division (Fig. 73G) was over, there were 14 nuclei in the sac, of which 3 organized into an egg apparatus, 2 functioned as polar nuclei, and the rest formed
THE FEMALE GAMETOPHYTE 115
the antipodal cells (Fig. 73H). The basal antipodal cell contained a variable number of nuclei, which subsequently fused to form 1 nucleus. Embryo sacs with fewer than 14 nuclei were also seen, but this was due to a degeneration and disappearance of some of the nuclei at the chalazal end.
In specimen 2, collected from a different locality in Sweden, the megaspore mother cells as well as the developing embryo sacs and their nuclei were found to lie of a larger size than in the first plant. The chalazal megaspore nucleus degenerated soon after its formation. The remaining 3 nuclei divided to form 6 and then 12 nuclei. In the mature embryo sac the basal antipodal cell was observed to have more than one nucleus, while the remaining antip- odal cells were uninucleate.
Material of the same species collected from the Brooklyn Bo- tanical Gardens, New York (Maheshwari and Haque, 1949), showed the usual 4- and 8-nucleate stages, after which all the nuclei were found to divide again, resulting in 16 nuclei, 4 at the micropylar end and 12 at the chalazal. These organize to form a three-celled egg- apparatus, two polar nuclei, and eleven uninucleate antipodal cells.
The embryo sac of Tanacetum vulgar e (Fagerlind, 1941) is funda- mentally similar to that of Chrysanthemum. The megaspore mother cell (Fig. 74 A) undergoes the usual reduction divisions to produce 2 (Fig. 745) and then 4 nuclei (Fig. 74C) which become arranged in a linear fashion (Fig. 74D). Vacuoles soon appear between the nuclei, which now increase in size and prepare for the next division (Fig. 74E), resulting in the formation of 8 nuclei (Fig. 74//"). In many cases, however, the basal nucleus does not take part in the division and soon begins to degenerate (Fig. 74G), and sometimes the subbasal nucleus also remains undivided (Fig. 74F). At this stage the embryo sac may, therefore, contain 8, 7, or only 6 nuclei. If all of them take part in the next division, the mature embryo sacs may be 16-, 14-, or 12-nucleate (Fig. 74/). But frequently there is a further degeneration of one or two of the chalazal nuclei so that embryo sacs with fewer than 12 nuclei are not uncommon (Fig. 74/).
Crucianella laiifolia (Fagerlind, 1937), a member of the Rubiaceae, also belongs to the Drusa type. After the reduction divisions are over, the coenomegaspore shows a pronounced elongation, rupturing the nucellar epidermis at its micropylar end (Fig. 737). The next
116 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
division proceeds normally (Fig. 73/), but of the 8 nuclei now formed the basal nucleus remains undivided (Fig. 73 Jv) so that the mature embryo sac shows only 15 nuclei (Fig. 73L) which be- come organized to form a three-celled egg apparatus, two polar nuclei, and 10 antipodals (Fig. 73M).
Fig. 74. Development of embryo sac in Tanacetum vulgare. A, mega spore mother cell. B,C, end of first and second meiotic division, respectively. D, megaspore nuclei. E, same, older stage, showing vacuolation. F-H, first postmeiotic divi- sion; two megaspore nuclei dividing in F, three in G, and all four in H. I, J, mature embryo sacs with varying number of nuclei. (After Fagerlind, 1941.)
In Maianthemum bifolium (Stenar, 1934) both the reduction divi- sions are accompanied by the formation of cell plates (Fig. 75 A-C). They soon become absorbed, however, resulting in a common tetra- nucleate cell (Fig. 75D). The four megaspore nuclei take up a 1+3 arrangement, so that the next stage shows 2 nuclei at the micropylar pole and 6 at the chalazal pole (Fig. 75E-F). There is one more division, resulting in 16 nuclei (Fig. 75G). These organize into an egg apparatus, two polar nuclei, and 11 antipodal cells.
THE FEMALE GAMETOPHYTE
117
Most of the antipodal cells soon degenerate, and only a few may be seen in the mature embryo sac (Fig. 75H).
Fritillaria Type. Following the work of Treub and Mellink
E F G H
Fig. 75. Development of embryo sac in Maianthemum bifolium. A, l.s. nucellus, showing megaspore mother cell. B, clyad stage. C, tetrad. D, four-nucleate embryo sac formed by dissolution of walls separating megaspores. E, eight- nucleate stage. F, same, nuclei in prophase of next division. G, embryo sac with 16 nuclei. H, mature embryo sac. (After Stenar, 1934)
(1880) on Lilium bulbiferum, several other investigators, notably Strasburger, Mottier, Guignard, Coulter, and Sargant, studied a number of species of this genus and repeatedly confirmed that the
118 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
4 megaspore nuclei undergo only one division to give rise to the 8 nuclei of the mature embryo sac. They no doubt observed certain peculiarities and curious appearances which could not be explained on this interpretation, but these were disregarded as abnormal or even "pathological" conditions.
Bambacioni (1928a, 6) showed that in Fritillaria and Lilium the formation of the 4 megaspore nuclei is not followed directly by the 8-nucleate stage but by a secondary I^-nucleate stage, in which the 2 chalazal nuclei are much larger than the micropylar. This comes about in a very peculiar manner. At first there is a 1+3 arrange- ment of the megaspore nuclei (Fig. 76A-D) so that the 3 chalazal nuclei come to lie very close to each other. During the next stage the micropylar nucleus divides normally, but the three chalazal spindles fuse to form a single common spindle (Fig. 7QE-F), so that at the close of the division there are two haploid nuclei at the micropylar end and two triploid nuclei at the chalazal (Fig. 76 G-H). One more division occurs, resulting in 8 nuclei, of which the 4 chalazal nuclei are triploid and the 4 micropylar are haploid (Fig. 767). The mature embryo sac thus consists of three haploid cells (the egg and two synergids), three triploid cells (the antip- odals), and a tetraploid secondary nucleus formed by the fusion of the two polar nuclei, one haploid and the other triploid (Fig. 76/). Of the antipodals, the two lowest frequently show a flattened and degenerated appearance — a condition originating from the fact that the basal nucleus of the secondary 4-nucleate stage often divides in a more or less abortive fashion.
Cooper (1935a) extended the observations of Bambacioni to sev- eral other species of Lilium, and since then the Fritillaria type has been demonstrated in a general way for the entire tribe Lilioideae and several other genera belonging to diverse families; Piper, Heekeria, Myricaria, Tamarix, Cornus (some spp.), Armeria, Statiee (most spp.), Rudbeckia (most spp.), GaiUardia, Cardiocrinum, Gagea, Erythronium (most spp.), Tulipa (some spp.), and Clintonia (see Maheshwari, 19466, for detailed information).
It may be noted that the fusion of the 3 chalazal megaspore nuclei may take place when they are either in the prophase stage or in early metaphase. In the former case the secondary 4-nu- cleate stage is preceded by a secondary 2-nucleate one, and the
THE FEMALE GAMETOPHYTE
119
sequence then is as follows: megaspore mother cell, primary 2-nu- cleate stage, primary 4-nucleate, secondary 2-nucleate, secondary 4-nucleate, and last of all the S-nucleate stage.
Fig. 76. Development of embryo sac in Fritillaria persica. A, l.s. nucellus, show- ing megaspore mother cell in prophase of Meiosis I. B, two-nucleate stage. C, primary four-nucleate stage. D, megaspore nuclei, showing 1 +3 arrangement. E, megaspore nuclei dividing. F, same, showing fusion of three chalazal spin- dles. G, telophase of same division. H, secondary four-nucleate stage in which the two micropylar nuclei are haploid and chalazal nuclei are triploid. 7, four nuclei dividing to form eight. J, eight-nucleate embryo sac. (After Bambacioni, 1928a.)
Normally all the 4 megaspore nuclei are of the same size, but in some plants the micropylar nucleus is the largest and the other 3 nuclei are considerably smaller. When this happens, the nuclei of
120 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
the secondary 2-nucleate and secondary 4-nucleate stages show no appreciable difference in size, and rarely the chalazal nuclei are smaller than the micropylar in spite of the triploid nature of the former.
Finally, the basal nucleus of the secondary 4-nucleate stage some- times fails to divide, resulting in a 7-nucleate gametophyte with two antipodal cells instead of three, as in some species of Gagea (Romanov, 1936) ; or, both the basal as well as the subbasal nucleus remain undivided and the embryo sac is 6-nucleate, as in Statice (Fagerlind, 19396). In Tulipa maximovitii (Romanov, 1939) the 3 chalazal megaspore nuclei undergo an abnormal division in which all the telophase chromosome groups become included in a common membrane, so that the mature embryo sac is 5-nucleate. In one genus, Clintonia (R. W. Smith, 1911; F. H. Smith, 1943; Walker, 1944), the chalazal megaspore nuclei degenerate as soon as they are formed, without undergoing any division at all.
Plumbagella Type. In this type also, which has so far been re- ported only in Plumbagella micrantha (Fagerlind, 19386; Boyes, 1939), the 4 megaspore nuclei take up a 1+3 arrangement (Fig. 77 A-C), and a large vacuole separates the 3 chalazal nuclei from the micropylar nucleus (Fig. 77 D). The former gradually approach one another and eventually fuse to give rise to a single triploid nucleus (Fig. 77 E). This results in a secondary 2-nucleate stage, followed by a secondary 4-nucleate one, in which the 2 micropylar nuclei are haploid and the chalazal are triploid (Fig. 77 F-G). There are no further divisions. The nucleus nearest the micro- pylar end organizes into the egg; the triploid nucleus nearest the chalazal end forms the single antipodal cell; and the remaining^ nuclei, one haploid and the other triploid, fuse to form a tetraploid secondary nucleus (Fig. 77/7-7).
This mode of development shows an evident relationship with
Fig. 77. Development of embryo sac in Plumbagella micrantha. A, megaspore mother cell. B, second meiotic division in megaspore mother cell. C, megaspore nuclei showing 1+3 arrangement; the three chalazal nuclei are of a smaller size. D, chalazal nuclei in process of fusion. E, fusion of the three chalazal nuclei is completed, resulting in formation of secondary two-nucleate stage. F,G, forma- tion of secondary four-nucleate sLage. H, wall formation in embryo sac. /, mature embryo sac showing egg, secondary nucleus, and single antipodal cell. (After Fagerlind, 1938b.)
THE FEMALE GAMETOPHYTE
121
122 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
the Fritillaria type, the only difference being that in Plumbagella the development is arrested at the secondary 4-nucleate stage and the fourth division is omitted.
Adoxa Type. The Adoxa type, formerly known as "Lilium type," is characterized by all 4 megaspore nuclei undergoing just one more division to form an 8-nucleate embryo sac having a normal egg apparatus, three antipodal cells, and two polar nuclei (Fig. 78). It was described for the first time by Jonsson (1879-1880) in Adoxa moschatellina and later by Lagerberg (1909) and Fagerlind (1938a).
Until only a few years ago there was a long list of plants under the Adoxa type. With the publication of Bambacioni's work and the consequent reinvestigation of Lilium, Fritillaria, and several other genera, its ranks have steadily diminished and there are now only five genera in which its occurrence is a more or less regular feature: Adoxa, Sambucus, and some species of Erythronium10, Tulipa, and Ulmus.
An interesting variation has been reported in some species of Tulipa. In T. sylvestris (Bambacioni-Mezzetti, 1931) vacuolation frequently commences even at the megaspore mother cell stage, and all the 4 megaspore nuclei gather at the micropylar end of the cell, where they divide to give rise to a group of six cells (one of which is to be interpreted as the egg) and 2 free nuclei. T. tet- raphylla (Romanov, 1938) is essentially similar. After the meiotic divisions are over, 3 nuclei go to the micropylar pole and one to the chalazal (Fig. 79A-D). All of them divide again (Fig. 79 E), so that there are 6 daughter nuclei in the upper part of the sac and 2 in the lower. Cell plates are laid down at the conclusion of the division, resulting in the formation of five cells at the micropylar end (one of these is to be regarded as the egg) and one cell at the chalazal, leaving 2 free nuclei (the polars) in the center (Fig. 79F).
Since this peculiar mode of development occurs only in the Eriostemones section of the genus Tulipa, it is known as the "Erio- stemones form" of the Adoxa type. Other species of the genus come under the Fritillaria or the Drusa type (see Maheshwari, 1948).
10 Haque's (1950) observations on E. americanum and Walker's (1950) on U. fulva, U. racemosa, and U. glabra show that the development sometimes follows the Adoxa type and sometimes the Fritillaria type.
THE FEMALE GAMETOPHYTE
123
Fig. 78. Development of embryo sac in Adoxa moschatellina. A, megaspore mother cell. B, two-nucleate stage. C, two nuclei dividing. D-G, four-nucleate embryo sacs. H, division of four nuclei. /, same, telophase. J,K, mature embryo sacs. (After Fagerlind, 1938a.)
124
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
One species of Leontodon, L. hispidus, also deserves mention in this connection (Bergman, 1935). Ordinarily a row of four mega- spores is formed, and the embryo sac is of the Polygonum type. But in more than 50 per cent of the ovules of one plant the sepa- rating walls between the megaspore nuclei frequently dissolved and disappeared, and all the 4 nuclei divided only once to give rise to the 8-nucleate stage (Fig. 80). Since here only three divisions intervened between the megaspore mother cell stage and the dif- ferentiation of the egg, this mode of development comes under the Adoxa type.
Fig. 79. Development of embryo sac in Tulipa tetraphylla. A-C, formation of megaspore nuclei. D, 3+1 arrangement of megaspore nuclei. E, all four nuclei F, mature embryo sac. (After Romanov, 1988.)
dividing
Plumbago Type. The embryo sac of Plumbago capensis, described by Haupt (1934), may be presented as a representative of the Plumbago type. The 2- and 4-nucleate stages (Fig. 81 A-B) are normal, and the 4 megaspore nuclei, which are arranged in a crosswise fashion, undergo a further division (Fig. 81C) resulting in 8 free nuclei arranged in four pairs (Fig. 81 D), One nucleus of the micropylar pair is now cut off to form the lenticular egg cell (Fig. &IE). Of the remaining 7 nuclei, 4 (presumably one member of each of the original four pairs) undergo a slight increase in size and gradually approach one another, functioning as polar nuclei (Fig. 81F). The remaining 3 nuclei degenerate at their original places, but occasionally 1, 2, or all 3 of them are cut off at the periphery to form cells which may persist and assume an egg-like appearance; synergids are entirely absent (Fig. 81G-H).
THE FEMALE GAMETOPHYTE
125
The Plumbago type occurs not only in other species of the genus Plumbago (Dahlgren, 1937; Fagerlind, 19386) but also in two other genera of the Plumbaginaceae, viz., Ceratostigma (D'Amato, 19406) and Vogelia (Mathur and Khan, 1941). It is so far unknown outside this family.
F
Fig. 80. Development of embryo sac in Leontodon hispidus. A, tetrad of mega- spores. B-E, dissolution of separating walls between megaspore nuclei. F, mature eight-nucleate embryo sac. (After Bergman, 1935.)
ABERRANT AND UNCLASSIFIED TYPES
In addition to the above fairly distinct and well-established types of embryo sac development, there are a few which appear to be more or less isolated. The more important of them are mentioned below.
Limnanthes douglasii. The embryo sac of this plant has been investigated by three different workers but without any complete agreement regarding the mode of development. Stenar (1925a) re- ported an Adoxa type of embryo sac and called attention to the
126 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
reduced size of the nuclei at the chalazal end. Eysel (1937) con- firmed this report but noted an occasional reduction in the number of nuclei at the chalazal end of the embryo sac, owing to a failure
H G F
Fig. 81. Development of embryo sac in Plumbago capensis. A, two-nucleate stage derived from first division of megaspore mother cell. B, four-nucleate stage. C, all four nuclei dividing; one of the mitotic figures is oriented at right angles to plane of sectioning. D, eight-nucleate stage, showing cutting off of egg cell. E, differentiation of the four polar nuclei. F, fusion of polar nuclei. G,H, later stages, showing egg at micropylar end and secondary nucleus in center. The two lateral cells in G are derived from nuclei which ordinarily disappear in earlier stages. (After Haupt, 1934.)
of the basal nucleus of the 4-nucleate stage to undergo the last division. In other cases he observed a disappearance of the wall separating the megaspore mother cell from the nucellar cell situated directly below it and the consequent incorporation of the latter into the embryo sac. One embryo sac showed 9 nuclei, of which
THE FEMALE GAMETOPHYTE
127
7 had organized into cells (four looking like synergids, two looking like eggs, and one of an undecided nature) and 2 resembled polar nuclei; antipodals were absent.
