UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Comparative developmental studies of the floret and embryo sac in five species of Oryzopsis (Gramineae) Kam, Yew Kiew 1973

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

Item Metadata

Download

Media
831-UBC_1973_A1 K34.pdf [ 8.3MB ]
Metadata
JSON: 831-1.0101004.json
JSON-LD: 831-1.0101004-ld.json
RDF/XML (Pretty): 831-1.0101004-rdf.xml
RDF/JSON: 831-1.0101004-rdf.json
Turtle: 831-1.0101004-turtle.txt
N-Triples: 831-1.0101004-rdf-ntriples.txt
Original Record: 831-1.0101004-source.json
Full Text
831-1.0101004-fulltext.txt
Citation
831-1.0101004.ris

Full Text

nm j COMPARATIVE DEVELOPMENTAL STUDIES OF THE FLORET AND EMBRYO SAC IN FIVE SPECIES OF ORYZOPSIS (GRAMINEAE) by RAM YEE KIEW B.Sc. (Hons.)«» University of Malaya, Malaysia, 1967 M.Sc, University of Malaya, Malaysia, 1969 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY Iii the Department of BOTANY We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA May, 1973 In presenting t h i s thesis i n p a r t i a l f ulfilment of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t freely available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of Botany The University of B r i t i s h Columbia Vancouver 8, Canada Date July 5. 1973 i i ABSTRACT -Development of the floret and embryo sac of Oryzopsis virescens and 0. hymenoides was studied. Evidence from this study and from other studies on grass floret and embryo sac development has brought the following interpretations. Histogenesis of the lemma, palea, posterior lodicule and the gynoecial wall is similar, and indicates their foliar nature. They are determinate organs, have a shallow site of initiation, and exhibit marginal growth. The anterior lodicules differ from them in having a deeper initiation site. The interpretation of the anterior and posterior lodicules as reduced perianth structures of one whorl rather than as structures 'de novo' is preferred. The callus is formed by the downward projection of the base of the lemma. Developmentally, the stamens are stem-like. The gynoecium consists of a unit ascidiform gynoecial wall surrounding a terminal ovule. There are two styles, each of which develops from the lateral portions of the gynoecial wall. The floret apex is not used up in the formation of the gynoecial wall. The residual floret apex develops into the ovule. The grass gynoecium may be considered acarpellate. The ovule is hemianatropous, bitegmic and pseudocrassinucellate. The micropyle is delimited by the inner integument. Embryo sac development is of the monosporic, 8-nucleate type. The antipodals are proliferated. The development of the floret and embryo sac of three other species of Oryzopsis was also studied. They are, namely, 0. micrantha, 0. kingii, and 0. asperifolia. Developmental features of a l l five species of Oryzopsis were compared with developmental features of i i i Oryzopsis miliacea, and of four species of Stipa, a closely related genus. These are, namely, S_. lemmoni, £5. hendersoni, S_. t o r t i l i s and S_. richardsoni. Cytotaxonomic studies by Johnson (1945. Bot. Gaz. 107: 1-31) in the genus Oryzopsis indicate that 0. virescens (n = 12) and 0. miliacea (n = 12) are members of the Old World section Piptatherum; 0. micrantha (n = 11), 0. k i n g i i (n = 11), and 0. asperifolia (n = 23), belong to the New World section Oryzopsis; 0. hymenoides (n = 24) belongs to the New World section Eriocoma. Intergradation of the genera Oryzopsis and Stipa occurs in North America in the sections Oryzopsis and Eriocoma. Qryzopsis micrantha resembles 0. miliacea in certain morphological features, while 0. k i n g i i i s a 'borderline' Oryzopsis-Stipa species. Oryzopsis hymenoides is known to hybridise with eleven species of Stipa. Thirty-one characters were abstracted from the developmental data and were analyzed s t a t i s t i c a l l y . The results indicate that 0. virescens is set apart from the five other species of Oryzopsis and the four species of Stipa. The a f f i n i t y of 0. hymenoides on the basis of development is with Stipa. This further supports data from mor-phology, distribution and hybridization studies and suggests that Oryzopsis hymenoides belongs to the genus Stipa. There does not appear to be a discontinuous variation in development between 0. miliacea, 0. micrantha, 0. asperifolia, 0. k i n g i i , £3. richardsoni, 0. hymenoides  (Stipa) , 0. hendersoni, S_. lemmoni and S_. t o r t i l i s . It would seem that more comprehensive studies of the genus Oryzopsis w i l l either lead to i t s mergence with Stipa or at least to a redefinition of the sections of Oryzopsis. iv TABLE OF CONTENTS Page ABSTRACT i i LIST OF TABLES v i LIST OF FIGURES v i i ACKNOWLEDGEMENTS x GENERAL INTRODUCTION 1 An evaluation of ontogenetic studies 1 A brief botanical history of Oryzopsis . 4 PART I 8 Introduction 8 Materials and methods 10 Observations (a) . Oryzopsis virescens 12 (b). Oryzopsis hymenoides 27 Discussion. 36 Conclusion 54 PART II 55 Introduction 55 Materials and methods..... 58 Observations (a). Oryzopsis micrantha 59 (b) . Oryzopsis k i n g i i 65 (c) . Oryzopsis asperif o l i a 71 Discussion 77 Conclusion 89 LITERATURE CITED 90 APPENDIX FIGURES v i LIST OF TABLES Table Page I. Reference table for comparison of stages in awn-lemma and callus development in Oryzopsis  virescens and 0. hymenoides, using the stage of gynoecium development as a marker 31 II. Reference table for comparison of stages in awn-lemma and callus development in Oryzopsis  virescens, 0. hymenoides, 0. micrantha, 0. kingii and 0. asperifolia, using the stage of gynoecium development as a marker... 75 III. Comparison of developmental features of Oryzopsis  virescens, 0. hymenoides, 0. micrantha, 0. kingii 0. asperifolia, 0. miliacea, Stipa lemmoni, S_. hendersoni, S_. tortilis and S_. richardsoni. 81 IV. Similarity matrix ( % ) of Oryzopsis virescens, 0. hymenoides, 0. micrantha, 0. kingii, 0. asperifolia, 0. miliacea, Stipa lemmoni, S. hendersoni, S. tortilis and S_. richardsoni 82 v i i LIST OF FIGURES Figure - Page 1 - 6 . Floral parts of Oryzopsis virescens and 0. hymenoides 102 7. Diagramatic transverse section of grass floret 11 8 - 14. Stages in floret development i n 0. virescens. 104 15 r,r20. Stages in floret development in 0. virescens 106 21 - 25. Stages in awn-lemma development in 0. virescens 108 26 - 30. Stages in callus development i n 0. virescens 110 31 - 39. Stages in floret development i n 0. virescens 112 40 - 56. Stages i n stamen and anterior lodicule development in 0. virescens. 114 57 - 72. Stages i n gynoecium development in 0. virescens 116 73 - 77. Stages in gynoecium development in 0. virescens 118 78 - 84. Stages in ovule and embryo sac development in 0. virescens 120 85 - 90. Stages in ovule and embryo sac development in 0. virescens 122 91 - 98. Stages in embryo sac development in 0. virescens 124 99 - 105. Stages in floret development in 0. hymenoides 126 106 - 112. Stages in awn-lemma development in 0. hymenoides 128 113 - 119. Stages in callus development in 0. hymenoides 130 120 - 128. Stages in floret development in 0. hymenoides 132 129 - 134. Stages in floret development in 0. hymenoides 134 v i i i 135 - 140. Ovule and early embryo sac development in 0, hymenoides. 136 141 - 143. Stages in embryo sac development in 0. hymenoides.... 138 144 - 146. Fertilization and early post-fertilization stages in 0. hymenoides 140 147 - 154. Floral parts of 0. micrantha, 0. kingii and 0. asperifolia... 