Fagerlind's (1939c) observations differ from those of both Stenar and Eysel. The megaspore mother cell has a highly vacuolated cytoplasm (Fig. 82 A). u As a result of the first division, 2 nuclei are formed of which the lower promptly degenerates and is reduced
Fig. 82. Development of embryo sac in Limnanthes douglasii. A, megaspore mother cell. B,C, two-nucleate embryo sacs ; note degeneration of primary chalazal nucleus. D, three-nucleate stage originating by division of primary micropylar nucleus. E, division of micropylar nucleus. F, embryo sac piercing the nucellar epidermis; note two nuclei at micropylar end, one nucleus in middle, and degene- rated nucleus at chalazal end. G, mature embryo sac showing egg apparatus, upper polar nucleus, lower polar nucleus (?), and degenerating antipodal cell. (After Fagerlind, 1939c.)
to a densely staining homogeneous blob which lies at the bottom of the embryo sac and takes no further part in the development (Fig. 82B-C). The upper nucleus divides to form 2 daughter nu- clei, of which the lower is much smaller and usually incapable of further division (Fig. 82D). Following meiosis, we thus have a 3-nucleate stage showing a micropylar nucleus, a middle nucleus, and a chalazal nucleus. Of these the micropylar nucleus divides
11 In the majority of angiosperms vacuolation takes place only after the meiotic divisions are over.
128 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
twice, to give rise to a group of 4 nuclei which form the egg ap- paratus and the upper polar nucleus (Fig. S2E-G). Of the re- maining 2 nuclei, one may be considered as an antipodal and the other as the lower polar nucleus.
Several variations in the development and organization of the embryo sac were found, however. Most of these seemed to have their origin in the behavior of the middle nucleus. In some cases it was found to take part in the third or fourth division, resulting in a 7-nucleate embryo sac with two antipodal nuclei instead of one. Less frequently it divided synchronously with the micropylar nu- cleus, but only one of its daughter nuclei divided again, resulting in an 8-nucleate embryo sac.
It is probable that at least some of the variations reported by Stenar, Eysel, and Fagerlind are due to environmental influences, and a more detailed study is necessary to decide the point.110
Balsamita vulgaris. A recent investigation of the embryo sac of this plant (Fagerlind, 1939c) has revealed several interesting features. As in other Compositae, the ovules are tenuinucellate. The archesporium is usually two-celled (Fig. S3 A), but sometimes three cells may be present and occasionally there is only one. After the first meiotic division 2 nuclei are formed of which the upper soon becomes larger than the lower (Fig. 831?). Both divide again with- out wall formation and the resulting 4 nuclei take up a 1+3 ar- rangement (Fig. 83C-D). Only the micropylar nucleus functions, while the other 3 nuclei soon begin to degenerate. Vacuolation takes place at this stage and is followed by the appearance of a lateral vesicular outgrowth, which assumes a tubular form and gradually makes its way upward into the micropyle (Fig. S3E-F). The functioning megaspore nucleus, which has by this time moved
lla Mason (1949), who has made a recent study of Limnanthes, regards the embryo sac as bisporic.
Fig. 83. Development of embryo sac in Balsamita vulgaris. A, l.s. nucellus showing two-celled archesporium. B, mother cell on right has two nuclei (end of Meiosis I); that on left has four nuclei (end of Meiosis II). C,D, megaspore nuclei take up 1+3 position; micropylar nucleus has enlarged; smaller chalazal nuclei are on way to degeneration. E, formation of vesicular outgrowth from chalazal end of the cell. F, functional megaspore nucleus has entered vesicle. G-H, two- and four-nucleate stages. /, eight-nucleate embryo sac; note three nonfunctioning megaspore nuclei at base. J, mature embryo sac, showing egg apparatus, second- ary nucleus, and multinucleate antipodal cells. (After Fagerlind, 1989c.)
THE FEMALE GAMETOPHYTE
129
Fig. 83.
130 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
into the apex of the tube, undergoes two divisions to form 4 nuclei, which lie in two pairs, one at each end of a large vacuole (Fig. SSG-H). The next division gives rise to 8 nuclei, of which the upper 4 form the egg apparatus and the upper polar nucleus, and the lower 4 give rise to the three antipodal cells and the lower polar nucleus (Fig. 83 1-J). The lowest antipodal cell connects the vesicular outgrowth with the body of the old megaspore mother cell in which the three degenerated megaspore nuclei are sometimes still distinguishable. The nuclei of the antipodal cells frequently undergo a few divisions but the daughter nuclei fuse once again to form a single lobed nucleus.
Chrysanthemum cinerariaefolium. Martinoli (1939) has dis- covered a peculiar mode of development in this plant. The em- bryo sac is tetrasporic and the megaspore nuclei take up a 1+2 + 1 arrangement so that there is 1 nucleus at each pole and 2 nuclei lie in the middle (Fig. 84A-C). The two central nuclei become separated from the terminal nuclei by vacuoles and may either fuse to form a single diploid nucleus (Fig. 84#) or may merely re- main close to one another without undergoing any fusion (Fig. 84D). The subsequent development differs, depending on which of the two conditions is present.
In the first case the next division gives rise to 6 nuclei (a haploid pair at either end and a diploid pair in the center) (Fig. 847) which divide again to form three groups of 4 nuclei each (Fig. 84/). The micropylar quartet now produces the egg apparatus and upper polar nucleus, all haploid. The chalazal quartet gives rise to four antipodal cells, also haploid. The central quartet is composed of diploid nuclei; one of these functions as the lower polar nucleus and the remaining 3 organize as additional antipodal cells (Fig. 84iv). Sometimes less than 12 nuclei are formed (10 or 7), either because of a failure of some divisions at the chalazal end or because the central diploid nucleus of the 3 -nucleate stage undergoes only one division instead of two.
In the second of the two previously mentioned alternatives, i.e., when the two central megaspore nuclei do not fuse but only lie in contact with each other, neither undergoes any further divisions and both function directly as the polar nuclei. Meanwhile the micropylar megaspore nucleus divides twice, to give rise to the micropylar quartet, but there is no regularity in the behavior of
THE FEMALE GAMETOPHYTE
131
the chalazal nucleus. The total number of nuclei in the mature embryo sac may therefore be 10 or 9 or even as few as 6, depending upon two divisions, or a single division, or a complete failure of division, of this nucleus (Fig. 84E-G).
Fig. 84. Two modes of development of embryo sac of Chrysanthemum cinerariae- folium. A-C, formation of the four megaspore nuclei. D-G, first type of develop- ment, in which the two central megaspore nuclei remain undivided and function directly as polar nuclei. H-K, second type of development, in which two central megaspore nuclei fuse to form diploid nucleus which undergoes two divisions to give rise to four nuclei; of these, one functions as polar nucleus and three form anti- podal cells. For details, see text. (Adapted from Martinoli, 1939.)
ORGANIZATION OF MATURE EMBRYO SAC
Although the origin of the mature embryo sac may differ, its eventual organization shows a surprisingly uniform pattern in the majority of angiosperms. The Polygonum, Allium, Fritillaria, and Adoxa types of embryo sacs all have a similar appearance at the time of fertilization (three -celled egg apparatus, three antipodals, and two polar nuclei). Even in the remaining types an egg ap- paratus, at least, is almost always present and it is only in a few genera like Peperomia, Plumbago, Plumbagella, and Acalypha indica that we see a radical departure from the basic plan. Ignoring for
132 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
the present the origin of the embryo sac (whether mono-, bi-, or tetrasporic), we shall now confine our attention to the organiza- tion of the mature stage only.
The Egg Apparatus. Typically
the egg apparatus is composed of an egg and two synergids. As a rule each of the synergids is notched by an indentation result- ing in the formation of a promi- nent hook (Fig. 85). The upper part of the cell is occupied by the so-called "filiform apparatus" which shows a number of stria- tions converging towards the apex. The nucleus lies in or just below the region of the hook and the lower part of the cell contains a large vacuole (Dahlgren, 1928a, 1938). In the egg, on the other hand, the nucleus and most of the cytoplasm lie in the lower part of the cell and the vacuole in the upper. Hooks and indentations are usually absent, having been described only in Plumbagella, Ditepalanthus, Hclosis, and a few members of the Ulmaceae and Urticaceae (Fagerlind, 1943).
Usually the synergids are eph- emeral structures which degener- ate and disappear soon after fer- tilization or even before it. In some cases, however, one or both of them may persist for a time and
Fig. 85. Mature embryo sac of Oeno- thera nutans, showing synergids with fili- form apparatus and indentations. Note that nucleus of synergids lies towards upper end of cell, and vacuole towards lower end. (After Ishikawa, 1918.)
show signs of considerable activ- ity. In Allium unifolium and A. rotundum (Weber, 1929) this be- havior is particularly pronounced, and one of the synergids begins to degenerate only after the development of the embryo is well under way. Nothoscordum (Stenar, 1932), Limnanthes (Fig. 86B) (Fager-
THE FEMALE GAMETOPHYTE
133
lind in 1939c), and Albuca (Eunus, 1950) are essentially similar; and in some Cucurbitaceae (Fig. 86^4) both the synergids become large and prominent and seem to play an important role in the nutrition of the embryo sac.
Fig. 86. Modifications of synergids. A, Luffa acutangula, embryo sac, showing extremely long synergids reaching down to level below middle of sac. {After Kirkwood, 1905.) B, Limnanthes douglasii, embryo sac showing three-celled embryo and persisting synergid. {After Fagerlind, 1939c.) C, Ursinia anthe- moides, beak-shaped synergids protruding through micropyle. {After Dahlgren, 1924.)
In none of the plants cited above do the synergids extend beyond the limits of the embryo sac wall. This condition has so far been noted to a pronounced extent only in the Compositae. Dahlgren (1924) found that in Ursinea (Fig. 86C) and Calendula the synergids elongate so much that their tips project to a considerable distance into the micropyle and outside it, sometimes reaching as far as the funiculus.
Certain other reports of the occurrence of synergid haustoria,
134 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
as in Lathraea, Lobelia, and Angelonia, have, however, to be in- terpreted differently, for there is now no doubt that the cells in question are really endosperm derivatives. The confusion was caused by the fact that the micropylar cells of the endosperm some- times show an appearance identical with that of the synergids — a vacuole lying in the lower part of the cell and the nucleus and cyto- plasm in the tapering upper part (see Rosen, 1947). In Myriophyl- lum (Stolt, 1928; Soueges, 1940) and Hypecoum (Soueges, 1943) even suspensor cells are known to show a surprising resemblance to synergids.
Antipodal Cells. Although usually short-lived, the antipodals frequently show a considerable increase in size or number. In some members of the Gentianaceae (Stolt, 1921) the three antipodal cells divide to form about 10 to 12 cells (Fig. 87 D), and in the Gramineae a still larger number of cells is produced (Fig. 87C). In Sasa paniculata (Yamaura, 1933), a member of the Bam- busae, an many as 300 antipodal cells have been reported.
In several genera of the Rubiaceae, like Putoria (Fagerlind, 1936a) and Galium (Fagerlind, 1937), the basal antipodal cell is often greatly elongated and acts as an aggressive haustorium (Fig. 87 B, E). In Phyllis (Fagerlind, 19366) all three of the antipodal cells are swollen; the basal becomes 8-nucleate and each of the upper two becomes 4-nucleate (Fig. 87 G,H).
An increase in the number of antipodal cells and the number of nuclei per antipodal cell is well known in the Compositae (Fig. 87 A, F). In Grindelia squarrosa, according to Howe (1926), only two antipodal cells are formed, the one nearer the micropyle being binucleate. One or both of these cells undergo further develop- ment, growing laterally into the integument for a considerable dis- tance. In Artemisia (Diettert, 1938) the number of antipodal cells varies from three to six and each cell may have 2 or more nuclei. The basal antipodal cell frequently elongates and penetrates through the chalazal tissue, finally entering the ovarian chamber. Rudbeckia bicolor (Maheshwari and Srinivasan, 1944), whose em- bryo sac follows the Fritillaria type of development, has triploid antipodal cells which attain a much larger size than the cells of the egg apparatus (Fig. 88). The central antipodal cell, in particular, persists for a long time, being recognizable even during embryonal development.
THE FEMALE GAMETOPHYTE
135
Fig. 87. Embryo sacs showing abnormal behavior of antipodal cells. A, Ligu- laria sibirica; embryo sac showing increase in number of antipodal cells, some of which are binucleate. (After Afzelius, 1924-) B, Putoria calabrica, three embryo sacs of which two are well organized; note extreme elongation of basal antipodal cell. (After Fagerlind, 1936a.) C, Zea mays, embryo sac showing mass of antipodal cells at lower end. (After Randolph, 1£83.) D, Gentiana campestris, embryo sac showing increase in number of antipodal cells. (After Stolt, 1921.) E, Galium moliugo, embryo sac showing elongation of basal antipodal cell. (After Fagerlind, 1987.) F, Aster novae-anglieae, several multinucleate antipodal cells, of which basal has undergone considerable enlargement. (After Chamberlain, 1895.) G, Phyllis nobla, l.s. ovule. H, embryo sac of same enlarged to show young embryo, endosperm, and three haustorial antipodal cells. (After Fagerlind, 1936b.)
136 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
The antipodal cells of some members of the Ranunculaceae be- come greatly enlarged and assume a glandular appearance (Fig. 89). Graft (1941) has shown that in Caltha palustris they attain a
high degree of polyploidy. At first each antipo- dal cell becomes binucleate. The two nuclei now divide again, but the spindles fuse dur- ing this process so that there are again only two nuclei which, however, possess the diploid number of chromosomes. This process may be repeated, leading to the formation of tetraploid and even octoploid nuclei. It gives an indica- tion of the high metabolic activity in these cells and offers a close analogy with the behavior of the anther tapetum.
Polar Nuclei. The central portion of the em- bryo sac containing the polar nuclei eventually gives rise to the endosperm and has therefore been called the Endospermanlage or "endo- sperm mother cell." Usually the two nuclei are so similar to each other that once they have come together it is difficult to distinguish the micro- pylar from the chalazal. When there is a dif- ference in size between the two, it is usually the micropylar which is the larger. In embryo sacs of the Fritillaria type, however, the chala- zal polar nucleus is the larger (see page 118).
The fusion of the polar nuclei may occur either before, or during, or sometimes after, the entry of the pollen tube inside the embryo sac. The secondary nucleus formed after fusion usually lies just below the egg and is separated from the antipodal cells by a large vacuole. In those plants in which it lies near the center, it is con- nected with the egg apparatus by a conspicuous cytoplasmic strand. A chalazal position is less frequent except in those plants which are characterized by a Helo- bial type of endosperm (see page 245).
Embryo Sacs with Disturbed Polarity. Rarely, embryo sacs may be found in which the usual polarity and organization are absent.
Fig. 88. Embryo sac of Rudbeckia bicolor, showing three large antipodal cells which are arranged like cells of egg apparatus. (After Maheshwari and Srinivasan, 1944-)
THE FEMALE GAMETOPHYTE
137
Fig. 89. Embryo sac of Aconitum napellus showing three large antipodal cells. Note young embryo at micropylar end and endosperm nuclei in various stages of division. (After Osterwalder, 1898.)
138 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
Sometimes one or all of the antipodal nuclei move up and function as supernumerary polar nuclei; or the secondary nucleus fragments to form a group of micronuclei of varying sizes. Supernumerary egg cells and synergids have also been noted. Very rarely, the embryo sac shows a reversed polarity, with the egg apparatus differentiating at the chalazal end and the antipodals at the micro- pylar. As examples may be cited Atamosco texana (Pace, 1913), Fuchsia marinka (Tackholm, 1915), Lindelofia longiflora (Svens- son, 1925), Saccharum officinarum (Dutt and Subba Rao, 1933; Narayanaswami, 1940), Woodfordia floribunda (Joshi and Venka- teswarlu, 1935), Eriodendron anfractuosum (Thirumalachar and Khan, 1941), Heptapleurum venulosum (Gopinath, 1943), and Crinum asiaticum (Swamy, 1946). In certain other plants, a nor- mal egg apparatus is differentiated at the micropylar end, but two of the antipodal cells also look like synergids and the third resembles an egg, (Fig. 94A) so that the embryo sac apparently shows two egg apparatuses, one at each end. Poa alpina (Hakansson, 1943) sometimes shows the reverse condition, i.e., the occurrence of two groups of antipodal cells, one at the micropylar end and the other at the chalazal. Embryo sacs of the latter type are functionless, however, and do not produce embryos.12
The embryo sacs of the Viscoideae (Fig. 65), some members of the Balanophoraceae (Figs. 60, 61), and a few saprophytic genera of the Gentianaceae (Fig. 90) also appear to be inverted. Oehler (1927) has given the correct explanation when he says that the ovules of Leiphaimos and Cotylanthera, although seemingly ortho- tropous, are in fact anatropous, and that the inversion in the polar- ity of the embryo sac is only apparent but not real.
Food Reserves in the Embryo Sac. It is usually taken for granted that the angiosperm embryo sac is devoid of any appre- ciable food reserves. While this is generally true, there are now several records of the occurrence of starch in embryo sacs, and in the families Aizoaceae, Cactaceae, Portulacaceae, Bruniaceae, Tilia- ceae, Crassulaceae, and Asclepiadaceae this is a common phenome- non. Dahlgren (1927, 1939) who has reviewed the subject in recent years, states that the reason why starch grains have not been re-
12 A fertilization of antipodal cells seems to have been recorded only in Nigella arvensis (Derschau, 1918), but it is quite likely that it also occurs sometimes in Ulmus (Shattuck, 1905; Ekdahl, 1941).
THE FEMALE GAMETOPHYTE
139
ported more frequently in embryo sacs is that they are not very distinct in the usual balsam mounts, and very few workers take the trouble of removing the coverslip and testing the sections with an iodine solution.
While reference must be made to Dahlgren's papers for fuller information on the subject, a few noteworthy cases of the occur- rence of starch grains in the embryo sac may be mentioned here.
E
Fig. 90. Development of ovule and embryo sac of Leiphaimos spectabilis. Oehler, 1927.)
(After
In Arachis (Reed, 1924), Tilia (Stenar, 19256), Pentstemon (Evans, 1919), and Acacia (Newman, 1934) the embryo sacs are so full of starch that it becomes difficult to study the nuclei inside them. In Styphelia (Brough, 1924) starch grains are so abundant in the vicinity of the egg that the latter is obscured by them. In Den- drophthora (York, 1913) their crowding is said to cause a degenera- tion and disappearance of the nuclei.
The stage at which the starch makes its appearance in the embryo sac varies in different plants. In Loranthus pentandrus, Treub (1883) saw starch grains even at the megaspore mother cell stage;
140 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
in Psychotria (Fagerlind, 1937) starch appears at the dyad cell stage; in Castalia (Cook, 1902), Acacia (Guignard, 1881), Sedum (D'Hubert, 1896), Pentas, Richardsonia, and Cephalantus (Fager- lind, 1937) at the functioning megaspore stage; in Portulaca oleracea (Cooper, 1940) at the binucleate stage; and in Corchorus trilocularis (Stenar, 19256), Cynanchum acutum (Francini, 1927), and Medicago saliva (Cooper, 19356) at the 4-nucleate stage. In the majority of
Fig. 91. Embryo sacs of Acacia baileyana, showing starch grains. A\,Ai, succes- sive sections of unfertilized embryo sac. B, postfertilization stage. (After New- man, 1934.)
plants, however, the starch appears when the embryo sac is mature and reaches a maximum shortly after fertilization, gradually de- creasing in postfertilization stages. Xyris indica (Weinzieher, 1914), Acacia baileyana (Newman, 1934) (Fig. 91), and Petunia (Cooper, 1946) (Fig. 112) are peculiar in having large quantities of starch even during endosperm formation.
A few cases are on record in which the starch occurs not merely in the cavity of the embryo sac but also in the cells of the egg ap-
THE FEMALE GAMETOPHYTE
141
paratus and rarely even in the antipodal cells. The following are some examples of the occurrence of starch grains in the egg: Astilbe grandis (Dahlgren, 1930), Aspidistra elatior (Golaszewska, 1934), Acacia baileyana (Newman, 1934), Medicago saliva (Cooper, 19356; Cooper, Brink, and Albrecht, 1937), Korthalsella opuntia (Rutis- hauser, 19376), Zea mays, Euchlaena mexicana (Cooper, 1938), Portulaca oleracea (Cooper, 1940), and Phryma leptostachya (Cooper, 1941). In Korthalsella (Rutishauser, 19376) starch is also found in antipodal cells.
In Sonneratia (Venkates- warlu, 1937; Mauritzon, 1939) certain oily bodies of an un- known nature persist from the megaspore mother cell stage to the formation of the mature embryo sac, and in Aspidistra (Fig. 92) (Golaszewska, 1934) large raphides have been seen in the mature stages. The significance of these structures in the economy of the embryo sac has not been elucidated up to this time.
Embryo Sac Haustoria. In the majority of angiosperms the entire surface of the em- bryo sac serves an absorptive function, demolishing the ad- jacent cells of the nucellus and even the inner layers of the integument. In some plants,
however, more active growth
., , „ ,, Fig. 92. Embryo sac of Aspidistra elatior,
is seen at the ends of the sac. , • + , • , i WA
showing starch grams and a large raphide.
In Phaseolus (Fig. 93 A) (Wein- (After Golaszewska, 1984).
142
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
stein, 1926) and Melilotus, (Cooper, 1933) the embryo sac ruptures the nucellar epidermis and grows beyond it, so that more than one- third of it lies in direct contact with the cells lining the micropy- lar canal. In Arechavaletaia (Ventura, 1937) and Kirengeshoma (Mauritzon, 1939) it protrudes out of the endostome and comes to lie in the exostome. In Philadelphus (Mauritzon, 1933), Thesium (Schulle, 1933), Galium (Fagerlind, 1937), Utricularia (Kausik,
Fig. 93. Some instances of embryo sacs protruding into and beyond micropyle. A, Phaseolus vulgaris, upward elongation of embryo sac, resulting in rupture of nucellar tissue; nucellar epidermis is intact, however, at apex of embryo sac. (After Weinstein, 1926.) B, Torenia hirsuta, embryo sac protruding out of micropyle. (After Krishna Iyengar, 1941.) C, Philadelphus coronarius, embryo sac protruding out of micropyle. (After Mauritzon, 1933.) D, Galium lucidum, one embryo sac completely outside micropyle; another in position, but in process of degeneration. (After Fagerlind, 1937.)
1938), and certain members of the Scrophulariaceae like Vandellia and Torenia (Krishna Iyengar, 1940, 1941) the nucellus breaks down completely at a rather early stage and the naked embryo sac protrudes out of the ovule, establishing direct contact with the placenta and digesting its way into the tissue of the latter (Fig. 93B-D). Strangest of all are some genera of the Loranthaceae, like Scurrula and Dendrophthoe (Rauch, 1936; Singh, 1950), in which ovules and integuments are absent in the usual sense and the embryo sacs undergo a remarkable elongation toward both the top and the bottom. At the lower end they are soon stopped by a pad of col-
THE FEMALE GAMETOPHYTE
143
lenchymatous cells, but the upper part continues to grow, sometimes reaching a considerable distance into the style. Fertilization occurs here by the incoming pollen tubes and the embryos are thrust down again by the elongating suspensors. The observations of Schaeppi and Steindl (1942), who have recently studied several other genera of the family, show that in Macrosolen the upper end of the embryo
D E
Fig. 94. Formation of embryo sac caeca in Allium paniculatum (A-C), and Digera arvensis (D, E). {A-C, after Modilewski, 192S; D-E, after Joshi, 1930.)
sac reaches up to the base of the style, in Elythranthe it grows beyond its base, in Lepeostegeres it is at about one-fourth of the height of the style, in Amyema somewhere near its middle region, in Taxillus, slightly above the middle, and in HelixantheraUa just below the papillate layer of the stigma.
In other plants it is a downward growth which is more striking.
Ua See also Johri and Maheshwari (1950).
144 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
In several members of the Centrospermales (Oksijuk, 1927; Art- schwager and Starrett, 1933; Joshi, 1936; Cooper, 1949) the em- bryo sac pushes forward at the chalazal end and digests its way through the nucellus, while the antipodal cells remain in situ and are left behind in a lateral position (Fig. 94E). This "caecum," which is also known in Allium (Modilewski, 1928) (Fig. 94A-C), Elegia (Borwein, et al., 1949), Macrosolen (Maheshwari and Singh, 1950), and certain other plants (see Finn, 1936), seems to be a very effective haustorial organ.
In Veltheimia (Stiffler, 1925; Buchner, 1948), Paradisia (Stenar, 1928), and Eucomis (Buchner, 1948) the haustorium arises laterally rather than from the pole of the embryo sac. As a rule, however, such a condition is met with only in postfertilization stages and will therefore be considered in the chapter on the endosperm.
References
Afzelius, K. 1916. Zur Embryosackentwicklung der Orchideen. Svensk Bot.
Tidskr. 10: 183-227. . 1924. Embryologische und zytologische Studien in Senecio und ver-
wandten Gattungen. Acta Horti Bergiani 8(7): 123-219. Arnoldi, W. 1912. Zur Embryologie einiger Euphorbiaeeen. Trav. Mus. Bot.
Acad. St. Petersburg 9: 136-154. Artschwager, E., and Starrett, R. C. 1933. The time factor in fertilization and
embryo development in the sugar beet. Jour. Agr. Res. 47: 823-847. Bambacioni, V. 1928a. Ricerche sulla ecologia e sulla embriologia di Fritillaria
persica L. Ann. di Bot. 18: 7-37. . 19286. Contribute alia embriologia di Lilium candidum L. R. C. Accad.
Naz. Lincei 8: 612-618. Bambacioni-Mezzetti,V. 1931. Nouve ricerche suH'embriologia delle Gigliaceae.
Ann. di Bot. 19:365-382. Bergman, B. 1935. Zytologische Studien tiber die Fortpflanzung bei den Gat- tungen Leontodon und Picris. Svensk Bot. Tidskr. 29: 155-301. Borwein, B., Coetsee, M. L., and Krupko, S. 1949. Development of the embryo
sac of Restio dodii and Elegia racemosa. Jour. South AM. Bot. 15: 1-11. Boyes, J. W. 1939. Development of the embryo sac of Plumbagella micrantha.
Amer. Jour. Bot. 26: 539-547. Brough, P. 1924. Studies in the Epacridaceae. I. The life history of Styphelia
longifolia (R.Br.). Proc. Linn. Soc. N. S. Wales 49: 162-178. Brown, W. H., and Sharp, L. W. 1911. The embryo sac of Epipactis. Bot.
Gaz. 52: 439-452. Buchner, L. 1949. Vergleichende embryologische Studien an Scilloideae. Osterr.
bot. Zeitschr. 95: 428-450.
THE FEMALE GAMETOPHYTE 145
Campbell, D. H. 1899a. Die Entwicklung des Embryosackes von Peperomia
pellucida. Ber. deutsch. bot. Gesell. 17: 452-456. . 18996. A peculiar embryo sac in Peperomia pellucida. Ann. Bot. 13:
626.
1901. The embryo sac of Peperomia. Ann. Bot. 15: 103-117.
Carlson, M. C. 1945. Megasporogenesis and development of the embryo sac
of Cypripedium parviflorum. Bot. Gaz. 107: 107-114. Chamberlain, C. J. 1895. Contributions to the embryology of Aster novae-
angliae. Bot. Gaz. 20: 205-212. Chiarugi, A. 1933. Lo sviluppo del gametofito femmineo della Weddelina squam-
ulosa Tul. (Podostemonaceae) . R. C. Accad. Naz. Lincei 17: 1095-1100. Cook, M. T. 1902. Development of the embryo sac and embryo of Castalia
odorata and Numphaea advena. Bui. Torrey Bot. Club 29: 211-220. Cooper, D. C. 1933. Macrosporogenesis and embryology of Melilotus. Bot.
Gaz. 95: 143-155. . 1935a. Macrosporogenesis and development of the embryo sac of
Lilium henryi. Bot. Gaz. 97: 364-355. . 19356. Macrosporogenesis and embryology of Medicago. Jour. Agr.
Res. 51:471-477. . 1938. Macrosporogenesis and embryo sac development in Euchlaena
mexicana and Zea mays. Jour. Agr. Res. 55: 539-551. . 1940. Macrosporogenesis and embryo ogy of Portulaca oleracea. Amer.
Jour. Bot. 27: 326-330. . 1941. Macrosporogenesis and the development of the seed of Phryma
leptostachya. Amer. Jour. Bot. 28: 755-761.
. 1946. Double fertilization in Petunia. Amer. Jour. Bot. 33: 53-57.
. 1949. Flower and seed development in Oxybaphus nyctagineus. Amer.
Jour. Bot. 36: 348-355. -, Brink, R. A., and Albrecht, H. R. 1937. Embryo mortality in relation
to seed formation in alfalfa {Medicago sativa). Amer. Jour. Bot. 24: 203-213. Dahlgren, K. V. O. 1924. Studien iiber die Endospermbildung der Kompositen.
Svensk Bot. Tidskr. 18: 177-203. . 1927. tJber das Vorkommen von Starke in den Embryosacken der
Angiospermen. Ber deutsch. bot. Gesell. 45: 374-384. . 1928a. Hakenformige Leistenbildungen bei Synergiden. Ber. deutsch.
bot. Gesell. 46: 434-443. . 19286. Die Embryologie einiger Alismataceae. Svensk Bot. Tidskr.
22: 1-17. . 1930. Zur Embryologie der Saxifragagoideen. Svensk Bot. Tidskr.
24: 429-448. . 1934. Die Embryosackentwicklung von Echinodorus macrophyllus und
Sagittaria sagittifolia. Planta 21: 602-612. . 1937. Die Entwicklung des Embryosackes bei Plumbago zeylanica.
Bot. Notiser 1937, pp. 487-498.
146 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
Dahlgren, K.V.O. 193S. Hakenbildungen bei Synergiden. Zweite Mitteilung.
Svensk Bot. Tidskr. 52: 221-237. . 1939. Sur la presence d'amidon dans h sac embryonnaire chez bs
angiospermes. Bot. Notiser 1939, pp. 437-49S. D'Amato, F. 1940a. Embriologia di Ulmus campestris L. Nuovo Gior. Bot.
Ital. N.S. 47: 247-233. . 19406. Contributo all'embriologia delle Plumbaginaceae. Nuovo Gior.
Bot. Ital. N.S. 47:349-382. Derschau, M. v. 1918. Uber disperme Befruchtung der Antipoden von Nigella
arvensis. Ber. deutsch. bot. Gesell. 36: 260-262. D 'Hubert, E. 1896. Recherches sur le sac embryonnaire des plantes grasses.
Ann. des Sci. Nat., Bot. 2: 37-128. Diettert, R. A. 1938. The morphology of Artemisia tridentata Nutt. Lloydia
1 : 3-74. Dodds, K. S. 1945. Genetical and cytological studies of Musa. VI. The develop- ment of female cells of certain edible diploids. Jour. Genet. 46: 161-179. Dutt, N. L., and Subba Rao, K. S. 1933. Observations on the cytology of the
sugarcane. Indian Jour. Agr. Sci. 3 : 37-56. Ekdahl, I. 1941. Die Entwicklung von Embryosack und Embryo bei Ulmus
glabra Huds. Svensk Bot. Tidskr. 35: 143-156. Ernst, A. 1903. Ergebnisse neurer Untersuchungen uber den Embryosack der
Angiospermen. Verh. schvveiz. naturf. Ges. 91, Jahresv. I: 230-263. Eunus, A. M. 1950. Contributions to the embryology of the Liliaceae. X.
Albuca transvalensis Moss-Verdoorn. Jour. Indian Bot. Soc. 29: 68-78. Evans, A. T. 1919. Embryo sac and embryo of Pentstemon secundiflorus. Bot.
Gaz. 67: 426-437. Eysel, G. 1937. "Die Embryosackentwicklung von Limnanthes douglasii R.
Br." Diss. Marburg. Fagerlind, F. 1936a. Die Embryologie von Putoria. Svensk Bot. Tidskr. 30:
362-372. . 19366. Embryologische Beobachtungen liber die Gattung Phyllis. Bot.
Notiser 1936, pp. 577-584. . 1937. Embryologische, zytologische und bestaubungsexperimentelle
Studien in der Familie Rubiaceae nebst Bemerkungen uber einige Polyploidi-
tatsprobleme. Acta Horti Bergiani 11: 195-470. . 1938a. Wo kommen tetrasporische durch drei Teilungsschritte vollen-
entwickelte Embryosacke unter den Angiospermen vor? Bot. Notiser 1938,
pp. 461-498. . 19386. Embryosack von Plumbagella und Plumbago. Arkiv. for Bot.
29B: 1-8. . 1939a. Die Entwicklung des Embryosackes bei Peperomia pellucida.
Arkiv. for Bot. 29A: 1-15. -. 19396. Drei Beispiele des Fritillaria-Typus. Svensk Bot. Tidskr. 33:
188-204.
THE FEMALE GAMETOPHYTE 147
Fagerlind, F. 1939c. Kritische und revidierende Untersuchungen uber das
Vorkommen des Adoxa ("Lilium")-Typus. Acta Horti Bergiani 13: 1-49. . 1940. Stempelbau und Embryosackentwicklung bei einigen Pandana-
zeen. Ann. Jard. Bot. Buitenzorg 49 : 55-78. , 1941. Die Embryosackentwicklung bei Tanacetum vulgare und einigen
Chrysanthemum-Arten. Svensk Bot. Tidskr. 35: 157-176. . 1943. Vorkommen und Entstehung von Hakenleisten an Synergiden
und Eizellen. Svensk Bot. Tidskr. 37: 339-351. . 1944. Der tetrasporische Angiospermen-Embryosack und dessen Bedeu-
tung fur das Verhaltnis der Entwicklungsmechanik und Phylogenie des
Embryosacks. Arkiv. for Bot. 31A: 1-71. .. 1945a. Bildung und Entwicklung des Embryosacks bei sexuellen und
agamospermischen Balanophora-Arten. Svensk Bot. Tidskr. 39: 65-82.
19456. Bau der floralen Organe bei der Gattung Langsdorffia. Svensk
Bot. Tidskr. 39: 197-210. Finn, W. W. 1936. Zur Entwicklungsgeschichte der chalazogamen Ostrya carpi-
nifolia Scop. Jour. Inst. Bot. Acad. Sci. Ukraine 8: 15-25. Francini, D. E. 1927. L'embriologia del Cynanchum acutum L. Nuovo Gior.