142 155 - 162. Stages in floret development in 0. micrantha 144 163 - 167. Stages in awn-lemma development in 0. micrantha 146 168 - 173. Stages in callus development in 0. micrantha 148 174 - 185. Stages in floret development in 0. micrantha 150 186 - 191. Ovule and embryo sac development in 0. micrantha 152 192 - 196. Stages in embryo sac development in 0. micrantha 154 197 - 204. Stages in floret development in 0. kingii 156 205 - 211. Stages in awn-lemma development in 0. kingii 158 212 - 218. Stages in callus development in 0. kingii. 160 219 - 231. Stages in floret development in 0. kingii 162 232 - 239. Ovule and embryo sac development in 0. kingii. 164 240 - 244. Stages in embryo sac development in 0. kingii 166 245 - 251. Stages in floret development in 0. asperifolia. 168 252 - 257. Stages in awn-lemma development in 0. asperifolia.... 170 258 - 265. Stages in callus development in 0. asperifolia....... 172 266 - 279. Stages in floret development in 0. asperifolia....... 174 280 - 284. Ovule and early embryo sac development in 0. asperifolia 176 ix 285 - 288. Stages in embryo sac development and fertilization in 0. asperifolia 178 289. Inter-relationships of 0. virescens, 0. hymenoides, J2« micrantha. 0. kingii, 0. asperifolia, 0. miliacea, Stipa lemmoni, S_. hendersoni, S_. tortilis and £>. richardsoni 85 290. Scanning electronmicrograph of young florets of 0. virescens 180 291. Scanning electronmicrograph of young florets of 0. virescens 180 292. Scanning electronmicrograph of young florets of 0. virescens 182 293. Sagittal section of floret of 0. virescens 184 294. Transverse section of top of ovary in 0. hymenoides.. 186 295. Frontal longitudinal section of ovary of 0. hymenoides. 186 ACKNOWLEDGEMENTS This research was carried out during the tenure of a Canadian Commonwealth Scholarship at The University of British Columbia for the years 1970-1973. It is with pleasure that the author expresses her appreciation to Dr. Jack Maze for guidance and counsel during the course of this work, and the use of his library. 1 GENERAL INTRODUCTION This thesis consists essentially of two parts. Part I is an ontogenetic study of the floret and embryo sac of Oryzopsis virescens and 0. hymenoides, as such data would lead to a better understanding of the floret of Oryzopsis in particular, and of the Gramineae in general. Part II is an attempt to elucidate the relationships of 0. virescens, 0. hymenoides, 0. micrantha, 0. kingii and 0. asperifolia on the basis of data from developmental studies of the floret and embryo sac. It is hoped that developmental data can be used effectively as an adjunct to other characteristics in working out taxonomic problems. An evaluation of ontogenetic studies It is generally accepted by most taxonomists that taxa are properly established on the basis of multiple correlations of characters (Cronquist, 1968). Davis and Heywood (1965) , stressed that a l l stages of development of the plant should be examined for possible characters. Yet ontogenetic features have not received much attention from taxonomists- Embryologi-cal studies have been employed, successfully, to solve certain taxonomic problems (Maheshwari, 1950, 1961). Recent work by Mehlenbacher (1970), and Maze and co-workers (1971) , has shown that comparative studies on floret development in Oryzopsis and Stipa are useful in understanding the relationships of these two genera. A few reasons could be offered for this lack of popularity of applied ontogenetic studies. Results from ontogenetic investigations have a history of not bearing out the popular existing dogma of the time. This 2 i s especially true i n the interpretation of the flower. This branch of research was set on a controversial course with the pronouncements of Schleiden in 1839 and Payer in 1857 (both cited by Mbelir5.no, 1970), that ovules were not borne on carpels but were axis-borne. This was of course contrary to the prevalent carpel theory of that time. The fact that Schleiden was later shown to be wrong in his interpretation of f e r t i l i z a -tion did not help to foster faith in the ontogenetic method. Nonetheless, ontogenetic research has continued to be one of the chief lines along which attempts are made to solve the problem of organisation of angiosperm reproductive structures. Even with today's improved techniques, no agree-ment has been reached regarding the interpretation of the stamens and the carpels. The interpretation of the stamen on the basis of histogenetic analysis has resulted i n diametrically opposed conclusions. Satina and Blakeslee (1943), Barnard (1957a, b.), and Sharman (1960b) concluded that the stamens they studied were cauline, while Kaussman (1941), McCoy (1940), Boke (1947, 1949), and Kaplan (1968) hold that their investigations point to a f o l i a r interpretation for stamens. Likewise, in the Cyperaceae the ovules are phyllosporous according to Schultze-Motel (1959), and cauline according to Barnard (1957b). In Datura stramonium, Satina and Blakeslee, (1943) concluded that the ovules are axis-borne, but Verzar-Petri and Baranyai Szentpetery (1960) stated that the ovules are carpellary. It would appear, therefore, that the ontogenetic method cannot be used with any measure of r e l i a b i l i t y i n studies on descriptive f l o r a l morphology, and much less so in comparative studies. It is the writer's contention that such a view-point is erroneous. The place of ontogeny in f l o r a l enquiry has been staunchly defended by Thompson (1937). One of the 3 causes for a l l these contradictions is that most workers have not analysed their results s t r i c t l y on the basis of the histogenetic data they have obtained. An example is seen in Schultze-Motel 1s paper (1959). In his diagrams (Abb. 7, lOd, 12, 17, 18, 20) the ovules are formed directly from apical cells and are therefore terminal, but he interprets them to be phyllosporous. He adheres to the concept of the 'Entwicklungsfeld des Karpells' of Goebel (or 1 scheitelbedeckendes Feld 1 of Eckardt, 1957), as shown by the following excerpts "Die Primordien der Karpelle treten als 'scheitelbedeckendes Feld' (Eckardt, 1957, S. 76) auf. Die Samenanlage (o in Abb. 7) entwickelt sich aus diesem Feld, is also phyllospor". The description of his results is dictated by the use of certain interpretive terms. As long as terms are used in such a way, the discussion about phyllosporous- versus axis- borne ovules becomes a travesty (Fagerlind, 1958). Another example is illustrated by the studies of Roth (1959), and Pankow (1959), on the Primula placenta. Using the same method (micro-technique) , they obtained almost identical results, which were interpreted by them in diametrically opposed directions. Pankow, on the basis of his histogenetic results, interpreted the placenta as a continuation of of the f l o r a l apex. Roth, however, by a series of 'reasoned* deductions concluded that the placenta is of carpellary tissue. The value of developmental studies has been stated cogently by Maze and co-workers (1971, 1972). However, one must guard against an over-emphasis of the results obtained by developmental studies. In this connection, i t is well to bear in mind the opinions of von Guttenberg (1960), and Rohweder (1963), on the significance and limitations of histogenetic methods. 