Bot.' Ital. N.S. 34: 381-395. . 1931. Ricercheembriologicb.ee cariologichesulgenere Cypripedium. Nu- ovo Gior. Bot. Ital. N.S. 38: 155-212. Frisendahl, A. 1927. tJber die Entwicklung chasmogamer und kleistogamer
Bluten bei der Gattung Elatine. Acta Hort. Gothoburg. 3 : 99-142. Geerts, J. M. 1908. Beitrage zur Kenntnis der cytologischen Entwicklung von
Oenothera lamarckiana. Ber. deutsch. bot. Gesell. 26A: 608-614. Gerassimova, H. 1933. Fertilisation in Crepis capillaris. Cellule 42: 103-148. Golaszewska, Z. 1934. Die Entwicklung des Embryosackes bei Aspidistra ela-
tior. Acta Soc. Bot. Poloniae 11: 399-407. Goodspeed, T. H. 1947. Maturation of the gametes and fertilization in Nico-
tiana. Madrono 9: 110-120. Gopinath, D. M. 1943. Reversed polarity in the embryo sac of Heptapleurum
venulosum Seem. Current Sci. [India] 12 : 58. Grafl, I. 1941. Uber das Wachstum der Antipodenkerne von Caltha palustris.
Chromosoma 2: 1-11. Guignard, L. 1881. Recherches d'embryogenie vegetale compared. I. Legumi-
neuses. Ann. des Sci. Nat., Bot. 12: 5-166. Hakansson, A. 1923. Studien iiber die Entwicklungsgeschichte der Umbellife-
ren. Acta Univ. Lund N.F. Avd. II, 18(7) : 1-120. . 1943. Die Entwicklung des Embryosacks und die Bsfruchtung bei
Poa alpina. Hereditas 29: 25-61. Hammond, B. L. 1937. Development of Podostemon ceratophyUum. Bui. Tor-
rey Bot. Club 64: 17-36. Haque, A. 1950. The embryo sac of Erythronium amervanum. (In press.) Harling, G. 1946. Studien uber den Blutenbau und die Embryologie der Familie
Cyclanthaceae. Svensk Bot. Tidskr. 40: 257-272.
148 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
Haupt, A. W. 1934. Ovule and embryo sac of Plumbago capensis. Bot. Gaz.
95:649-659. Hoare, G. 1934. Gametogenesis and fertilization in Scilla nonscripta. Cellule
42 : 269-292. Hoeppener, E., and Renner, 0. 1929. Genetische und zytologische Oenotheren-
studien. II. Bot. Abh. Goebel 15: 1-66. Hofmeister, W. 1847. Untersuchungen des Vorgangs bei der Befruchtung der
Oenotheren. Bot. Ztg. 5 : 785-792.
. 1849. "Die Entstehung des Embryo der Phanerogamen." Leipzig.
Howe, T. D. 1926. Development of embryo sac in Grindelia squarrosa. Bot.
Gaz. 81: 280-296. Ishikawa, M. 1918. Studies on the embryo sac and fertilisation in Oenothera.
Ann. Bot. 32: 279-317. Jonsson, B. 1879-80. Om embryosackens utveckling hos angiospermerna.
Lunds Univ. Arskr. 16 : 1-86. Johansen, D. A. 1929. Studies on the morphology of the Onagraceae. I. The
megagametophyte of Hartmannia tetraptera. Bui. Torrey Bot. Club 56:
285-298. . 1931a. Studies on the morphology of the Onagraceae. II. Anogra
pallida. Amer. Jour. Bot. 18: 854-863.
19316. Studies on the morphology of the Onagraceae. V. Zauschneria
latifolia, typical of a genus characterised by irregular embryology. Ann. N. Y.
Acad. Sci. 33: 1-26. Johnson, D. S. 1900. On the endosperm and embryo of Peperomia pellucida.
Bot. Gaz. 80: 1-11. . 1914. Studies in the development of the Piperaceae. II. The structure
and seed development of Peperomia hispidula. Amer. Jour. Bot. 1: 323-
339, 357-397. Johri, B. M. 1935a. Studies in the family Alismaceae. I. Limnophyton obtusi-
folium Miq. Jour. Indian Bot. Soc. 14: 49-66. . 19356. Studies in the family Alismaceae. II. Sagittaria sagittifolia L.
Proc. Indian Acad. Sci. Sect. B. 1: 340-348. . 1935c. Studies in the family Alismaceae. III. Sagittaria guayanensis
H.B.K. and S. latifolia Willd. Proc. Indian Acad. Sci. Sect. B. 2: 33-48. . 1936a. Studies in the family Alismaceae. IV. Alisma plantago L.;
A. plantago-aquatica L., and Sagittaria graminea Mich. Proc. Indian Acad.
Sci. Sect. B. 4: 128-138. . 19366. The life history of Butomopsis lanceolata Kunth. Proc. Indian
Acad. Sci. Sect. B. 4: 139-162.
— . 1938a. The embryo sac of Limnocharis emarginata L. New Phytol.
37:279-285.
19386. The embryo sac of Hydrocleis nymphoides Buchen. Beihefte
bot. Centbl. 48A: 165-172.
Johri, B. M. and Maheshwari, P. 1950. The development of the embryo sac,
embryo, and endosperm in Helixanthera ligustrina (Wall.) Dans. (In press.)
THE FEMALE GAMETOPHYTE 149
Jones, H. A., and Emsweller, S. L. 1936. Development of the flower and macro-
gametophyte of Allium cepa. Hilgardia 10: 415-454. Joshi, A. C. 1936. A note on the antipodals of Digera arvensis Forsk. Current
Sci. [India] 4: 741-742. and Venkateswarlu, J. 1935. A case of reversed polarity in the embryo
sac. Ann. Bot. 49:841-843. Juliano, J. B. 1934. Studies on the morphology of the Meliaceae. II. Sterility
in santol, Sandoricum kootjape (Burm. F.) Merill. Philippine Jour. Agr. 23:
253-266. Kausik, S. B. 1938. Pollen development and seed formation in Utricularia
coerulea L. Beihefte bot. Centbl. 58A: 365-578. Khan, R. 1942. A contribution to the embryology of Jussieua repens Linn.
Jour. Indian Bot. Soc. 21: 267-2S2. Kirkwood, J. E. 1905. The comparative embryology of the Cucurbitaceae.
Bui. N. Y. Bot. Gard. 3: 313-402. Krishna Iyengar, C. V. 1940. Development of embryo sac and endosperm-
haustoria in some members of Scrophularineae. IV. Vandellia hirsuta Ham.
and V. scabra Benth. Jour. Indian Bot. Soc. 18: 179-189. . 1941. Development of the embryo sac and endosperm haustoria in
Torenia cordifolia Roxb. and T. hirsuta Benth. Proc. Natl. Inst. Sci. India
7:61-71. Lagerberg, T. 1909. Studien iiber die Entwicklungsgeschichte und systematische
Stellung von Adoxa moschatellina L. Svenska Vet.-Akad. Handl. 44(4) : 1-86. Madge, M. 1934. Nuclear migrations in Hcdychium. Proc. Linn. Soc. Lon- don 146: 108-109. Magnus, W. 1913. Die atypische Embryosackentwicklung der Podostomaceen.
Flora 105: 275-336. Maheshwari, P. 1937. A critical review of the types of embryo sacs in angio-
sperms. New Phytol. 36: 359-417. . 1941. Recent work on the types of embryo sacs in angiosperms — a
critical review. Jour. Indian Bot. Soc. 20: 229-261. . 1946a. The Fritillaria type of embryo sac: a critical review. Jour.
Indian Bot. Soc, M. O. P. Iyengar Comm. Vol., pp. 101-119. . 19466. The Adoxa type of embryo sac — a critical review. Lloydia 9:
73-113.
. 1947. Tetranucleate embryo sacs in angiosperms. Lloydia 10: 1-18.
. 1948. The angiosperm embryo sac. Bot. Rev. 14: 1-56.
and Haque, A. 1949. The embryo sac of Chrysanthemum parthenium L.
(Bernh). New Phytol. 48: 255-258.
and Johri, B. M. 1941. The embryo sac of Acalypha indica L. Beihefte
bot. Centbl. 61A: 125-136. Maheshwari, P. and Singh, Bahadur. 1950. The embryology of Macrosolen
cochinchinensis Van Tieghem. (In press.) and Singh, Balwant. 1943. Studies in the family Alismaceae. V. The
embryology of Machaerocarpus californicus (Torr.) Small. Proc. Natl. Inst.
Sci. India 9: 311-322.
150 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
Maheshwari and Srinivasan, A. R. 1944. A contribution to the embryology of
Rudbeckia bicolor Nutt. New Phytol. 43: 135-142. Martinoli, G. 1939. Contribute all'embriologia dalle Asteracaae. I— III. Nu-
ovo Gior. Bot. Ital. N.S. 45: 259-298. Mason, C. T. 1949. Development of the embryo sac in the genus Limnanthes.
Amer. Jour. Bot. 36: 799. Mathur, K. L., and Khan, R. 1941. The development of the embryo sac in
Vogelia indica Lamk. Proc. Indian Acad. Sci. Sect. B. 13: 330-388. Mauritzon, J. 1933. "Studien liber die Embryologie der Familien Crassulaceae
und Saxifragaceae." Diss. Lund. . 1939. Contributions to the embryology of the orders Rosales and
Myrtales. Lunds Univ. Arsskr. N.F. Avd. II, 35(2) : 1-120. Messeri, A. 1931. Ricerche embriologiche e cariologiche sopra i gsnri Allium
e Nothoscordum. Nuovo Gior. Bot. Ital. N.S. 33: 409-441. Modilewski, J. 1908. Zur Embryobildung von Gunnera chilensis. Ber. deutsch.
bot. Gesell. 26A: 550-556. . 1909. Zur Embryobildung von einiger Onagraceen. Ber. deutsch. bot.
Gesell. 27: 287-291. . 1910. Weitere Beitrage zur Embryobildung einiger Euphorbiaceen.
Ber. deutsch. bot. Gesell. 28: 414-418. . 1911. Uber die abnormale Embryosackentwicklung bai Euphorbia p%-
lustris L. und anderen Euphorbiaceen. Bar. deutsch. bot. Gasall. 23: 433-436. . 1918. [Cytological and embryological studies on Neollia nidus aris (L.)
Rich.] Verh. Kiewer Ges. Naturf. 23 : 1-55.
1928. Weitere Beitrage zur Embryologie und Cytologie von Allium-
Arten. Bui. Jard. Bot. de Kieff 7/8: 57-64. Narayanaswami, S. 1940. Megasporogenesis and the origin of triploids in Szc-
charum. Indian Jour. Agr. Sci. 10: 531-551. Newman, I. V. 1934. Studies in the Australian acacias. IV. The life history
of Acacia baileyana F.v.M. Proc. Linn. Soc. N. S. Wales 53: 277-313. Oehler, E. 1927. Entwicklungsgeschichtlich-zytologische Untersuchungen an ei-
nigen saprophytischen Gentianaceen. Planta 3: 611-733. Oksijuk, P. 1927. Entwicklungsgeschichte der Zuckerriibe (Beta vulgaris). Bui.
Jard. Bot. de Kieff 5/6: 145-164. . 1937. Vergleichende zytologisch-embryologische Untersuchung dar Fa-
milie Resedaceae. I. Reseda und Astrocarpus. Jour. Inst. Bot. Acad. Sci.
Ukraine 12: 3-81. Osterwalder, A. 1898. Beitrage zur Embryologie von Aconitum nipellus L.
Flora 85: 254-292. Pace, L. 1907. Fertilization in Cypripedium. Bot. Gaz. 41: 353-374.
. 1913. Apogamy in Atamosco. Bot. Gaz. 53: 376-334.
Palm, B. 1915. "Studien liber Konstruktionstypen und Entwicklungswege des
Embryosackes der Angiospermen." Diss. Stockholm. Prosina, M. N. 1930. Uber die vom Cypripedilum-Typus abweichende Embryo- sackentwicklung von Cypripedium guttatum SW. Planta 12: 532-544.
THE FEMALE GAMETOPHYTE 151
Puri, V. 1939. Studies in the order Parietales. II. A contribution to the morph- ology of Garcinia livingstonii T. Anders. Proc. Indian Acad. Sci. Sect. B. 9:
74-86. . 1941. Life history of Moringa oleifera Lamk. Jour. Indian Bot. Soc.
20: 263-284. Randolph, L. F. 1936. Developmental morphology of the caryopsis in Maize.
Jour. Agr. Res. 53: 881-916. Rauch, K. V. 1936. Cytologisch-embryologische Untersuchungen an Scumda
atropurpurea Dans, und Dendrophthoe pentandra Miq. Ber. schweiz. bot.
Gesell. 45:5-61. Razi, B. A. 1949. Embryological studies of two members of the Podostomaeeae.
Bot. Gaz. 111:211-218. Reed, E. L. 1924. Anatomy, embryology and ecology of Arachis hypogea. Bot.
Gaz. 78: 289-310. Romanov, I. D. 1936. Die Embryosackentwicklung in der Gattung Gagea
Salisb. Planta 25: 438-459. . 1938. Eine neue Form des Embryosackes von Adoxa-Typus bei Tulipa
tetraphylla und T. ostrovskiana. Compt. Rend. (Dok.) Acad, des Sci. U.R.S.S.
19: 113-115. . 1939. Two new forms of embryo sac in the genus Tidipa. Compt.
Rend. (Dok.) Acad, des Sci. U.R.S.S. 22: 139-141.
Ros6n, W. 1947. The female gametophyte in Nolana and endosperm develop- ment in Tubiflorae. Bot. Notiser 1947, pp. 372-382.
Rutishauser, A. 1935. Entwicklungsgeschichtliche und zytologische Untersuch- ungen an Korthalsella dacrydii (Ridl.) Danser. Ber. schweiz. bot. Gesell. 44: 389-436.
. 1937a. Entwicklungsgeschichtliche Untersuchungen an Thesium rostra-
tum M.u.K. Mitt. Naturf. Gesell. Schaffhausen 13: 25-47.
19376. Blutenmorphologische und cytologische Untersuchungen an den Viscoideen Korthalsella opuntia Morr. und Ginalloa linearis Dans. Ber. schweiz. bot. Gesell. 47: 5-28.
1945. Zur Embryologie amphimiktischer Potentillen. Ber. schweiz.
bot. Gesell. 55: 19-32. Samuels, J. A. 1912. Etudes sur le deVeloppement du sac embryonnaire et sur
la f^condation du Gunnera macrophylla Bl. Arch. f. Zellforsch. 8: 52-120. Schaeppi, H., and Steindl, F. 1942. Blutenmorphologische und embryologische
Untersuchungen an Loranthoideen. Vrtljschr. naturf. Gesell. Zurich 87:
301-372. and . 1945. Blutenmorphologische und embryologische Unter- suchungen an einigen Viscoideen. Vrtljschr. naturf. Gesell. Zurich 99: 1-46. Schulle, H. 1933. Zur Entwicklungsgeschichte von Thesium montanum Ehrh.
Flora 27: 140-184. Sharp, L. W. 1912. The orchid embryo sac. Bot. Gaz. 51: 373-3S5. Shattuck, C. H. 1905. A morphological study of Ulmus americana. Bot. Gaz.
49:209-223.
152 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
Singh, Bahadur. 1950. The embryology of Dendrophthoe fzlcata (Linn, fil.)
Ettingshausen. Proc. 37th Indian Sci. Cong. Sect. Bot. Smith, R. W. 1911. The tetranucleate embryo sac of Clintonia. Bot. Gaz.
52:209-217. Smith, F. H. 1943. Megagametophyte of Clintonia. Bot. Gaz. 105: 233-257. Soueges, E. C. R. 1940. Embryog6nie des Haloragacees. Developpement de
l'embryon chez le Myriophyllum aUerniflorum DC. Compt. Rend. Acad, des
Sci. Paris 211: 185-188. ■ . 1943. Embryogenie des Fumariac6es. L'origine et les premieres divi- sions de la cellule embryonnaire proprement dite chez YHypecoum procumbens
L. Compt. Rend. Acad, des Sci. Paris 216: 310-311. Steindl, F. 1935. Pollen- und Embryosackentvvicklung bei Viscum album L. und
V. articulatum Burm. Ber. schweiz. bot. Gesell. 44: 343-388. Stenar, H. 1925a. Embryologische und zytologische Studien liber Limnanthes
douglassi R.Br. Svensk Bot. Tidskr. 19 : 133-152. . 19256. "Embryologische Studien. I. Zur Embryologie der Columni-
feren. II. Die Embryologie der Amaryllideen." Diss. Upsala. . 1928. Zur Embryologie der Asphodeline-Gruppe. Ein Beitrag zur sys-
tematischen Stellung der Gattungen Bulbine und Paradisia. Svensk Bot.
Tidskr. 22: 145-159. . 1932. Studien liber die Entwicklungsgeschichte von N othoscordum frag-
rans Kunth u. N. striatum Kunth. Svensk Bot. Tidskr. 26: 25-44. . 1934. Embryologische und zytologische Beobachtungen liber Majan-
themum bifolium und Smilacina stellata. Arkiv. for Bot. 26: 1-20. . 1937. Zur Embryosackentvvicklung einiger Malpighiazeen. Bot. Noti-
ser 1937, pp. 110-118. . 1940. Biologiska och embryologiska notiser rorande Calypso bulbosa
(L) Oakes. Jamten Heimbygdas Forlag, Ostersund 1940, pp. 184-189.
1941. tlber die Entwicklung des Embryosackes bei Convallaria majalis
L. Bot. Notiser 1941, pp. 123-128. Stephens, E. L. 1909. The embryo sac and embryo of certain Penaeaceae. Ann.
Bot. 23:363-378. Stiffler, E. G. 1925. Development of embryo sac in Gasteria, Cyrtanthus and
Veltheimia. Bot. Gaz. 79: 207-216. Stolt, K. A. H. 1921. Zur Embryologie der Gentianaceen und Menyanthaceen.
K. Svenska Vet.-Akad. Handl. 61(14): 1-56. ■ . 1928. Die Embryologie von Myriophyllum aUerniflorum DC. Svensk
Bot. Tidskr. 22: 305-319. Strasburger, E. 1879. "Die Angiospermen und die Gymnospermen." Jena. Subba Rao, A. M. 1940. Studies in the Malpighiaceae. I. Embryo sac develop- ment and embryogeny in the genera Hiptage, Banisteria and Stigmatophyl-
lum. Jour. Indian Bot. Soc. 18: 145-156. . 1941. Studies in the Malpighiaceae. II. Structure and development
of the ovules and embryo sacs of Malpighia coccigera Linn, and Tristellateia
australis. Proc. Natl. Inst. Sci. India 7: 393-404. Svensson, H. G. 1925. "Zur Embryologie der Hydrophyllaceen, Borraginaceen
und Heliotropiaceen." Diss. Upsala
THE FEMALE GAMETOPHYTE 153
Swamy, B. G. L. 1945. Embryo sac and fertilization in Cypripedium spectabile. Bot. Gaz. 107: 291-295.
. 1946. Inverted polarity of the embryo sac of angiosperms and its rela- tion to the archegonium theory. Ann. Bot. 9: 171-183.
. 1949a. Embryological studies in the Orchidaceae. I. Gametophytes.
Amer. Midland Nat. 41: 184-201.
. 19496. A reinvestigation of the embryo sac of Maianthemum canadense.
Bui. Torrey Bot. Club 76: 17-23. Tackholm, G. 1915. Beobachtungen liber die Samenentwicklung einiger Onagra-
ceen. Svensk Bot. Tidskr. 9 : 294-361 . Tateishi, S. 1927. On the development of the embryo sac and fertilization of
Acalypha australis L. (preliminary note). Bot. Mag. [Tokyo] 51: 477-485. Thirumalachar, M. J., and Khan, B. A. 1941. Megasporogenesis and endosperm
formation in Eriodendron anfractuosum DC. Proc. Indian Acad. Sci. Sect. B.
14:461-465. Treub, M. 1883. Observations sur les Loranthacees. IV. Loranthus pentan-
drus L. Ann. Jard. Bot. Buitenzorg 3 : 184-189. and Mellink, J. 1880. Notice sur le deVeloppement du sac embryonnaire
dans quelques angiospermes. Arch. N6erld. des Sci. 15: 452-457. Venkateswarlu, J. 1937. A contribution to the embryology of the Sonneratia-
ceae. Proc. Indian Acad. Sci. Sect. B. 5: 206-223. Ventura, M. 1934. Sulla poliembryonia di Mallotus japonicus Muell. Arg.
Ann. di Bot. 20: 568-578. . 1937. Osservazioni embryologiche su Arechavalelaia uruguayensis Speg.
Ann. diBot. 21:527-533. Walker, R. I. 1944. Chromosome number, megasporogenesis and development
of embryo sac of Clintonia. Bui. Torrey Bot. Club 71 : 529-535. Walker, R. I. 1950. Megasporogenesis and development of megagametophyte
in Ulmus. Amer. Jour. Bot. 37: 47-52. Weber, E. 1929. Entwicklungsgeschichtliche Untersuchungen liber die Gattung
Allium. Bot. Arch. 25: 1-44. Weinstein, A. J. 1926. Cytological studies on Phaseolus vulgaris. Amer. Jour.
Bot. 13:248-263. Weinzieher, S. 1914. Beitrage zur Entwicklungsgeschichte von Xyris indica
L. Flora 106: 393-432. Went, F. A. F. C. 1910. Untersuchungen liber Podostomaceen. I. Verhandel.
K. Akad. van Wetensch. te Amsterdam II, 16(1). . 1912. Untersuchungen liber Podostomaceen. II. Verhandel. K. Akad.
van Wetensch. te Amsterdam II, 17(2).
1926. Untersuchungen liber Podostomaceen. III. Verhandel. K. Akad.
van Wetensch. te Amsterdam II, 25(1). Yamaura, A. 1933. Karyologische und embryologische Studien liber einige
Bambusa -Arten. (Vorlaufige Mitteilung). Bot. Mag. [Tokyo] 47: 551-555. York, H. H. 1913. The origin and development of the embryo sac and embryo
of Dendrophthora opuntioides and D. gracile. I and II. Bot. Gaz. 56: 89-111,
200-216.
CHAPTER 5 THE MALE GAMETOPHYTE1
The development of the male gametophyte is remarkably uniform in angiosperms. The microspore, which is the first cell of the gametophyte generation, undergoes only two divisions. The first division gives rise to a large vegetative cell and a small generative cell (Fig. 95 A-F). The second, which concerns only the generative cell, may take place either in the pollen grain (Fig. 95G-H) or in the pollen tube (Fig. 95/-/) and gives rise to the two male gametes. Details of the process may be considered under the following heads : microspore, formation of the vegetative and generative cells, divi- sion of the generative cell, male "cells" or "nuclei," and vegetative nucleus.
Microspore. The newly formed microspore has a very dense cytoplasm with a centrally situated nucleus, but the cell rapidly increases in volume and the accompanying vacuolation is followed by a displacement of the nucleus from the center to a place adjacent to the wall. In most tropical plants the nucleus begins to divide almost immediately, but in plants belonging to colder regions there is often a resting stage lasting from a few days to several weeks. To mention a few instances, in Tradescantia reflexa the resting period of the microspore is about four days or less, in Styrax obassia about a week, and in Himantoglossum hircinum between two to three weeks. In Uvularia sessilifolia, Empetrum nigrum, and Betula odorata, the microspores are said to pass the entire winter in the uninucleate stage (for further information, see Dahlgren, 1915; Finn, 1937a).
Formation of Vegetative and Generative Cells. The first division of the microspore gives rise to the vegetative and generative cells. Geitler (1935) noted that the metaphase spindle usually shows a
1 For more detailed information on the development and organization of the male gametophyte, see Wulff and Maheshwari (1938) and Maheshwari (1949). The technique for the study of the male gametophyte has been reviewed in another paper by Maheshwari and Wulff (1937).
154
THE MALE GAMETOPHYTE
155
pronounced asymmetry, the wallward pole being blunt and the free pole acute. More recent studies (Brumfield, 1941) seem to indicate that this asymmetry is associated with the form of the prophase nucleus. In Allium, where the nucleus is strongly flattened on the
Fig. 95. Diagram to illustrate the more important stages in development of male gametophyte. A, newly formed microspore. B, older stage, showing vacuolation and wallward position of microspore nucleus. C, microspore nucleus dividing.
D, division completed; two-celled stage, showing vegetative and generative cells.
E, generative cell losing contact with wall. F, generative cell lying free in cyto- plasm of vegetative cell. G, H, division of generative cell in pollen grain. /, J, division of generative cell in pollen tube. (After Maheshwari, 19/f9.)
wallward side, the asymmetry is extreme (Fig. 96); in Pancratium, where it is only slightly flattened, the asymmetry is much less pro- nounced; and Tradescantia shows an intermediate condition. The direct cause of the asymmetry has been attributed to a difference in
156
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
the time of development of the two spindle poles, the wallward or generative pole developing more slowly than the vegetative, pre- sumably because of the smaller amount of cytoplasm associated with the former. With the onset of the anaphase, the asymmetry becomes less pronounced. In the telophase the generative chromo- somes are arranged in a plane surface parallel to the inner wall of the microspore, while the vegetative ones form a somewhat hemi- spherical pattern.
Symmetrical spindles have been observed only occasionally. To mention a few examples, in Asclepias (Gager, 1902) and Anthericum (Geitler, 1935) both the poles of the spindle are blunt; in Adoxa (Lagerberg, 1909) both are more or less pointed; and in Podophyllum
Fig. 96. Allium cermium, first division of microspore, phase. C, end of anaphase. {After Brumfield, 1941.)
meta-
(Darlington, 1936) both symmetrical and asymmetrical spindles are said to occur in the same loculus. Further, in Adoxa (Lagerberg, 1909), Myricaria (Frisendahl, 1912), Sambucus (Schurhoff, 1921), Colylanthera (Oehler, 1927), and Uvularia (Geitler, 1935) the spindle is not situated near the wall of the pollen grain but occupies almost the entire width of the latter. In any case the cells formed by the division are always unequal, although the conditions which bring about this result are not clearly understood. In Cuscuta (Fedort- schuk, 1931) and Slrychnos (Mohrbutter, 1936), where the daughter cells are sometimes of the same size, this is clearly an abnormality leading to the formation of double microspores, each of which is des- tined to divide again to give rise to the vegetative and generative cells (Fig. 99F). Double pollen grains comprising two units, each with its own generative and vegetative cells, have also been figured in Podos- temon subulalus (Magnus, 1913) (Fig. 9QH, I). The separating wall between the two pollen grains is pitted (Fig. 99/) and only one of them
THE MALE GAMETOPHYTE
157
produces a pollen tube, the other presumably serving as a source of food material.
It may be noted that unlike the reduction divisions, which occur more or less simultaneously in all the microspore mother cells of an anther, the microspores usually divide without any such synchron- ization, and the same loculus may show different although not widely separated stages of division and development. In those plants in which the microspores remain together in a tetrad, all
C
E
H
1
K
L
Fig. 97. A-I, Zostera marina, division of microspore to form vegetative and generative cells. Pollen grains are so long that only a part of each is shown. (After Rosenberg, 1901 .) J, Vaccinium vitis idaea, pollen tetrad showing generative cell cut off toward outer side of each microspore. (After Samuelsson, 1918.) K, Xyris indica, pollen tetrad, showing generative cell cut off toward inner side of each microspore. (After Weinzieher, 1914.) L, Acacia baileyana, pollinium, showing various stages in division of microspore. (After Newman, 1934.)
four cells in a tetrad are usually in the same stage of division, but not all the tetrads of an anther. A complete synchronization may perhaps be expected only where the microspores are united into pollinia (Mimosaceae, Asclepiadaceae, and Orchidaceae), for here the cells probably exercise some influence over one another through the uncuticularized walls which lie between them (Barber, 1942). Exceptions do occur, however, even in such cases. Figure 97 L of the pollinium of Acacia baileyana (Newman, 1934) shows one of the microspores in prophase, another with the tube and generative cells already formed, and the rest in various intermediate stages.
158 INTRODUCTION TO EMBRYOLOGY OF ANG10SPERMS
Goebel (1933) thought that in the angiosperms the generative cell is always cut off on the distal (i.e., ventral) side of the micro- spore. Geitler (1935) showed, however, that there is no such uni- formity and that the generative cell may be cut off either on the outer side (Fig. 97 J, L), or on the inner side (Fig. 97 K, 98), or on a radial wall (Fig. 97 A-I), or in a corner instead of the middle of the radial wall. To cite a few examples, the first-named condition has
Fig. 98. Microsporogenesis and development of male gametophyte in Juncus filiformis (A-F) and J. squarrosus (G-I). A, interkinesis after Meiosis I, showing formation of ephemeral cell plate. B, microspore nuclei. C, microspore nuclei in prophase; note intervening plasma membranes. D, microspore nuclei in ana- phase. E, F, formation of vegetative and generative cells. G, one member of tetrad, showing generative cell in late anaphase. H, same, division nearly com- pleted. 7, older stage, showing vegetative nucleus and two sperm cells. (After Wulff, 1939a.)
been reported in Elodea (Wylie, 1904), Vaccinium (Samuelsson' 1913), Albizzia (Maheshwari, 1931), Acacia (Newman, 1934), Asimina (Locke, 1936), and most members of the Orchidaceae (Swamy, 1949); the second in Symplocarpus (Duggar, 1900), Xyris (Weinzieher, 1914), Erica (Geitler, 1935), Juncus (Wulff, 1939a), Cyanastrum (Nietsch, 1941), and most members of the Cyperaceae (Piech, 1928); the third in Allium (Geitler, 1935); and the fourth
THE MALE GAMETOPHYTE 159
in Lilium (Strasburger, 1908), Anthericwn, and Convallaria (Geitler, 1935).
Whatever the position may be, it is usually constant in individuals of the same species and sometimes in all the species of a genus or family and is thus a character of some systematic significance. Un- fortunately, it can be recognized most clearly only in those plants in which the pollen grains remain together in tetrads. In most genera the microspores round up at such an early stage that it be- comes impossible to distinguish one side from the other, although even here the position of the germ pores and furrows often serves as a useful guide. An important point to keep in mind, however, is that very soon the generative cell loses contact with the wall of the microspore, and after this has happened it may change its posi- tion in the pollen grain and come to lie in almost any part of it.2
There is considerable variation in the form of the generative cell. Usually it is elliptical, lenticular, or spindle-shaped, but in Cuscuta (Finn, 19376) and Ottelia (Islam, 1950) it becomes long enough to occupy the entire width of the pollen grain, coming quite close to the inner wall of the latter on either side. In Monochoria (Banerji and Haldar, 1942) it is one and a half times as long as the diameter of the pollen grain and is accommodated in the latter only by the incurving of its whip-like ends. In Campanula ranunculoides (Schnarf, 1937) the two ends are dissimilar, one being pointed and the other more or less blunt and swollen so as to look like a "head." There are also occasional reports of changes in the form of the generative cell. More frequently, however, such appearances are merely due to the plane of sectioning. A spindle-shaped cell appears round when cut across and oval when cut obliquely.
In fixed material the cytoplasm of the generative cell is usually distinguishable from that of the vegetative by its hyaline appearance and general lack of food materials. Plastids (Ruhland and Wetzel, 1924; Krupko, 1926) (Fig. 990) and chondriosomes have, however, been demonstrated in a few cases, and some recent studies on living pollen grains and pollen tubes (Benetskaia, 1939; Kostriukova, 1939a, b; Kostriukova and Benetskaia, 1939) have confirmed the
2 This gradual extension of the vegetative cytoplasm around the generative cell and the consequent "engulfing" of the latter has been referred to by several workers, viz., Friemann (1910), Wefelscheid (1911), Capoor (1937a), and others.
160 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
THE MALE GAMETOPHYTE 161
presence of a vacuome and mitochondria in the generative as well as the sperm cells of Narcissus, Asclepias, Vinca, Crinum, and Lilium. Mention may also be made of the "colored bodies" de- scribed by Kostriukova (19396) in living pollen tubes of Lilium martagon. He saw two structures of a pale greenish color, one at each end of the generative nucleus. In older stages these bodies were found to divide and occupy similar positions in the sperm cells (Fig. 100). They were not recognizable in fixed material, but in their places small areolae were seen which stained black with osmic acid. The author concludes that they probably correspond with the structures described as Golgi bodies, but a further study is of course necessary to confirm this.3
Regarding the contents of the vegetative cell, starch and fat are the most conspicuous substances. The distinction between starchy and fatty pollen has been recognized for a long time and their pos- sible ecological significance has been a subject of much interest (see Tischler, 1917; Kuhlwein, 1937). Luxemburg (1927) traced the origin of the starch grains and fat bodies from plastids in the pollen grains of several members of the Malvaceae, and believes that the
3 In his book "The Cytoplasm of the Plant Cell" Guilliermond (1941) remarks that "there is no Golgi apparatus in plants" and that all formations described as Golgi apparatus are elements belonging either to the vacuolar system or to the chondriome.
Fig. 99. Pollen grains of various angiosperms. A, Cuscuta epithymum, mature pollen grain, showing vegetative nucleus and two male cells. B, pollen grain, showing two vegetative nuclei and dividing generative nucleus. C, vegetative nucleus and three sperms. D, two vegetative nuclei. E, vegetative nucleus, two sperms, and prothallial cell. F, microspore has divided into two parts, of which one on right shows vegetative as well as generative nucleus. (After Fedort- schuk, 1931.) G, Atriplex hymenelytra, pollen grain showing prothallial cell (?), vegetative nucleus, and two male nuclei. (After Billings, 1934.) H, Podostemon subulatus, double pollen grain. /, older stage, in which microspore nucleus of each has divided to form vegetative and generative nuclei; small bodies outside nuclei are starch grains. /, partition wall between the two cells, showing pits. (After Magnus, 1913.) K, Vinca herbacea, dumbbell-shaped pollen grain with two pairs of sperm cells and two vegetative nuclei. (After Finn, 192S.) L, Erythronium americanum, pollen grain, showing vegetative and generative cells. M, same, showing amoeboid nature of generative cell. (After Schaffner, 1901.) N, Wormia suffruticosa, pollen grain, showing large crystal. (After Paetow, 1931.) 0, Lupinus luteus, pollen grain, showing chloroplasts in generative cell as seen after silver impregnation. (Ruhland and Wetzel, 1924.)