4 A brief botanical history of Oryzopsis The genus Oryzopsis was established in 1803 by Michaux, based on a single New World species, Oryzopsis asperifolia. The genus has a very complicated taxonomic history, an excellent account of which was given by Johnson (1945a). Only a brief resume of the taxonomic relationships in the genus is given here. As presently understood, the genus Oryzopsis consists of approximately 1 6 species, occurring i n the cool and temperate regions of both the eastern and western hemispheres. The species are placed in three main morphologi-cal groups, each a taxonomic section (Hackel, 1889; Johnson, 1945a). The taxonomic arrangement is given below: Section Piptatherum 0. miliacea (L.) Benth. and Hook. 0. virescens (Trin.) Beck. 0. paradoxa (L.) Nutt. 0. coerulescens (Desf.) Hack. 0. holciformis (Biel.) Hack. 0. racemosa (J.E.Smith) Ricker Section Oryzopsis 0. micrantha (Trin. and Rupr.) Thurb. 0. pungens (Torr. ) Hitchc. 0. canadensis (Poir.) Torr. 0. exigua Thurb. 0. k i n g i i (Boland.) Beal 0. asperifolia Michx. 0. swallenii Hitchcock and Spellenberg 5 Section Eriocoma 0. hymenoides (Roem. and Schult.) RIcker 0. contracta (B.L.Johnson) Shechter The Old World section, Piptatherum, consists of 5 diploid species (2n = 24) which range in distribution from southern Europe and Northern Africa to Asiatic Russia. The sixth member, 0. racemosa, is a polyploid (2n = 46) occurring in North America. Species of the New World section Oryzopsis are rather distinct at the diploid level (2n = 22) from those of Piptatherum. The two polyploids in the section are 0. asperifolia (2n «= 46) and 0. swallenii (2n = 34). Eriocoma is a New World section of two polyploids, 0. hymenoides (2n = 48) and 0. contracta (2n = 48). The latter species i s purported to represent a stabilized hybrid line derived from an 0. hymenoides x 0. micrantha cross (Shechter and Johnson, 1968). The genus Oryzopsis is extremely d i f f i c u l t to circumscribe. Consider-able uncertainty s t i l l exists i n delimiting Oryzopsis from Stipa, a closely a l l i e d genus which tends to merge with Oryzopsis through the section Oryzopsis. Characters of the lemma, awn, and callus, are traditionally relied on to distinguish Oryzopsis from Stipa. In Ory zopsis the 1emma is relatively broad, short, and indurate, the callus blunt and the awn deciduous. In Stipa the lemma is long and slender, less indurate, the callus is sharp-pointed and the awn is persistent. Also, in most species of Oryzopsis the inflorescence i s an open panicle, while in Stipa the panicle is contracted. Taken independently, these characters are highly unreliably. Even when considered together they do not effectively 6 distinguish between the two genera, for certain species resemble Oryzopsis in some features, and Stipa in others. It has been the tendency for systematists to assign uncertaind cases to Oryzopsis. A recent example is 0. swallenii (Hitchcock and Spellenberg, 1968). Studies in floret development and embryology in the two genera have been undertaken by Maze et a l . in recent years (1970, 1971, 1972) and are continuing. Their data have proved to be useful in assessing relation-ships of three species of Stipa. Mehlenbacher (1970), after a thorough study of 0. hendersoni Vasey found that this species i s closer to Stipa than to Oryzopsis, and subsequently transferred i t to Stipa (Spellenberg and Mehlenbacher, 1971). The development of the floret and the embryo sac (up to fe r t i l i z a t i o n ) of 5 species of Oryzopsis was studied by the author. The 5 species are, namely, 0. virescens, 0. hymenoides, 0. micrantha, 0. k i n g i i and 0. asperifolia. They were chosen for the following reasons: 0. virescens has the vegetative and f l o r a l features traditionally associated with Oryzopsis (f l a t broad leaves, open panicle, short deciduous awn, plump lemma and blunt callus), and i s considered to be most representative of the genus. Oryzopsis hymenoides is of interest because of i t s resem-blance to species of the section Piptatherum in several features, and because i n nature i t forms hybrids with several species of Stipa, as Johnson demonstrated i n a series of papers (1943, 1945b, 1960, 1962, 1963). It is easily distinguished from the Old World species by i t s pubescent lemma and sharp pointed callus. Oryzopsis micrantha, 0. k i n g i i and 0. asperifolia are members of the section Oryzopsis — a section 7 marked by specialization in characters which merge with the genus Stipa. Oryzopsis micrantha possesses an admixture of Piptatherum and Oryzopsis characters. Its a f f i n i t y to the section Piptatherum is recognized by Elias (1942), who placed i t i n that section. Oryzopsis k i n g i i can best be described as a borderline Stipa-Oryzopsis species. It possesses features which are associated regularly with Stipa (narrow involute leaves, closed panicle, twisted awn, narrow lemma and sharp callus). Its assign-ment to either genus islargely arbitrary. Oryzopsis asperifolia, the type species of the genus, seems to occupy a position, morphologically speaking, between the two extreme ends of the section Oryzopsis, as expressed by 0. micrantha and 0. k i n g i i (see Johnson, 1945a). Morphological and cytological studies so far have not resulted in a satisfactory delineation of the genus Oryzopsis, or a taxonomic arrange-ment that reflects the intricate relationships between species of Oryzopsis, and between Oryzopsis and Stipa. More comprehensive studies of the genus Oryzopsis are needed, and towards that end the author has chosen developmental studies. A thorough study of floret development and embryology of two species, 0. virescens and 0. hymenoides, is attempted — 0. virescens because i t approaches the traditional 'Oryzop-zoid 1 description; 0. hymenoides because of i t s a f f i n i t i e s with section Piptatherum and i t s hybridization with species of Stipa. It is hoped that such a study w i l l lead to a better understanding of the Oryzopsis floret. A logical extension of the above investigation i s to seek charac-ters of a developmental nature for purposes of comparison. This is done for the 5 species mentioned. 8 PART I Developmental studies of the floret and embryos sac of Oryzopsis  virescens and 0. hymenoides. INTRODUCTION In recent years knowledge of the grass floret has increased through contributions from histogenetic studies. Published descriptions of such studies are found in the papers by Cannon (1900), Bonnett (1953, 1961), Barnard (1955, 1957a), Sharman (1960a, 1960b), Klaus (1966), Mehlen-bacher (1970), and Maze et a l . (1970, 1971, 1972). However, there is s t i l l no unanimous interpretation of the grass floret, which remains a controversial topic in plant morphology. The controversy centres mainly on the nature of the lodicules, the stamens and the gynoecium, that i s , whether they are of a phyllome or caulome nature. Because of the imprecision of morphological terminology, i t is imperative that the author c l a r i f i e s her use of terminology, especially a the terms phyllome and caulome. Used in Adescriptive sense, morphologi-cal terms such as leaf, bud, etc., are merely words for structural entities. In a comparative sense the same words signify mutually exclu-sive categories. While It is not wrong to use the word ' l e a f in the sense of a category, the word conjures in most minds the picture of a foliage leaf. An alternative term of a more generalized character — phyllome — is preferred. Likewise, the structure described as a 'bud* is a member of the category caulome. Defined in general terms, a 9 phyllome is an abstract entity to which belong lateral appendages of determinate growth. These lateral appendages are dorsiventral, exhibit marginal growth and have a shallow site of i n i t i a t i o n . The category caulome encompasses structures of an axial nature. These are of indeter-minate growth, usually radial in shape, do not undergo marginal growth, and have a deeper site of i n i t i a t i o n . It is rather fortunate that the vegetative structures of the grasses that have been studied development-al l y lend themselves easily to categorization. The preceding paragraph w i l l undoubtedly unleash a torrent of criticism at the author, identified by 'Croizatian' and 'Sattlerian' tags. Sattler (1966) made a very eloquent and valid plea for a more precise approach to comparative morphology. But his semi-quantitative -not homology concept does^lend i t s e l f readily to application, as he himself admitted. For lack of additional methods, the author has to use the •traditional tools of the trade' — the established vocabulary of comparative morphology is s t i l l the most useful for communication of ideas (at least at the present time). In this part of the thesis, the author reports her studies on the f l o r a l histogenesis in two species of Oryzopsis, 0. virescens and 0. hymenoides. Her present observations on f l o r a l histogenesis w i l l be applied to the interpretation of the Oryzopsis floret in particular, and to a c r i t i c a l discussion of the Gramineae floret in general. 1 0 MATERIALS AND METHODS The f l o r e t of 0. v i r e s c e n s i s short and plump, the awn i s short and deciduous, and the c a l l u s i s b l u n t ( F i g s . 1, 2, 3 ) . Oryzopsis hymenoides i s a wide-ranging grass i n a r i d areas of western North America. In the shape of i t s f l o r e t and the deciduous nature of i t s awn, i t approaches 0. v i r e s c e n s ( F i g s . 4, 5, 6 ) , but i t s sharp c a l l u s and i t s p i l o s e lemma are ch a r a c t e r s found i n S t i p a , a genus c l o s e l y a l l i e d w i t h O r y z o p s i s , (see Johnson, 1945a f o r the i n t r i c a t e r e l a t i o n s h i p s w i t h i n Oryzopsis and between i t and S t i p a ) . M a t e r i a l s used i n t h i s study were c o l l e c t e d from p l a n t s growing i n the Botany Garden at the U n i v e r s i t y of B r i t i s h Columbia. P l a n t s of 0. v i r e s c e n s were grown from seed obtained through the I n t e r n a t i o n a l Seed Exchange; p l a n t s of 0. hymenoides were t r a n s p l a n t s of populations c o l l e c t e d i n Canal F l a t s , B r i t i s h Columbia. F l o r e t s , e n t i r e young i n f l o r e s c e n c e s , or p o r t i o n s of young i n f l o -rescences were f i x e d i n a F o r m a l i n - A c e t i c A c i d - A l c o h o l m i x t u r e , u s i n g 507. e t h y l a l c o h o l (Johansen, 1940). They were dehydrated i n a 2,methoxyethanol - absolute a l c o h o l - n-propanol - n-butanol s e r i e s (Feder and O'Brien, 1968), and embedded i n P a r a p l a s t . F l o r e t s longer than 1 mm. were o r i e n t e d before s e c t i o n i n g . This was done very simply by us i n g a p a i r of f i n e forceps and a s m a l l s p a t u l a . These instruments were heated and used to melt the wax around the f l o r e t . With the a i d of a stereo-microscope the f l o r e t was then moved around and p o s i t i o n e d f o r s e c t i o n i n g e i t h e r i n a f r o n t a l or s a g i t t a l l o n g i t u d i n a l p l ane. Sec t i o n s were cut at 6 - 7 u and s t a i n e d i n a combination of s a f r a n i n , t a n n i c a c i d , orange G and i r o n alum (Sharman, 1943). Drawings were 11 traced from a projection head on a Zeiss Ultraphot II microscope. A l l drawings are of sagittal sections unless otherwise stated, with the front (= anterior) side of the floret to the reader's left. Information from sections of florets in early stages of development was supplemented by scanning electron micrographs. The terms anterior (= front) vs. posterior (= back), and abaxial vs. adaxial are used to describe the orientation of the various parts of the floret. These are used with reference to two axes: the rachilla axis and the floret axis.(Fig. 7). In those spikelets with more than one floret, the lemma is situated away from the rachilla, while the palea is next to the rachilla. That side of the floret with the lemma is said to be the anterior side; the palea side is correspondingly the posterior side. Adaxial and abaxial denote two opposing surfaces of the same lateral structure, with reference to the axis on which i t is borne (in this case the floret axis). For example, the lemma has an adaxial surface (adjacent to the floret axis), and an abaxial surface (away from the floret axis). • —. anterior stamen floret axis palea anterior (= front) lodicule (anterior side of floret) (posterior side of floret) posterior (= back) lodicule rachilla axis FIG. 7. Cross-section of grass floret (diagramatic). 12 OBSERVATIONS Oryzopsis virescens (Trin.) Beck. General floret organogenesis The f i r s t structure to be init i a t e d on the floret apex Is the awn-lemma (Figs. 8, 15). This occurs after the i n i t i a t i o n and early growth of the f i r s t and second glumes. The palea and the stamens appear next (Figs. 9, 18). There are three stamens, one anterior and two l a t e r a l , and unless otherwise stated, the stamen featured in the drawings is the anterior one. The lodicules, two anterior and one posterior, follow closely (Figs. 10, 19). The callus and the gynoecial wall then Initiate, at approximately the same time (Figs. 11, 20). The last f l o r a l structure to develop is the ovule (Fig. 13). The shape of the floret.at the megaspore stage in embryo sac development is shown in Figure 14. Floret apical meristem The floret apical meristem has a tunica-corpus organization. The tunica is one-layered. The outermost corpus cells often seem to form a distinct layer beneath the tunica (Fig. 15). The floret apex is dome-shaped and remains so throughout organ formation, though i t under-goes some displacement in the course of floret development (cf. Figs. 15, 18, 19, 20). Early in floret development, the floret apex i s more or less aligned with the floret axis (Fig. 15). During early growth of the anterior stamen and anterior lodicule the floret apex becomes posteriorly directed (Figs. 10, 19), and continues to be so oriented during early growth of the gynoecial wall (Figs. 12, 34). After the completion of 13 I n i t i a t i o n of the gy n o e c i a l w a l l , the f l o r e t apex continues growth as the o v u l e , but i t g r a d u a l l y becomes r e - o r i e n t e d to a v e r t i c a l and then to an a n t e r i o r l y - d i r e c t e d p o s i t i o n ( F i g s . 78, 79, 80). Awn-lemma Fo l l o w i n g the terminology and reasoning of Maze et a l . (1971), the awn and the lemma are considered together i n e a r l y development, and are r e f e r r e d to c o l l e c t i v e l y as the awn-lemma. Awn-lemma i n i t i a t i o n i s r-i n d i c a t e d by a p e r i c l i n a l d i v i s i o n i n the protoderm on the a n t e r i o r s i d e of the f l o r e t a p i c a l meristem, j u s t below the apex ( F i g . 