162 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
plastids in turn arise either from preexisting plastids or from chon- driosomes. In very young pollen grains the reserve food consists almost entirely of droplets of fat, and starch formation begins only after the pollen grains have increased in size.
WW
A
Fig. 100. Portions of pollen tubes of Lilium martagon showing behavior of "colored bodies" inside generative cell and sperm cells, as seen in living condition. A, generative cell in division, showing colored body at either end. B, enlarged view of one end of generative cell, showing detail of colored body. C-E, stages in division of generative cell. F, G, sperm cells, showing the colored bodies. (After Kostriukova, 1939b.)
Certain proteinaceous bodies have also been reported in pollen grains and pollen tubes (Fig. 101), but their exact origin remains unknown. They probably arise in plastids but soon become liber- ated in the general cytoplasm of the pollen grain and pollen tube. Most remarkable of all are the large transparent protein crystals of Wormia suffruticosa (Paetow, 1931) (Fig. 99N), although these are of a transitory nature and disappear during the later stages in the maturation of the pollen grain.
THE MALE GAMETOPHYTE
163
Division of Generative Cell. The generative cell may divide either in the pollen grain (Fig. 95G-H) or in the pollen tube (Fig. 95I-J). Formerly the second condition was believed to be the more frequent, but during recent years three-celled pollen grains
B
D
4aV
G
H
E F
Fig. 101. Stages in development of male gametophyte of Asclepias, showing- protein granules inside cytoplasm of pollen grain and pollen tube. A, microspore mother cells. B, tetrad of microspores. C-F, microspores. G, pollen grain, showing vegetative nucleus and two sperm nuclei. H, terminal portion of pollen tube. (After Guignard, 1922.)
have been reported in several genera (see Schnarf, 1939) and it seems certain that many of the older records were based on a study of immature pollen.4
4 As the pollen grain grows older, the vacuoles become smaller and more evenly distributed and finally they disappear almost entirely so that with the usual methods of fixing and staining the mature pollen grain, like the young microspore, again shows a dense cytoplasm devoid of all conspicuous vacuolation. This is such a constant feature in most angiosperms (excluding some aquatics) that it serves as a useful check for judging whether a pollen grain is fully mature or not (see Schnarf, 1937).
164 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
There is also considerable evidence to indicate that even in those plants in which the pollen grains are shed in the two-celled condi- tion, the generative nucleus is already in the prophase stage and the process of division is merely continued in the pollen tube. Some- times the nucleus may even show a pro-metaphase stage which is distinguishable from a typical metaphase only by the delay in the dissolution of the nuclear membrane and the organization of the spindle. This has been demonstrated very clearly in Impatiens (Wulff, 1934; Heitz and Resende, 1936), Bulbine (Geitler, 1942) and other plants.
Occasionally both two- and three-celled pollen grains have been reported in the same plant, as in the cleistogamous flowers of Viola (West, 1930), in Dionaea (Smith, 1929), Circaeaster (Junell, 1931), Nicotiana (Poddubnaja-Arnoldi, 1936), Epimedium, and Iris (Schnarf, 1937), but this is probably due to environmental influ- ences. Poddubnaja-Arnoldi (1936) found that, in several kinds of pollen grains which are normally two-celled, the generative nucleus divided before germination if the grains were kept for some time on a sugar-agar substrate. Eigsti (1941) was similarly able to induce a precocious division of the generative cell in the pollen grains of Polygonatum canaliculatum. In Holoptelea integrifolia (Capoor, 19376) the pollen grains are shed at the two-celled stage, but the generative cell divides on the surface of the stigma before the pollen tube has started to grow.6
Details of the division of the generative cell vary depending on whether it takes place in the pollen grain or in the pollen tube. In the former case, spindle fibers and a normal metaphase plate have been regularly observed, and the process does not seem to differ in any essential way from a normal mitosis. Cytokinesis, resulting in a bipartitioning of the cell, may take place either by a process of furrowing as in Juncus (Wulff, 1939a) (Fig. 98//"), or by the laying down of a cell plate as in Asclepias (Finn, 1925) and Portulaca (D. C. Cooper, 1935). Witmer (1937), who observed both cell plates and constriction furrows in Vallisneria, states that in his material these two factors varied in importance.6 In some pollen
6 In Euphorbia terracina (D'Amato, 1947), which is at the other extreme, the division occurs only after the pollen tube has entered the embryo sac and its tip has come to lie by the side of the egg.
6 See also Kausik and Rao (1942).
THE MALE GAMETOPHYTE 165
grains a definite cell plate was laid down in the beginning, but it soon faded away, leaving the final separation of the sperms to a constriction furrow which arose soon afterwards. In others the cell plate persisted, and the progress of the constriction furrow was arrested in this region although evident on either side of it; here the splitting of the cell plate divided the generative cell before the constriction could make much progress.
It has proved more difficult to understand the mechanism of the division when it occurs in the pollen tube. The chief points in question are: (1) whether a regular metaphase plate is formed during the division, (2) whether spindle fibers are present or absent, and (3) whether cytokinesis takes place by constriction or by cell-plate formation.
Nawaschin (1910), O'Mara (1933), Wulff and Raghavan (1937), Raghavan et al. (1939), and several other workers failed to find any regular metaphase plates in the plants studied by them, viz., Lilium martagon, L. regale, Nemophila insignis, and Impatiens balsamina. On the other hand, Cooper (1936) reported their occurrence to be a regular feature in Lilium regale, L. auratum, and L. philippinense (Fig. 102), and believes that O'Mara's (1933) figures of an "irregular metaphase" really represent a late prophase, the true metaphase having been missed by him. Upcott (1936) and Madge (1936) also found a metaphase plate in Tulipa and Hedychium respectively, the only important difference being its oblique orientation which gives more space to the chromosomes for their proper alignment. More recently, well-differentiated metaphase plates have been re- corded in Eichhornia (Banerji and Gangulee, 1937), Tulipa, Amar- yllis, Nicotiana, Forsythia, Camellia, Bryophyllum (Johnston, 1941), and Eschscholtzia (Beatty, 1943) (Fig. 103).
Regarding the presence or absence of spindle fibers, Nawaschin (1909) in Lilium martagon, Welsford (1914) in L. auratum and L. martagon, O'Mara (1933) in L. regale, Trankowsky (1931) in Convallaria majalis and Galanthus nivalis, Fuchs (1936) in Elaeag- nus angustifolius , Wunderlich (1937) in Muscari racemosum and M. comosum, Finn (1939) in Phlomis tuberosa, Raghavan et al. (1939) in Impatiens balsamina, and several other workers failed to find a spindle. On the other hand, Trankowsky (1931) in Hemero- callis fulva, Cooper (1936) in Lilium auratum, L. regale, and L. philippinense (Fig. 102), Madge (1936) in Hedychium gardnerianum,
166
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
Upcott (1936) in several species of Tulipa, Eigsti (1939) in Lilium canadense, L. speciosum, L. auratum, Polygonatum commutatum, Convallaria majalis, and Tradescantia reflexa, Johnston (1941) in Tulipa gesneriana, Amaryllis spp., Nicotiana tabacum, Forsythia viridissima, Camellia japonica, and Bryophyllum pinnatum, and
E
Fig. 102. Division of generative cell of Lilium regale as seen in pollen tubes grown in culture. A, prophase. B-C, metaphase chromosomes advancing to equatorial plate. D-F, metaphase. G, three chromosomes at metaphase. II, two chromo- somes at early anaphase. /, late anaphase showing cell-plate formation. J, two male gametes; cell plate completed. {After Cooper, 1936.)
Beatty (1943) in Eschscholtzia calif ornica (Fig. 103) have emphasized that spindle fibers are present and perform the same functions as in normal mitosis.
Coming finally to the mode of cytokinesis, Raghavan ct al. (1939) in Impaliens, Banerji and Gangulee (1937) in Eichhomia, and several other authors have reported that the division of the generative cell occurs by a constriction. Eigsti (1940) also states that cell plates
THE MALE GAMETOPHYTE
4S7
are difficult to find and are temporary structures without any special significance. On the other hand, very distinct cell plates have been figured in Lilium regale (Cooper, 1936) (Fig. 102/) and a number of other plants, and their occurrence has also been confirmed from studies on living pollen tubes of Crinum hildebrandtii (Kostriukova, 1939a), Lilium martagon (Kostriukova, 19396), and Narcissus poeticus (Kostriukova and Benetskaia, 1939).
In an important and extensive work on pollen tubes, Johnston (1941) suggests that the inability to see cell plates is to be attributed
A
1
Fig. 103. Division of generative cell in Eschscholtzia californica. {After Beatty, 1943.)
to the exclusive use of nuclear stains in most studies of this type. The same explanation holds good for the frequently reported absence of spindle fibers in the division of the generative cell. Delafield's haematoxylin, he says, is much superior to Heidenhain's haemato- xylin for such purposes and should always be used for comparison. As far as present evidence goes, it may therefore be concluded that the division of the generative cell, whether it takes place in the pollen grain or in the pollen tube, occurs in a fairly regular fashion. However, in cases in which the tube is very narrow and the chromosomes are rather large, there may be some disturbance of the metaphasic alignment, resulting in their crowding or buckling. It is also probable that the metaphase stage, owing to its very short
168 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
duration, has been entirely missed in some plants, thereby creating the false impression that the nucleus passes directly from the pro- phase into the anaphase. Regarding the spindle, an intensity of staining which is adequate or just right for the chromosomes often fails to bring out the fibers, which are more clearly seen in over- stained material. Finally, the division of the cell may take place either by means of a constriction furrow or by the laying down of a cell plate.
Male "Cells" or "Nuclei." Formerly it was believed that when- ever the division of the generative cell occurs in the pollen grain it is followed by the formation of sperm cells, but if it takes place in the streaming cytoplasm of the pollen tube only nuclei are formed. Recent work has shown, however, that in all cases the male gametes are definite cells and the cytoplasmic sheath persists throughout their course in the pollen tube (see Schnarf, 1941).
Considering first the case of Lilium, which has been the favorite object for such studies, Guignard (1889) figured the male gametes of L. martagon as cells, but Koernicke (1906), Strasburger (1908) and Nawaschin (1910) believed that the cytoplasmic sheath is lost during the division of the generative cell. As to the exact time of dis- appearance of the sheath, however, these authors are not in agree- ment with one another. According to Koernicke it is lost when the generative cell is in prophase; according to Strasburger, at the metaphase stage; and according to Nawaschin, only during the telo- phase. Later, Welsford (1914) and O'Mara (1933) reported that in L. martagon and L. auratum definite sperm cells are formed, al- though eventually the cytoplasm dissolves away so that only the naked nuclei enter the embryo sac. Cooper (1936) showed, how- ever, that the male gametes persist as cells right up to the time they enter the embryo sac. This has received further confirmation from the work of Anderson (1939), who finds that the cytoplasmic sheath around the male nuclei possesses all the inclusions normally present in the vegetative cytoplasm. He explains that the failure of other workers to see the sheath is due to their use of nuclear stains, which are not suited for bringing out the cytoplasm to the best advantage.
Not only in Lilium but also in other plants, the cytoplasmic sheath around the male nuclei has been followed up to the time of their discharge in the embryo sac. Nawaschin and Finn (1913) figured a clear space around the male nuclei of Juglans, which, as Finn (1925) subsequently explained, represents a thin layer of cytoplasm
THE MALE GAMETOPHYTE 169
around them. Tschernojarow (1915), Dahlgren (1916) and Ishi- kawa (1918) demonstrated the occurrence of male cells in Myosurus, Plumbagella, and Oenothera respectively. Wulff (1933) and Finn (1935, 1940, 1941), who are the most active workers in this field, categorically state that the occurrence of male cells may be assumed in all angiosperms, and assert that in those plants in which only male nuclei have been reported, proper methods of fixing and staining will eventually reveal the thin cytoplasmic sheath around them.
Vegetative Nucleus. Earlier authors took it for granted that the vegetative nucleus (often called "tube" nucleus) had an important role in directing the growth of the pollen tube. Present evidence seems to indicate, however, that its functional importance had been greatly exaggerated.
The vegetative nucleus is not always in the distal end of the pollen tube (where it would be most expected if it had any important function in directing the growth of the tube) but frequently lies considerably behind the male gametes. When the tube becomes branched as in Aconitum, Cucurbita, and Papaver (Poddubnaja- Arnoldi, 1936), the individual branches continue their growth for an appreciable period, although only one of them contains the vegeta- tive nucleus. In Ulmus (Shattuck, 1905), Senecio, Crepis, and Secale (Poddubnaja-Arnoldi, 1936) it degenerates even before the pollen grains begin to germinate and does not enter the tube at all; nevertheless the tube continues to function normally.7 In Cheno- podium, Atriplex, and Salsola it seems to break up and diffuse into the surrounding cytoplasm (G. O. Cooper, 1935), and in Musa (Juliano and Alcala, 1933) and Senecio (Poddubnaja-Arnoldi, 1933) it fragments into small bits which seem to be quite functionless. In some other plants also the vegetative nucleus assumes a very abnormal appearance. For instance, in the pollen tubes of Viola odorata (Madge, 1929) it becomes 4 times, in Cymbidium bicolor (Swamy, 1941) 18 times, and in Vallisneria americana (Wylie, 1923) 27 times longer than broad. In a few members of the Labiatae (Finn, 1939) and in Nicotiana (Goodspeed, 1947) the elongation is sufficiently pronounced to give it a filamentous outline.
On the basis of these and other data Poddubnaja-Arnoldi (1936) regards the vegetative nucleus as a vestigial structure without any important function in the growth of the pollen tube. This view is
7 See Hewitt (1939) for other examples of an early degeneration of the vegetative nucleus.
170 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
supported by Suita (1936, 1937a, b), who studied the pollen grains of Crinum with the Feulgen method. He states that soon after its formation the vegetative nucleus increases in size and becomes amoeboid. Later it begins to stain very faintly, indicating a de- composition of the chromatin. He agrees, therefore, that it is a degenerating structure without any important function in the life of the pollen tube.
While further evidence would be welcome, it seems safe to con- clude that the old view attributing a leading role to the vegetative nucleus in the growth and direction of the pollen tube now needs modification. It is likely that these functions are really discharged by the nucleus of the generative cell itself and later by the nuclei of the two male cells formed by its division.
Development of Pollen in the Cyperaceae. The course of de- velopment described above is generally characteristic of all angio- sperms, dicotyledons as well as monocotyledons, the family Cy- peraceae being the only notable exception. Juel (1900), Stout (1912), Piech (1928) and others have shown that, of the four micro- spore nuclei produced after meiosis, only one develops further, while the other three become pushed toward one end of the mother cell (Fig. 104 A, J5). The functional nucleus, which lies in the center, divides with its spindle oriented in the direction of the long axis of the cell (Fig. 104(7, D). The cell plate, which is laid down between the vegetative nucleus and generative nucleus, extends around the latter so as to give rise to a continuous plasma membrane. The generative cell (Fig. 104Z?) soon becomes spindle-shaped and divides to form the two sperm cells (Fig. 104jP).
A few doubtful points, which need further clarification, are the following: (1) whether the functioning microspore nucleus is sepa- rated from the three nonfunctioning nuclei by a wall, (2) whether the nonfunctioning nuclei are separated from one another by walls, and (3) what the fate of the nonfunctioning nuclei may be. Tanaka (1940, 1941), who has recently discussed these questions, believes that normally a plasma membrane separates the functioning micro- spore nucleus from the three nonfunctioning nuclei and that subse- quently similar membranes arise between the latter. The non- functioning nuclei sometimes undergo one division, resulting in a pair of daughter nuclei in each of the three cells. No separating wall is formed between them, however, and they are soon absorbed.
THE MALE GAMETOPHYTE
171
D E F
Fig. 104. Development of male gametophyte of Scirpvs paluster. A, telophase of Meiosis II resulting in formation of four microspore nuclei. B, three of micro- spore nuclei pushed to one end of pollen grain; functioning nucleus in center. C, D, functioning nucleus dividing; remaining three nuclei in process of degeneration. E, pollen grain, showing vegetative and generative cells. F, generative cell dividing to form two male cells. (After Piech, 192S.)
Embryo-sac-like Pollen Grains. In 1898 Nemec noted that in the petaloid anthers of Hyacinthus orientalis the pollen grains some- times form large eight-nucleate structures showing a surprising resemblance to embryo sacs. He believed that they arose as the
172 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
result of a degeneration of the generative nucleus and three divisions of the vegetative nucleus.