1 5 ) . The i n i t i a t i n g d i v i s i o n i s followed by f u r t h e r p e r i c l i n a l d i v i s i o n s i n the protoderm and ground meristem, i n v o l v i n g three to four c e l l s i n the v e r t i c a l d i r e c t i o n . The awn-lemma i s formed from c e l l s of the protoderm and the f i r s t l a y e r of the ground meristem. From the a n t e r i o r s i d e , i n i t i a t i n g d i v i s i o n s spread around the l a t e r a l f l a n k s of the f l o r e t apex. The l a t e r a l spread of d i v i s i o n s r e s u l t s i n the formation o f a c r e s c e n t i c p r o t u s i o n which i s higher on the a n t e r i o r s i d e and slopes down p o s t e r i o r -l y ( F i g s . 16, 16a). Upward growth of the primordium i s mainly the r e s u l t of s u b - a p i c a l a c t i v i t y ( F i g . 1 7 ) . Increased a p i c a l a c t i v i t y a t the p o i n t of o r i g i n of the awn-lemma primordium r e s u l t s i n a f r e e extension ( F i g s . 18, 290). This a n t e r i o r extension i s the young awn. For ease of d i s c u s s i o n , growth of the a n t e r i o r p o r t i o n of the awn-lemma primordium w i l l be considered f i r s t , and the l a t e r a l p o r t i o n s l a t e r . In the young awn in c r e a s e i n height by a p i c a l a c t i v i t y i s soon fo l l o w e d by i n t e r c a l a r y growth. C e l l u l a r d i f f e r e n t i a t i o n , u s i n g c e l l 14 vacuolation as an indication, occurs very early in awn development. The cells at the distal end become vacuolated when the awn-lemma primordium is ca. 150 u in height. Growth in girth of the awn is the result of c e l l vacuolation and pericl i n a l divisions in the ground meristem (Fig. 19). At this stage i t i s not possible to distinguish the awn base from the lemma below i t . The cells in the proximal one-third of the awn-lemma primordium are smaller than those i n the distal two-thirds of the primordium, and are also vacuolate to a lesser degree. Certain interest-ing growth phenomena are observed at the time the gynoecial wall i n i t i a t e s . In the adaxial ground meristem of the awn-lemma, at a point more or less level with the i n i t i a t i n g gynoecial division, oblique divisions are seen (Fig. 20). The adaxial ground meristem cells in this region become actively mitotic and form a group of small, densely cytoplasmic cells with no particular arrangement (Figs. 21, 22). At the same time the corresponding abaxial ground meristem cells undergo pe r i c l i n a l and anti-c l i n a l divisions. Figure 22 shows the awn-lemma junction when the gynoecial wall appears on the posterior side of the f l o r a l apex. Further divisions in the ground meristem on the adaxial side are mostly peri-c l i n a l , forming regular c e l l f i l e s . The lemma apex becomes slightly expanded (Fig. 23). Adaxial protodermal cells of the lemma apex also start to divide p e r i c l i n a l l y , while the abaxial protoderm remains one-layered. Continued meristematic activity, together with re-oriented planes of growth, greatly increases the thickness of the lemma apex (Fig. 24). Cell enlargement and vacuolation of the c e l l f i l e s push each side of the apex outwards so that the fu l l y expanded lemma apex has a double convex shape, with the apex of each convexity directed 15 upwards. This is seen in Figure 25, at a stage when the integuments of the ovule are formed. The abaxial protoderm has adjusted to the increase in bulk of the abaxial ground meristem by an increased number of anti-clinal divisions. On the adaxial side there is a multiple protoderm of about two to four cells deep, spread over the upper one-third of the convexity. Accomodation of the adaxial protodermal cells to the increased volume of the ground meristem is through cell division and enlargement. The portion of the lemma that eventually encloses the palea and the rest of the floret develops from the lateral portions of the lemma (Fig. 292). Growth of the lateral portions is of course co-ordinated with that of the rest of the lemma. The lateral portions of the lemma grow around to the posterior side of the floret but do not meet. Initially they are taller on the anterior side and barely cover the base of the floret (Fig. 291). Concurrent with the growth of the anterior portion of the lemma, the lateral portions grow upwards, elongate tremendously and overtop the rest of the floret. Just prior to the formation of the integument in.ovule development, the upper portions of the free margins of the lemma have increased in height more than that part of the lemma to which the awn is attached (Fig. 24). This results in the free edges of the lemma forming two lobes (or 'ears') in front of the awn. Cell enlargement in the lemma finally obscures the patterns of cell divisions that resulted in the characteristic shape of the lemma apex. Cells at the awn-lemma junction remain small, and eventually become heavily lignified, with abundant simple pits in the cell walls. They do not form a cambium-like layer as that reported for Stipa lemmoni by Maze et a l . (1972). 1 6 C a l l u s The c a l l u s i s i n i t i a t e d through c e l l enlargement and c e l l d i v i s i o n i n the ground meristem at the base of the lemma, on the a n t e r i o r s i d e of the f l o r e t ( F i g . 26). This occurs at about the time of i n i t i a t i o n of the g y n o e c i a l w a l l . At t h i s stage the base of the f l o r e t where i t i s attached t o the r a c h i l l a i s almost h o r i z o n t a l . I n i t i a l growth stages of the c a l l u s i n v o l v e p e r i c l i n a l d i v i s i o n s , v a c u o l a t i o n and expansion, i n a d i r e c t i o n p a r a l l e l w i t h the adjacent protoderm of the c e l l s of the ground meristem ( F i g . 27). The protoderm c e l l s n e i t h e r d i v i d e p e r i c l i n a l l y nor enlarge p e r c e p t i b l y . They d i v i d e a n t i c l i n a l l y to keep pace w i t h the i n c r e a s e i n volume of the groumd meristem ( F i g s . 27, 28). By the time the g y n o e c i a l w a l l i s f u l l y i n i t i a t e d the c a l l u s has developed i n t o a rounded hump w i t h a downward-d i r e c t e d apex ( F i g s . 27, 28). Concomitant w i t h the growth of the c a l l u s , the formerly h o r i z o n t a l f l o r e t - r a c h i l l a j u n c t i o n s t a r t s to t i l t i n an a n t i - c l o c k w i s e d i r e c t i o n . The r a c h i l l a a l s o increases i n width ( F i g s . 27, 28, 29, 30). Re-o r i e n t a t i o n of the f l o r e t - r a c h i l l a j u n c t i o n can be a t t r i b u t e d i n pa r t to the f o l l o w i n g growth phenomena. On the a n t e r i o r and p o s t e r i o r s i d e s of the r a c h i l l a , p e r i c l i n a l d i v i s i o n s i n the ground meristem produce f i l e s of c e l l s p e r p e n d i c u l a r to the protoderm ( F i g s . 27, 28). A n t i c l i n a l d i v i s i o n s i n the ground meristem on the p o s t e r i o r s i d e of the r a c h i l l a form f i l e s of c e l l s p a r a l l e l w i t h the adjacent protoderm ( F i g . 2 9 ). C e l l enlargement i n the ground meristem on the p o s t e r i o r s i d e of the r a c h i l l a i n the a x i l of the second glume ( F i g . 30) and p o s s i b l y c e l l enlargement i n the t i s s u e s below the second glume (unfigured) l e a d to growth upward 17 of the p o s t e r i o r s i d e s of the r a c h i l l a and the lemma base. The continued enlargement, p a r a l l e l w i t h the protoderm, of the ground meristem c e l l s i n the c a l l u s , together w i t h the e l o n g a t i o n of the p r o t o -dermal c e l l s , pushes the c a l l u s f u r t h e r down ( F i g . 30). This downward growth i s concomitant w i t h f u r t h e r t i l t i n g upwards of the lemma base and the r a c h i l l a on the p o s t e r i o r s i d e , as discussed above ( F i g s . 29, 3 0 ) . When f u l l y mature, the rounded t i p of the c a l l u s i s on, or c l o s e t o , the v e r t i c a l a x i s of the f l o r e t ( F i g . 1 4 ) . Pale a The p a l e a a r i s e s a f t e r e a r l y growth of the awn-lemma primordium through the p e r i c l i n a l d i v i s i o n s of s e v e r a l protodermal c e l l s on the p o s t e r i o r s i d e of the f l o r e t apex ( F i g s . 18, 31). The i n i t i a t i n g d i v i s i o n s spread around the f l a n k s of the apex to the a n t e r i o r s i d e , but do not e n c i r c l e the apex. F o l l o w i n g i n i t i a t i o n , the c e l l s under-l y i n g the protodermal i n i t i a l s d i v i d e p e r i c l i n a l l y , but the major c o n t r i -b u t i o n to the pal e a i s made by d e r i v a t i v e s of the protoderm ( F i g s . 32, 33) . Growth of the pal e a i s through a p i c a l and su b a p i c a l a c t i v i t y . O b l i q u e - a n t i c l i n a l d i v i s i o n s of the a p i c a l i n i t i a l s g i v e r i s e t o the pa l e a protoderm, w h i l e meristmatic a c t i v i t y of the su b a p i c a l c e l l s g ives r i s e t o the i n t e r n a l t i s s u e s of the p a l e a ( F i g s . 34, 35). Marginal growth i s due to marginal and submarginal i n i t i a l s . Growth of the p a l e a i n e a r l y stages i s mainly due to c e l l d i v i s i o n . A f t e r the gyno e c i a l w a l l i s completely i n i t i a t e d around the apex, the pal e a grows mainly through c e l l e l o n g a t i o n . At ma t u r i t y the pal e a has a short b i s e r i a t e a p i c a l p o r t i o n and a broad base ( F i g s . 36a,b,c). The c e l l s which 18 enlarge the most are those on the adaxial surface. Posterior l o d i c u l e The posterior l o d i c u l e (Figs. 291, 292) appears shortly after the palea. Its i n i t i a t i o n i s indicated by p e r i c l i n a l divisions i n the proto-derm on the posterior side of the f l o r e t apical meristem , just above the palea (Figs. 20, 33). In other words, i t i n i t i a t e s d i r e c t l y from the f l o r e t a p ical meristem. Protodermal derivatives contribute solely to the tissues of the posterior l o d i c u l e . Early growth i s the r e s u l t of apical and subapical a c t i v i t y (Figs. 34, 35), and r e s u l t s i n an organ which i s considerably thinner than the palea (Fig. 37). At maturity the posterior l o d i c u l e i s a t h i n f l a p of homogenous, non-vascularised tissue, about three to four c e l l s thick (Figs. 38, 39). Anterior l o d i c u l e The inception and development of the anterior lodicules i s different from that of the posterior lodicules. I n i t i a t i o n of the anterior lodicules i s through p e r i c l i n a l divisions i n the f i r s t layer of the ground meristem, i n between the bases of the developing stamens. The s i t e of i n i t i a t i o n occurs over 2 - 3 c e l l s i n longitudinal section (Figs. 19, 40). I t has a deeper s i t e of i n i t i a t i o n than the posterior l o d i c u l e . Follow-ing the i n i t i a t i n g d i v i s i o n s , the protoderm c e l l s divide p e r i c l i n a l l y (Figs. 34, 42). In cross section, two separate s i t e s of i n i t i a t i o n of the anterior lodicules are seen (Fig. 45). Through repeated p e r i c l i n a l divisions i n the ground meristem short f i l e s of c e l l s are formed (Fig. 41). The anterior lodicules at t h i s stage appear as bulges (Fig. 42). 19 Growth i n height of the anterior lodicules i s intercalary (Figs. 43, 44). Apical growth i s not d i s t i n c t . Each anterior l o d i c u l e i s attached to the f l o r e t axis by a very broad base (Fig. 48). The c e l l s of the mature anterior l o d i c u l e form a homogenous tiss u e , supplied by two vascular strands (Fig. 49). Marginal growth i s i n i t i a t e d when p e r i c l i n a l divisions i n the protoderm spread from the s i t e of lo d i c u l e i n i t i a t i o n i n a posterior d i r e c t i o n (Fig. 46). A d i s t i n c t posterior margin i s soon formed (Fig. 47). When thi s i s discernible an anterior margin starts to grow (Fig. 47). Growth i s maximum i n the anterior portion of the lod i c u l e so that this portion of the l o d i c u l e i s thickest (Figs. 48, 49, 50). Stamens Stamen i n i t i a t i o n starts with p e r i c l i n a l divisions i n the f i r s t layer of the ground meristem (Figs. 18, 51). At i n i t i a t i o n and during subsequent stages of development the protodermal c e l l s divide a n t i c l i n a l l y only. Early growth results from divisions i n the ground meristem (Figs. 19, 40). The young stamen primordium very early assumes a globose form (Fig. 291). Anther formation involves almost a l l the c e l l s of the stamen primor-dium. Only a few c e l l s at the basal portion of the primordium undergo extensive elongation to form the filament. A young anther primordium i n cross section soon becomes s l i g h t l y 4-lobed (Fig. 52). Several hypodermal c e l l s become d i f f e r e n t i a t e d i n each lobe and are recognised by t h e i r larger size and conspicuous nuclei (Fig. 53). These form 20 the archesporial c e l l s . Stages in anther development are shown in Figures 54, 55, and 292. The f u l l y developed anther has four wall layers. The innermost wall layer, or tapetum, is of the secretory type. Gynoecium _^  The gynoecium is the last structure to develop from the floret apical meristem. The gynoecial wall i s f i r s t indicated by one or two pe r i c l i n a l divisions in the protoderm on the anterior side of the floret primordium, between the adaxial furface of the anterior stamen and the apex of the primordium (Fig. 57). That i s , i t initiates as a lateral appendage. This is followed by similar divisions in the ground meristem underlying the protodermal i n i t i a l s . A fold develops lat e r a l l y on the apex (Figs. 58, 293). Such mitotic a c t i v i t i e s , starting on the anterior side of the floret apex, extend around both flanks of the apex (Figs. 59, 60, 61), and eventually appear on the posterior side (Figs. 62, 63). The gynoecial wall i s thus transformed from a crescent-shaped primordium to a ring-shaped one. The posterior rim of the gynoecial wall becomes the site of active mitotic divisions (Figs. 64, 65, 66). The gynoecial wall then grows upward as a tube (Figs. 67, 68, 69, 70). The total f l o r a l apex is not used up in the formation of the gynoe-c i a l wall. Following the i n i t i a t i o n of the ring-shaped primordium, the apex of the floret primordium increases in size (Figs. 64, 65). It develops directly into the ovule. The two lateral sides of the gynoecial wall develop into style branches (Figs. 68, 71, 72). The style branches are solid and i n i t i a t e stigmatic hairs through c e l l elongation i n the protoderm cells (Fig. 73), 2 1 at a time when the megaspore mother c e l l i s formed in embryo sac develop-ment. These enlarged club-shaped cells undergo antic l i n a l and oblique divisions to form stigmatic hairs (Fig. 74). The stigmatic hairs are mostly localized on the inner surfaces of each style. Each style has two areas of specialized tissue, the stigmatoid and the vascular. In Figure 74, the unf i l l e d space in the style, towards the outer surface of