De Mol (1923) observed this so-called "Nemec-phenomenon" in the anthers of other varieties of Hyacinthus orientalis which had been subjected to certain special conditions in order to obtain early flowering. He attributed the origin of the abnormality to a dupli- cation of the generative nuclei.
Stow (1930, 1934) found similar embryo-sac-like pollen grains or "pollen-embryo sacs" in the anthers of a variety called "La Victor" whose bulbs had been subjected to a temperature of 20°C. at the time of meiosis and were further "forced" in a greenhouse. He traced their development more fully than either Nemec or De Mol. At first the microspores increase in size to form large sac-like bodies (Fig. 105 A, B), after which the nucleus undergoes three successive divisions (Fig. 105C-F) to form 8 daughter nuclei. Of these, 3 lie at the end where the exine is still intact, 3 at the opposite end, and 2 in the middle. The 6 nuclei at the two poles organize into cells, while the remaining two fuse in the center (Fig. 105(7). Since the three cells at the exine end were found to remain healthy for a much longer time than those at the opposite end, Stow regards the former as corresponding to the egg and synergids, and the latter to the antipodals. In addition certain abnormal pollen-embryo sacs were also seen, showing the following types of organization: (1) 8 nuclei forming an egg, two polars, and five antipodal cells; (2) 4 nuclei forming an egg, two polars, and one antipodal cell; (3) 4 nuclei forming a polar and three antipodal cells but no egg; (4) 16 nuclei forming a 5- to 10-celled egg apparatus, one or two polars, and a few antipodal cells; and (5) more than 16 nuclei without any definite arrangement.
According to Stow, it is not the divisions of the vegetative or generative nucleus which give rise to the pollen-embryo sacs but those of the microspore nucleus itself. Once the vegetative and
Fig. 105. Development of pollen-embryo sacs in Hyacinthus orientalis. A, microspore in metaphase of first division. B, microspore showing tendency toward formation of pollen-embryo sac; nucleus is in metaphase. C, second nuclear divi- sion in pollen-embryo sac; on right, young pollen-embryo sac with undivided nucleus. D, four- and two-nucleate pollen-embryo sacs. E, division of four nuclei; metaphase. F, same; anaphase. G, well-developed pollen-embryo sac. (After Stow, 1930.)
THE MALE GAMETOPHYTE
173
k
Fig. 105.
174 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
generative cells have been differentiated, further development is quite normal and no pollen-embryo sacs are formed. Further, the pollen-embryo sacs were always accompanied by a large number of dead pollen grains, leading Stow to suggest that the latter secrete a "necrohormone" which causes an abnormal growth of the surviving pollen grains.
Stow also observed that when the pollen-embryo sacs were placed on an agar medium, together with some normal pollen grains of another variety, the pollen tubes formed from the latter coiled
B
Fig. 106. Fertilization of pollen-embryo sacs in Hyacinthus orientalis. A, normal pollen grain, showing vegetative and generative cells. B, pollen-embryo sac.
C, pollen-embryo sac affected by pollen tube from pollen grain of another variety.
D, fertilized pollen-embryo sac; the smaller nuclei are presumed to be products of division of triple fusion nucleus. (After Stow, 198/,.)
around the former (Fig. 106). Once a sperm nucleus was observed to be in process of entering the pollen-embryo sac; and in another case the pollen-embryo sac showed 16 nuclei, believed to have been derived from the divisions of a triple fusion nucleus.
In conclusion Stow says that all pollen grains are potentially capable of assuming either the male or the female form. Under normal conditions the "male potency" is dominant over the "female potency" leading to the formation of the generative cell and the male gametes; but under abnormal conditions, when there is a re- lease of necrohormones, the female potency gets the upper hand resulting in the formation of embryo-sac-like structures.
THE MALE GAMETOPHYTE 175
Shortly after the publication of Stow's papers, Naithani (1937) found embryo-sac-like pollen grains in the variety "Yellow Hammer" whose bulbs had been treated for early flowering. He confirms Stow's observations regarding the mode of development of these abnormal pollen grains, but believes their formation to be a tempera- ture effect and not the result of a liberation of necrohormones. According to him, the degeneration of the other pollen grains is not the cause but the effect of a hypertrophied growth of the more favored ones, which use up all the available food for their own growth.
More recently, pollen -embryo sacs with 8 and 16 nuclei have also been observed in another plant, Orniihogalum nutans (Geitler, 1941). Those with 8 nuclei showed the typical embryo-sac-like organization, but, contrary to Stow, Geitler interprets the three cells at the exine end of the pollen grain as the equivalents of anti- podals and the other three as equivalents of the egg and synergids.
References
Anderson, L. E. 1939. Cytoplasmic inclusions in the male gametes of Lilium'
Amer. Jour. Bot. 26: 761-766. Banerji, I., and Gangulee, H. C. 1937. Spermatogenesis in Eichhornia crassipes
Solms. Jour. Indian Bot. Soc. 16: 289-296. and Haldar, S. 1942. A contribution to the morphology and cytology
of Monochoria hastaefolia Presl. Proc. Indian Acad. Sci. Sect. B. 16: 91-106. Barber, H. N. 1942. The pollen grain division in the Orchidaceae. Jour. Genet.
43: 97-103. Beatty, A. V. 1943. The division of the generative nucleus in Eschscholtzia.
Amer. Jour. Bot. 30: 378-382. Benetskaia, G. K. 1939. Observations in vivo des gametes males dans les tubes
polliniques d'Asclepias cornutii Decsn., de Vinca major L. et de Vinca minor L.
Jour. Bot. de l'U.R.S.S. 24: 273-281. Billings, F. H. 1934. Male gametophyte of Atriplex hymenelytra. Bot. Gaz.
95: 477-484. Brumfield, R. T. 1941. Asymmetrical spindles in the first microspore division
of certain angiosperms. Amer. Jour. Bot. 28: 713-722. Capoor, S. P. 1937a. Contributions to the morphology of some Indian Liliaceae.
II. The gametophytes of Urginea indica Kunth. Beihefte bot. Centbl. 56 A:
156-170. . 19376. The life history of Holoptelea integrifolia Planch. Beihefte bot.
Centbl. 57A: 233-249. Cooper, D. C. 1935. Microsporogenesis and the development of the male
gametes in Portulaca oleracea. Amer. Jour. Bot. 22: 453-459. . 1936. Development of the male gametes in Lilium. Bot. Gaz. 98:
389-177.
176 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
Cooper, G. 0. 1935. Cytological studies in the Chenopodiaceae. I. Micro- sporogenesis and pollen development. Bot. Gaz. 97: 169-178.
Dahlgren, K. V. O. 1915. liber die Uberwinterungsstadien der Pollensacke und der Samenanlagen bei einigen Angiospermen. Svensk Bot. Tidskr. 9: 1-12.
. 1916. Zytologische und embryologische Studien liber die Reihen Primu-
lales und Plumbaginales. K. Svenska Vet.-Akad. Handl. 56(4): 1-80.
D'Amato, F. 1947. Nuove ricerche embriologiche e cariologiche sul genere Euphorbia. Nuovo Gior. Bot. Ital. N.S. 53: 405-436.
Darlington, C. D. 1936. The analysis of chromosome movements. I. Podo- phyllum versipella. Cytologia 7: 242-247.
De Mol, W. E. 1923. Duplication of generative nuclei by means of physiological stimuli and its significance. Genetica 5: 225-272.
Duggar, B. M. 1900. Studies in the development of the pollen grain in Symplo- carpus foetidus and Peltandra undulata. Bot. Gaz. 29: 81-98.
Eigsti, O. J. 1939. The morphology of pollen tubes in angiosperms. Proc. Okla. Acad. Sci. 19: 105-107.
. 1940. The effects of colchicine upon the division of the generative cell
in Polygonatum, Tradescantia and Lilium. Amer. Jour. Bot. 27: 512-524. 1941. The occurrence of a pollen tube with four sperms and two tube
nuclei in Polygonatum. Proc. Okla. Acad. Sci. 21: 134-136. Fedortschuk, W. 1931. Embryologische Untersuchung von Cuscuta monogyna
Vahl. und Cuscuta epithymum L. Planta 14: 94-111. Finn, W. W. 1925. Male cells in angiosperms. I. Spermatogenesis and fertil- ization in Asclepias cornutii. Bot. Gaz. 80: 1-25. . 1928. Spermazellen bei Vinca minor und V. herbacea. Ber. deutsch.
bot. Gesell. 46: 235-246. . 1935. Einige Bemerkungen iiber den mannlichen Gametophyten der
Angiospermen. Ber. deutsch. bot. Gesell. 53: 679-686. . 1937a. Entwicklungsgeschwindigkeit des mannlichen Gametophyten
bei den Angiospermen. Trav. Inst. Rech. Sci. Biol., Univ. Kiew 1937, pp.
71-86. . 1937o. Vergleichende Embryologie und Karyologie einiger Cuscuta-
Arten. Jour. Inst. Bot. Acad. Sci. Ukraine 12: 83-99. . 1939. On the history of the development of the male gametophyte in
Labiatae. Jour. Inst. Bot. Acad. Sci. Ukraine 20: 77-96. . 1940. Spermazellen bei Angiospermen. Jour. Bot. de l'U.R.S.S. 25:
155-175.
1941. Mannlichen Gameten bei Angiospermen. Compt. Rend. (Dok.)
Acad, des Sci. U.R.S.S. 30: 451-458. Friemann, W. 1910. "Uber die Entwicklung der generativen Zelle im Pollenkorn
der monocotylen Pflanzen." Diss. Bonn. Frisendahl, A. 1912. Cytologische und Entwicklungsgeschichtliche Studien an
Myricaria germanica Desv. K. Svenska Vet.-Akad. Handl. 48(7): 1-62. Fuchs, A. 1936. Untersuchungen iiber den mannlichen Gametophyten von
Elaeagnus anguMifolia. Osterr. bot. Ztschr. 85: 1-16.
THE MALE GAMETOPHYTE 177
Gager, C. S. 1902. The development of the pollinium and sperm cells in Asclepias
cornutii Decaine. Ann. Bot. 16: 123-148. Geitler, L. 1935. Beobachtungen iiber die erste Teilung im Pollenkorn der
Angiospermen. Planta 24: 361-386. . 1941. Embryosacke aus Pollenkornern bei Ornithogalum. Ber. deutsch.
bot. Gesell. 59: 419-423. . 1942. Uber die Struktur des generativen Kerns im zweikernigen Angio-
spermenpollen. Planta 32: 187-195. Goebel, K. 1933. "Organographie der Pflanzen. III. Samenpflanzen." 3d ed.
Jena. Goodspeed, T. H. 1947. Maturation of the gametes and fertilization in Nico-
tiana. Madrono 9: 110-120. Guignard, L. 1889. Etudes sur les phenomenes morphologiques de la teconda-
tion. Bui. Soc. Bot. de France 36: 100-146.
. 1922. Sur l'existence de corps proteiques dans le pollen de diverses
Angiospermes. Compt. Rend. Acad. Sci. Paris 175: 1015-1020. Guilliermond, A. 1941. "The Cytoplasm of the Plant Cell." Chronica Botanica
Company. Heitz, E., and RSsende, F. 1936. Zur Methodik der Pollenkorn- und Pollen-
schlauch-untersuchung. Bol. Soc. Broteriana, Coimbra II, 11: 5-15. Hewitt, W. C. 1939. Seed development of Lobelia amoena. Jour. Elisha Mitchel
Sci. Soc. 55: 63-82. Ishikawa, M. 1918. Studies on the embryo sac and fertilization in Oenothera.
Ann. Bot. 32: 279-317. Islam, A. S. 1950. The embryology of Ottelia alismoides Pers. Jour. Indian
Bot. Soc. 28: 79-91. Johnston, G. W. 1941. Cytological studies of male gamete formation in certain
angiosperms. Amer. Jour. Bot. 28: 306-319. Juel, H. O. 1900. Beitrage zur Kenntnis der Tetradenteilung. Jahrb. f. wiss.
Bot. 35: 626-649. Juliano, J. B., and Alcala, P. E. 1933. Floral morphology of Musa errans (Blanco)
Teodoro var. Botoan Teodoro. Philippine Agr. 22: 91-126. Junell, S. 1931. Die Entwicklungsgeschichte von Circaeaster agrestis. Svensk
Bot. Tidskr. 40: 111-126. Kausik, S. B., and Rao, P. V. K. 1942. The male gametophyte of Halophila
ovata Gaudich. Jour. Mysore Univ. B. 3: 43-49. Koernicke, M. 1906. Zentrosomen bei Angiospermen? Zugleich ein Beitrag
zur Kenntnis der generativen Elemente im Pollenschlauch. Flora 96: 501-522. Kostriukova, K. 1939a. Spermatogenesis in Crinum hildebrandtii Vatke. Ob- servations in vivo. Jour. Inst. Bot. Acad. Sci. Ukraine 21/22: 157-164. . 19396. Observations in vivo on the formation of the male sex cells in
Lilium martagon L. Compt. Rend. (Dok.) Acad, des Sci. U.R.S.S. 22: 444-447. and Benetskaia, G. 1939. Spermatogenesis in Narcissus poeticus L.
Observations in vivo. Jour. Bot. de l'U.R.S.S. 24: 209-220. Krupko, S. 1926. Les plastides et le chondriome pendant la gonogen&se dans le Gagea lutea. Acta Soc. Bot. Poloniae 4: 77-86.
ITS INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
Kuhlwein, H. 1937. Zur Physiologie der Pollenkeimung, insbesondere der Frage nach dem Befruchtungsverzug bei Gymnospermen. Beihefte bot. Centbl. 57A.
Lagerberg, T. 1909. Studien liber die Entwicklungsgeschichte und systematische Stellung von Adoxa moschatellina L. K. Svenska Vet.-Akad. Handl. 44(4):
1-86. Locke, J. F. 1936. Microsporogenesis and cytokinesis in Asimina triloba. Bot.
Gaz. 98: 159-168. Luxemburg, A. 1927. Recherches cytologiques sur les grains de pollen chez les
Malvaceaes. Bui. Internatl. Acad. Polonaise Sci. et Lettr.: CI. Sci. Math, et
Nat. B. 1927, pp. 363-396. Madge, M. 1929. Spermatogenesis and fertilisation in the cleistogamous flowers
of Viola odorata var. praecox. Ann. Bot. 43: 545-577. . 1936. Division of the generative cell in Hedychium gardnerianum.
Cellule 45: 171-176. Magnus, W. 1913. Die atypische Embryonalentwicklung der Podostomaceen.
Flora 105: 275-336. Maheshwari, P. 1931. Contribution to the morphology of A Ibizzia lebbek. Jour.
Indian Bot. Soc. 10: 241-264.
. 1949. The male gametophyte of angiosperms. Bot. Rev. 15: 1-75.
and Wulff, H. D. 1937 Recent advances in microtechnic. I. Methods
of studying the male gametophyte of angiosperms. Stain Technol. 12: 61-70. Mohrbutter, C. 1936. Embryologische Studien an Loganiaceen. Planta 26:
64-80. Naithani, S. P. 1937. Chromosome studies in Hyacinthus orientalis L. III.
Reversal of sexual state in the anthers of H. orientalis L. var. Yellow Hammer.
Ann. Bot. 1: 369-377. Nawaschin, S. G. 1909. Uber dasselbstandige Bewegungsvermogen des Sper-
makerns bei einigen Angiospermen. Osterr. bot. Ztschr. 59: 457-467. . 1910. Naheres liber die Bildung der Spermakerne bei Lilium martagon.
Ann. Jard. Bot. Buitenzorg, 3rd Suppl. 2: 871-904.
and Finn, W. W. 1913. Zur Entwicklungsgeschichte der Chalazogamen,
Juglans regia, und Juglans nigra. M&n. Acad. Imp. des Sci. St. Petersburg VIII, 31: 1-59.
N6mec, B. 1898. Uber den Pollen der petaloiden Antheren von Hyacinthus orientalis L. Rozpravy Ceske Akad. Prag II, 7(17).
Newman, I. V. 1934. Studies in the Australian acacias. IV. The life history of Acacia baileyana F.V.M. Part 2. Gametophytes, fertilisation, seed produc- tion and germination, and general conclusion. Proc. Linn. Soc. N. S. Wales 59: 277-313.
Nietsch, H. 1941. Zur systematische Stellung von Cyanastrum. Osterr. bot. Ztschr. 90: 31-52.
Oehler, E. 1927. Entwicklungsgeschichtlich-cytologische Untersuchungen an einigen saprophytischen Gentianaceen. Planta 3: 641-733.
O'Mara, J. 1933. Division of the generative nucleus in the pollen tube of Lilium. Bot. Gaz. 94: 567-578.
THE MALE GAMETOPHYTE 179
Paetow, W. 1931. Embryologische Untersuchungen an Taccaceen, Meliaceen
unci Dilleniaceen. Planta 14: 441-470. Piech, K. 1928. Zytologische Studien an der Gattung Scirpus. Bui. Internatl.
Acad. Polonaise Sci. et Lettr. 1928 (1/2): 1-43. Poddubnaja-Arnoldi, V. A. 1933. Kunstliche Kultur und zytologische Unter-
suchung des Pollenschlauches vom Senecio platanifolius Benth. Planta 19:
299-304.