the style, represents the vascular region, and the stigmatoid region, which is not represented, is situated towards the inner surface of the style, near the stigmatic hairs. Illustrations of the stigmatoid region are given i n the account on 0. hymenoides. b e The loculus of the ovary does not appear to Aclosed prior to or during f e r t i l i z a t i o n (Fig. 76). Tissue proliferated on the inner surfaces of the ovary wall bring the edges of the ovary wall closer together, but a c l e f t i s l e f t at the top of the ovary (Figs. 75, 75a, 76). The closing tissue consists of a group of small c e l l s , and is the 'stylar core' described by Arber (1934). The vasculature of the gynoecium was studied from seri a l sections of florets. Figure 77 shows a series of transverse sections of a floret at the megaspore stage. Subsequent to the detachment of a trace to each of the three stamens, the provascular tissue of the f l o r a l axis is in the form of a cylinder (Fig. 77a). At a higher lev e l , an anterior gynoecial trace is detached from this provascular cylinder (Fig. 77b), and a l i t t l e higher than this, two lateral traces become detached. The remainder of the central provascular axis continues directly to the ovule (Figs. 77c,d). There are, as shown in Figure 77d, four vascular 22 bundles in the ovary: two l a t e r a l , one anterior and one posterior. Each of the lateral vascular bundles enters a style (Figs. 77e,f,g). The posterior vascular bundle is the largest, and supplies the ovule. Both the anterior and posterior vascular bundles terminate in the ovary wall (Figs. 82a, 83a). Early ovule and embryo sac development  Ovule The whole floret apex is not consumed in the formation of the gynoecial wall. The residual apex forms a small convex dome, consisting of dense meristematic cells (Fig. 78). It develops directly into the ovule. With the formation of the ring-shaped gynoecial wall there is a slight s h i f t in position of the floret apex so that instead of i t s being directed towards the palea i t becomes anteriorly directed (Fig. 79). Concomitant with the growth of the posterior side of the gynoecial wall, the apex increases rapidly in size, and continues to t i l t anterior-ly (Fig. 80). By the time the integuments are initiated, the ovule appears to be borne laterally on the posterior side of the gynoecial wall (Fig. 81). As the ovule continues to grow, i t t i l t s towards the lemma (Fig.82a), becoming orthotropous at the megaspore mother c e l l stage (Fig. 83a), then bending downward (Figs. 85a, 88a, 90a), f i n a l l y becoming hemianatropous at the 8-nucleate stage in embryo sac develop-ment (Fig. 91a). Integuments Both the inner and outer integuments are of protodermal origin. They 23 are i n i t i a t e d as r i n g meristerns. The inner integument i n i t i a t e s f i r s t and i s indicated by protodermal p e r i c l i n a l divisions on both the upper and lower sides of the nucellus (Fig. 81). I n i t i a t i n g divisions of the outer integument are f i r s t seen on the upper side (Fig. 82), by which time the megaspore mother c e l l i s d i s t i n c t . The inner integument forms the micropyle. In early stages of i t s development the inner integument i s two c e l l s thick. By the megaspore stage the micropyle i s formed. Cell s at this end of the inner integu-ment st a r t to divide. The r e s u l t i s an inner integument with a thickened micropylar portion while the rest of i t remains two c e l l s thick (Figs.90, 91). This thicker portion delimits the micropyle. Throughout develop-ment the inner integument stains more intensely than the outer integu-ment. This feature, plus the smaller c e l l s , implies that the inner integument adjusts to the increase i n size of the ovule more by c e l l d i v i s i o n than by c e l l expansion. The outer integument i s two c e l l s thick throughout except on i t s upper side at the chalazal region. In th i s region there are two bumps (Fig. 88), the resu l t of a few inte r n a l c e l l layers. The bump nearest the micropyle i s discerned very early, at the time when the megaspore mother c e l l i s di f f e r e n t i a t e d (Fig. 83). The one further away from the micropyle appears at the tetrad stage, through p e r i c l i n a l divisions i n the outermost c e l l layer (Fig. 86). The c e l l s of the outer integument are larger and more vacuolate than those of the inner integument. Accomodation of the outer integu-ment to increasing ovule size i s mainly through c e l l enlargement. As 24 the ovule grows, the bumps increase i n s i z e . The f i r s t bump (that i s , the one nearer the micropyle) assumes an inverted cone shape ( Fig. 90). The second bump (that i s , the one further away from the micropyle) becomes a flange of tissue adpressed against the f i r s t bump. Both the bumps remain close to the point of ovule attachment. At the 8-nucleate stage, the two bumps are so closely adpressed that they appear as one structure (Fig. 91). This structure projects up into the c l e f t between the stylebranches (Fig. 91a), and may, as has been postulated, (see True, 1893; Weier and Dale, 1960), direct pollen tubes towards the micropyle. The outer integument begins to disintegrate soon after f e r t i l i z a t i o n . I t i s almost obliterated by the time a 4 - c e l l embryo i s developed. Nucellus Growth of the nucellus and re-orientation of the ovule occur simultaneously. During early megasporogenesis the ovule i s pushed from an u p t i l t to an orthotropous position through divisions of the nucellar c e l l s adjacent to the megaspore mother c e l l (Fig. 82). Subsequent divisions of the nucellar c e l l s around the megaspore, and l a t e r around the embryo sac, are oriented at an angle to the axis of the embryo sac (Fig. 83). This growth contributes to increase i n thickness of the ovule. By the 8-nucleate stage, mitotic a c t i v i t i e s of these c e l l s cease and the c e l l s elongate p a r a l l e l to the longitudinal axis of the embryo sac, thereby increasing the length of the ovule (Fig. 91). Some of the c e l l s next to the embryo sac are crushed by the transverse expansion of the l a t t e r . 25 The nucellar protoderm starts to divide pe r i c l i n a l l y at the late megaspore mother c e l l stage (Fig. 83). By the 4-nucleate stage most of the protodermal cells have divided at least once. Their contribution to the bulk of the ovule is most obvious at this stage (Fig. 90). From then on, as the ovule further increases in size the protodermal contrl-bution becomes less obvious (Fig. 91). Most of the growth of the nucellus i s actually due to activity i n the chalazal region. Early growth is in length (Fig. 84). Later growth is the outcome of c e l l divisions i n variously oriented planes, contributing to increase in bulk of the ovule as well as shifting the ovule to a hemianatropous position (Figs. 85, 88, 91). Further differentiation of the nucellar cells mostly involves c e l l expansion. Embryo sac The embryo sac is of the Polygonum type (sensu Maheshwari, 1950). The archesporial c e l l is differentiated as the megaspore mother c e l l before the floret is extruded from the inflorescence sheath. Figure 84 shows a stage in megasporogenesis. The tetrad that is formed is usually linear (Fig. 85), but is occasionally T-shaped (Fig. 87). The functional e megaspore is the chalazal one. The f i r s t megaspore to degenrate is either the one above the chalazal megaspore, or