. 1936. Beobachtungen iiber die Keimung des Pollens einiger Pflanzen
auf kunstlichen Nahrboden. Planta 25: 502-529. Raghavan, T. S., et al. 1939. Division of the generative cell in Impatiens balsa-
mina L. Cytologia 9: 389-392. Rosenberg, 0. 1901. Uber die Pollenbildung von Zostera. Meddel. Stock-
holma Hogskolas Bot. Inst.: 1-21. Ruhland, W., and Wetzel, K. 1924. Der Nachweis von Chloroplasten in den
generativen Zellen von Pollenschliiuchen. Ber. deutsch. bot. Gesell. 42: 3-14. Samuelsson, G. 1913. Studien iiber die Entwicklungsgeschichte einiger Bicornes-
Typen. Svensk Bot. Ticlskr. 7: 97-188. Schaffner, J. H. 1901. A contribution to the life history and cytology of
Erythronium. Bot. Gaz. 31: 369-387. Schnarf, K. 1937. Studien iiber den Bau der Pollenkorner der Angiospermen.
Planta 27: 450-465. . 1939. Variation im Bau des Pollenkornes der Angiospermen. Tab. Biol.
Periodicae 27: 72-89. . 1941. "Vergleichende Zytologie des Geschlechtsapparates des Kormo-
phyten. Monographien zur vergleichenden Zytologie." Berlin. Schiirhoff, P. N. 1921. Uber die Teilung des generativen Kernes vor der Kei- mung des Pollenkorns. Arch. f. Zellforsch. 15: 145-159. Shattuck, C. H. 1905. A morphological study of Ulmus americana. Bot. Gaz.
40: 209-223. Smith, C. M. 1929. Development of Dionaea muscipula. I. Flower and seed.
Bot. Gaz. 87: 508-530. Stout, A. B. 1912. The individuality of the chromosomes and their serial ar- rangements in Carex aquatilis. Arch. f. Zellforsch. 9: 114-140. Stow, I. 1930. Experimental studies on the formation of embryo sac-like giant
pollen grain in the anther of Hyacinthus orientalis. Cytologia 1: 417-439. . 1934. On the female tendencies of the embryo sac -like giant pollen
grain of Hyacinthus orientalis. Cytologia 5: 88-108. Strasburger, E. 1908. Chromosomenzahlen, Plasmastrukturen, Vererbungstrager
und Reduktionsteilung. Jahrb. f. wiss. Bot. 45: 477-570. Suita, N. 1936. Studies on the male gametophyte in Angiosperms. I. Studies
in the degenerating process of pollen tube nucleus, with special reference to
Feulgen's reaction. Bot. and Zool. [Tokyo] 4: 2033-2044. . 1937a. On the mature pollen grains in angiosperms. Bot. Mag. [Tokyo]
51: 524-530. . 19376. Studies on the male gametophyte in angiosperms. II. Differ-
180 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
entiation and behavior of the vegetative and generative elements in the pollen
grains of Crinum. Cytologia, Fujii Jubl. Vol., pp. 920-933. Swamy, B. G. L. 1941. The development of the male gametes in Cymbidium
bicolor Lindl. Proc. Indian Acad. Sci. Sect. B. 14: 454-460. . 1949. Embryological studies in the Orchidaceae. I. Gametophytes.
Amer. Midland Nat. 41: 184-201. Tanaka, N. 1940. Chromosome studies in Cyperaceae. VI. Pollen develop- ment and additional evidence for the compound chromosome in Scirpus la-
custris L. Cytologia 10: 348-362. . 1941. Chromosome studies in Cyperaceae. XII. Pollen development
in five genera, with special reference to Rhyncospora. Bot. Mag. [Tokyo] 55:
55-65. Tischler, G. 1917. Pollenbiologische Studien. Ztschr. f. Bot. 9: 417-488. Trankowsky, D. A. 1931. Zytologische Beobachtungen liber die Entwicklung
der Pollenschlauche einiger Angiospermen. Planta 12: 1-18. Tschnernojarow, M. 1915. Les nouvelles donnees dans l'embryologie du Myo-
surus minimus L. M6m. Soc. Nat. Kiew 24: 95-170. Upcott, M. 1936. The mechanism of mitosis in the pollen-tube of Tulipa. Proc.
Roy. Soc. London, Ser. B. 121: 207-220. Wefelscheid, G. 1911. "Uber die Entwicklung der generativen Zelle im Pollenkorn
der dikotylen Angiospermen." Diss.1 Bonn. Weinzieher, S. 1914. Beitrage zur Entwicklungsgeschichte von Xyris indica
L. Flora 106: 393-432. Welsford, E. J. 1914. The genesis of the male nuclei in Lilium. Ann. Bot. 28:
265-270. West, G. 1930. Cleistogamy in Viola riviniana, with especial reference to the
cytological aspects. Ann. Bot. 44: 87-109. Witmer, S. W. 1937. Morphology and cytology, of Vallisneria spiralis L. Amer.
Midland Nat. 18: 309-327. Wulff, H. D. 1933. Beitrage zur Kenntnis des mannlichen Gametophyten der
Angiospermen. Planta 21: 12-50. . 1934. Untersuchungen an Pollenkornern und Pollenschlauchen von
Impatiens parviflora. Ber. deutsch. bot. Gesell. 52: 43-47. . 1939a. Die Pollenentwicklung der Juncaceen. Jahrb. f. wiss. Bot. 87:
533-556. . 19396. Die Entwicklung der Pollenkorner von Triglochin palustris L.
und die verschiedenen Typen der Pollenkornentwicklung der Angiospermen.
Jahrb. f. wiss. Bot. 88: 141-168. and Maheshwari, P. 1938. The male gametophyte of angiosperms (a
critical review). Jour. Indian Bot. Soc. 17: 117-140.
and Raghavan, T. S. 1937. Beobachtungen an Pollenschlauchkulturen
von der Hydrophyllaceae Nemophila insignis. Planta 27: 466-473. Wunderlich, R. 1937. Zur vergleichenden Embryologie der Liliaceae-Scilloideae.
Flora 32: 48-90. Wylie, R. B. 1904. The morphology of Elodea canadensis. Bot. Gaz. 37: 1-22. . 1923. Sperms of Vallisneria spiralis. Bot. Gaz. 75: 191-202.
CHAPTER 6 FERTILIZATION
In gymnosperms the pollen grains usually land directly on the nucellus, while in angiosperms they are deposited on the stigma. There are various agencies which serve to bring about this transfer of pollen from the anthers to the stigma, but since this is primarily an ecological subject and information on it is readily available else- where, it need not be dealt with here. It is sufficient to say that, in the condition in which they are discharged from the anther, the pollen grains show considerable resistance to environmental changes. Sometimes they retain their viability for several weeks, and with proper methods of storage this period can be prolonged still further (see Chap. 12).
Germination of Pollen. Exact information on the time taken by pollen to germinate on the stigma is available for only a few plants, but the following examples will illustrate the range that has been observed: 2 days in Garrya elliptica (Hallock, 1930), 3 hours in Reseda spp. (Eigsti, 1937) ; 2 hours in Beta vulgaris (Artschwager and Starrett, 1933); and 5 minutes in Taraxacum kok-saghys (Poddub- naja-Arnoldi and Dianowa, 1934), Zea mays (Randolph, 1936), and Hordeum distichon (Pope, 1937). In Saccharum oflicinarum (Artschwager et al, 1929) and Sorghum vulgare (Artschwager and McGuire, 1949) germination takes place almost immediately.
The first step in germination is the expansion of the pollen grain by the absorption of liquid from the moist surface of the stigma and the protrusion of the intine through a germ pore. The small tubular structure which arises in this way then continues to elongate, mak- ing its way down the tissues of the stigma and style. Only the distal part of the tube has living cytoplasm, and as the nuclei pass forward callose plugs are left in the empty portions behind them.
Most pollen grains are monosiphonous, i.e., only a single pollen tube emerges from each pollen grain; others, like those of the Mal- vaceae, Cucurbitaceae and Campanulaceae, are polysiphonous. In Althaea rosea 10 tubes, and in Malva neglecta even 14 tubes, are
181
182 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
known to come out from the same pollen grain (Stenar, 1925). Eventually, however, only one of them makes further progress. Sometimes the same pollen tube may divide into one or more branches. Such a condition seems to be frequent in the Amenti- ferae, where the branching tubes give the appearance of a ramifying fungous mycelium (see Finn, 19286). In plants whose pollen grains are united into tetrads or into pollinia, several pollen tubes are produced at the same time (Fig. 107 A, B).
A B C
Fig. 107. Acacia baileyana, pollinium germinating on stigma (ct = cuticle; g = generative cell ; In = tube nucleus; in = intine). (After Newman, 1984.) B, Cymbi- dium bicolor, germination of pollen grains united in tetrads. (After Swamy, 1941.) C, Elatine triandra, t.s. anther of cleistogamous flower, showing pollen grains ger- minating in situ. (After Frisendahl, 1927.)
The stigma is believed to play an important part in the germina- tion of pollen, but in many plants germination can also be induced in a sugar solution of appropriate strength. Martin (1913) germi- nated the pollen of Trifolium pratense on hog's bladder moistened with distilled water and suggested that the only use of the stigma lies in controlling the water supply. Katz (1926) agreed with this view and said that the chief function of the stigmatic secretion is to protect the pollen as well as the stigma from desiccation. In her experiments the pollen germinated even on the cut surface of the style, provided the stigmatic secretion was applied to the stump and the latter was kept moist for some time.
FERTILIZATION 183
Pollen grains may also germinate on other parts of the flower be- sides the stigma. In cleistogamous flowers (Frisendahl, 1927; Madge, 1929; West, 1930; Maheshwari and Singh, 1934) germina- tion takes place within the anther loculi (Fig. 107C), and in Aeginetia indica (Juliano, 1935) even on the moist surface of the corolla tube. Frequently pollen grains germinate on a foreign stigma, i.e., stigma of a different species (see Eigsti, 1937; Sanz, 1945). If fertilization takes place, it results in the formation of interspecific and inter- generic hybrids.
Course of Pollen Tube. After the tube has emerged from the pollen grain, it makes its way between the stigmatic papillae into the tissues of the style. The latter is extremely variable in length. In some plants it is so short that the stigma is described as sessile, while in Zea mays the so-called "silk" may attain a length of 50 cm.
Depending on the presence or absence of the transmitting tissue and on the extent of its development, styles have been classified into three main types called open, half -closed, and closed (Hanf, 1935). In the first type there is a wide stylar canal and the inner epidermis itself assumes the function of the nutrition and conduction of the pollen tube, as in the Papaveraceae, Aristolochiaceae, Ericaceae, and many monocotyledons. In the second type the canal is sur- rounded by a rudimentary transmitting tissue of two or three layers of glandular cells, as in several members of the Cactaceae. In the third or closed type, illustrated by Datura and Gossypium, there is no open channel but instead a solid core of elongated and richly protoplasmic cells through which the pollen tube grows downward in order to reach the ovary. Finally, there are other plants like Salix, Acacia, and many grasses in which the styles are solid but are not provided with any specialized transmitting tissue.
In open styles the pollen tube grows on the surface of the cells lining the stylar canal (often in the mucilage secreted by them); and in solid styles through the intercellular spaces between the cells of the transmitting tissue, enlarging the spaces by the hydrostatic pressure of its contents and secreting some enzymatic substances which bring about a dissolution of the middle lamellae. Only rarely does the pollen tube pass through the cells themselves.
After arriving at the top of the ovary, the tube may enter the ovule either through the micropyle or by some other route. The former is the usual condition and is known as porogamy, but even
184
INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
in plants ordinarily classed as porogamous there are several modi- fications. To mention a few examples, in Acacia (Newman, 1934) the integuments are still below the apex of the nucellus at the time of fertilization so that a micropyle does not exist at this stage and in Philadelphus , Utricularia, Vandellia, and Torenia the embryo sac
Fig. 108. Course of pollen tube in ovule of Casuarina equisetifolia. A, optical longitudinal section of entire ovary; pollen tube is represented by heavy black line. B, l.s. ovule reconstructed from several sections to show path of pollen tube. (After Swamy, 1948.)
FERTILIZATION
185
protrudes out of the micropyle so that the pollen tube comes in direct contact with it. In several members of the Loranthaceae there is no integument and therefore nothing that can be called a micropyle. Here the embryo sacs undergo a remarkable elongation and meet the pollen tubes at some point in the stylar region (see also p. 143).
In some plants the pollen tube enters the ovule through the chalaza. This condition, known as chalazogamy (see page 17), was first reported in Casuarina (Treub, 1891) and soon afterwards
ABC
Fig. 109. Development of obturator in Acalypha indica. A, l.s. terminal flower of inflorescence, showing ovule at megaspore mother cell stage. Note beginning of formation of obturator (shaded). B, l.s. lateral flower at more advanced stage. C, ovule, enlarged to show hood-like obturator fitting over nucellus. (After Maheshwari and Johri, 1941.)
in several members of the Amentiferae. Nevertheless, it is not confined to them, being also known in Rhus (Grimm, 1912), Cir- caeaster (Junell, 1931), and a few other genera. Recent studies have, however, shown that even in such cases, where entry into the ovule is effected through the chalaza, the tube usually continues its growth over the surface of the embryo sac and penetrates it only after arriving near the egg apparatus. As examples may be men- tioned Ostrya carpinifolia (Finn, 1936), Juglans regia (Nast, 1941), and Casuarina equisetifolia (Swamy, 1948) (Fig. 108).
In Alchemilla (Murbeck, 1901), Cucurbita (Longo, 1901; Kirk-
186 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
wood, 1906), and Circaeaster (Junell, 1931) the pollen tube enters through the funiculus or the integument. This is known as mes-
ogamy.
Formerly considerable phylogenetic significance was attached to the route taken by the pollen tube during its entry into the ovule, but now this point is considered to be of physiological rather than phylogenetic importance, for we sometimes find con- siderable variation in this respect even in one and the same species. In Brassica oleracea (Thompson, 1933) the tube normally enters
through the micropyle, but some- times it may do so by way of the chalaza. In Ulmus, Shat- tuck (1905) speaks of its branch- ing and apparently aimless wand- ering through the funiculus, the integuments, and occasionally the nucellus. In Ep Hob turn (Werner, 1914; Tackholm, 1915) it may enter either through the micropyle, through the integu- ments, or by an intermediate route. In Boerhaavia (Mahesh- wari, 1929), although the tube actually enters through the mi- cropyle, it first makes a horizon- tal crossing through the funicu- lus. In Gossypium (Gore, 1932) it often passes from the funiculus to the base of the ovule and then travels up along the wall of the latter to enter the micropyle.
An organ of special signifi- cance in facilitating the entry of the pollen tube into the ovule is the so-called obturator, to which reference had already been made by Hofmeister in the year 1849.
Fig. 110. Origin and structure of obtu- rator in Myriocarpa longipes (A), Leu- cosyke capitellata (B), and Quisqualis
indica (C, D). (After Fagerlind 19U, Usually it is a swelling of the 1941.) placenta which grows towards
FERTILIZATION
187
the micropyle and fits like a hood or canopy over the nucellus, serving as a sort of bridge for the pollen tube (Fig. 109). Often the cells of the obturator may be greatly elongated or may have a glandular appearance (Fig. HOC, D).
Some other structures having a different origin but serving the same function may also be included under the general term ob- turator. In the Thymelaeaceae (Fuchs, 1938) the cells belonging to the base of the stylar canal elongate and grow down as hairy
Fig. 111. Development of obturator in some members of Thymelaeaceae. A, Daphne laureola, l.s. pistil. B, same, part of ovary with cells of obturator pro- truding downward into micropyle. C, more advanced stage, showing path of pollen tube. D, Passerina pectinata, l.s. part of ovary, showing obturator (mi = micropyle). (After Fuchs, 19SS.)
processes approaching the nucellus (Fig. 111). In Pilea (Fagerlind, 1944) a tuft of papillate cells extends from the base of the style to the apex of the ovule, coming in intimate contact with the latter. In Myriocarpa and Leucosyke (Fagerlind, 1944), on the other hand, it is the cells of the inner integument which elongate upward and penetrate into the stylar canal (Fig. 110A, B), forming what may be called an integumentary obturator.
Usually there are no special modifications in the cells lining the micropylar canal, but sometimes, as in Berkheya (Gelin, 1936),
188 INTRODUCTION TO EMBRYOLOGY OF ANGIOSPERMS
Grevillea (Brough, 1933), and Cynomorium (Steindl, 1945), they become mucilaginous or glandular and seem to contribute to the nutrition of the pollen tube. In Cardiospermum (Kadry, 1946) not only the cells belonging to the inner integument but also those forming the apical portion of the nucellus give rise to a mucilaginous mass which facilitates the entry of the pollen tube. In plants with a many-layered nucellar tissue, like Beta (Artschwager and Starrett, 1933), those of its cells which are in continuity with the micropyle become elongated and richly protoplasmic and give an impression as though they were designed to lead the pollen tube through the path of least resistance.
It is of interest to note that even during its passage through the nucellus the pollen tube usua