the next one above. The chalazal megaspore is the functional megaspore. The embryo sac undergoes changes in shape as i t develops. At the 2-nucleate stage i t is oblong (Fig. 89), and becomes broadly fusiform at the 4-nucleate stage (Fig. 90). At the beginning of the 8-nucleate stage the micropylar end has broadened out so that the embryo sac appears 2 6 broadly ovate at this end, while at the chalazal end i t abruptly narrows to a s l o t (Fig. 91). Just before f e r t i l i z a t i o n the s l o t widens (Fig. 94). The.embryo sac at t h i s stage i s ovate (Fig. 94). From th i s stage on, i t increases i n length (Figs. 97, 98). At the early 8-nucleate stage (Fig. 91), i t i s hot possible to distinguish the egg from the synergids. As the synergids d i f f e r e n t i a t e a large vacuole develops i n each of them at the lower (= chalazal) end, and the nucleus i s at the upper (= micropylar) end (Fig. 92). No f i l i f o r m apparatus i s seen i n the synergids. When the egg and the synergids have di f f e r e n t i a t e d the antipodals have divided once (Fig. 92). The a n t i -podals l i e i n the s l o t at the chalazal end of the embryo sac. Their cytoplasm i s dense, granular, and non-vacuolate. The polar nuclei at th i s stage are situated above the egg and synergids and are very close to them. Just prior to f e r t i l i z a t i o n certain v i s i b l e changes occur. The cytoplasm of the synergids becomes dense and granular. One synergid decreases i n size (Fig. 94), and then degenerates. The antipodals have pr o l i f e r a t e d even more and have moved to a l a t e r a l position. The outlines of the antipodals are d i f f i c u l t to distinguish. No mitotic figures were seen. Each antipodals appears to have one nucleus only. The polar nuc l e i move away from the i r position next to the egg and synergids to a higher position i n the central c e l l . F e r t i l i z a t i o n F e r t i l i z a t i o n occurs after one synergid has degenerated. The pollen 27 tube travels through the micropyle and nucellus and seems to enter the embryo sac between the base of the persistent synergid and the embryo sac wall (Figs. 96, 97). The t i p of the pollen tube appears to have a deposit of a darkly staining material. Actual discharge of the male gametes was not seen. I t could not be ascertained whether the pollen tube contents are discharged into the synergid before entering the egg, or are discharged d i r e c t l y into the egg. A quantity of chromatin-like materials i s seen on top of the egg. This may be pollen tube contents plus remnants of the degenerated synergid. The polar n u c l e i do not fuse before f e r t i l i z a t i o n . P o s t - f e r t i l i z a t i o n This was studied up to the 4 - c e l l embryo stage. The endosperm i s free-nuclear and i n i t i a l l y develops faster than the embryo (Fig. 98). The hitherto persistent synergid begins to degenerate. Oryzopsis hymenoides (Roem. and Schult.) Ricker A comparison of Oryzopsis hymenoides with 0. virescens w i l l be emphasized i n this account. Floret organogenesis Organ inception follows this sequence: awn-lemma (Fig. 99), palea and stamens (Fig. 100), lodicules and callus (Fig. 101), gynoecial wall (Fig. 102), and f i n a l l y , the ovule (Fig. 104). Oryzopsis hymenoides 28 d i f f e r s from 0. virescens i n the precocious development of the callus and the early d i f f e r e n t i a t i o n of the awn-lemma junction. By the time the gynoecial wall reaches the posterior side of the f l o r e t apex the awn-lemma junction and the callus are well-marked (Fig. 103). Further development of the f l o r e t i s seen i n Figures 104 and 105. Floret apical meristem As i n 0. virescens there seems to be no inner tunica i n 0. hymenoides (Figs. 99, 100). The f l o r e t a p i cal meristem also becomes displaced and re-oriented i n the course of f l o r e t development. Awn-lemma I n i t i a t i o n of the awn-lemma primordium i s as i n 0. virescens (Fig. 99). Subapical a c t i v i t y i s involved i n the growth of the young primor-dium (Fig. 100), but cessation of this a c t i v i t y occurs e a r l i e r than i n 0. virescens. A shorter awn i n 0. hymenoides i s perhaps an expression of the e a r l i e r cessation of subapical a c t i v i t y . Intercalary growth and c e l l u l a r d i f f e r e n t i a t i o n have produced a broader awn by the time of l o d i c u l e i n i t i a t i o n (Fig. 106; cf. Fig. 19). D i f f e r e n t i a t i o n of the awn-lemma junction i s seen when the gynoecial wall i n i t i a t e s (Fig. 107). This process i s comparable with that seen i n 0. virescens. Ce l l s i n the region of the presumptive junction are small, have dense cytoplasm, and undergo oblique divisions (Fig. 108). Re-orientation of d i v i s i o n a l planes and continued divisions result i n an expanded lemma apex which i s si m i l a r to that i n 0. virescens except for two features: i n 0. hymenoides 29 fewer p e r i c l i n a l divisions occur i n the ground meristem; the adaxial protoderm remains one-layered (Figs. 109, 110, 111). Hairs on the mature lemma are protodermal outgrowths. As i n 0. virescens the upper portion of the free margins of the lemma i s longer than the portion of the lemma attached to the awn and forms two 'ears' i n front of the awn (Fig. 105). At maturity, where the awn base i s attached to the lemma apex, the c e l l s remain small (Fig. 112). This zone presents a l i n e of weakness and accounts for the e a s i l y deciduous nature of the awn. Callus The cal l u s of 0. hymenoides d i f f e r s from that of 0. virescens i n i t s e a r l i e r development and i t s shape at maturity. But they are s i m i l a r i n the i r i n i t i a l stages of development and i n their c e l l u l a r composition. I n i t i a t i o n occurs at the time that the lodicules appear, and i s through expansion, vacuolation and d i v i s i o n of ground meristem c e l l s (Fig. 113). At the base of the lemma on the anterior side, ground meristem c e l l s near the protoderm become vacuolate and enlarge, i n planes p a r a l l e l with and perpendicular to the adjacent protoderm (Fig. 113), forming a s l i g h t bulge outwards. Continued c e l l enlargement of the ground meristem c e l l s , accompanied by p e r i c l i n a l and oblique d i v i s i o n s , increases the size of of the bulge (Fig. 114). The protoderm adjusts to this increase i n size by dividing a n t i c l i n a l l y . The callus shown i n Figure 114 i s at the stage when the gynoecial wall i n i t i a t e s on the anterior side of the f l o r e t apex. I t i s comparable with the callus of 0. virescens seen i n Figure 27, which i s during the growth of the anterior portion of the 30 gynoecial w a l l . During the early growth of the anterior portion of the gynoecial wall i n 0. hymenoides, increase i n size of the callus i s mainly through c e l l expansion, i n a dir e c t i o n perpendicular to the adjacent protoderm (Fig. 115). Further growth of the callus sees a s h i f t i n the dire c t i o n of expansion of the ground meristem c e l l s . When the gynoecial wa l l appears on the posterior side of the f l o r e t apex, expansion of the c e l l s of the callus i s predominantly i n a direc t i o n p a r a l l e l with the protoderm (Fig. 116). Thus the callus i s distended downward instead o-f outward, as i n the e a r l i e r stages. At the same time, some of the proto-dermal c e l l s begin to form h a i r s . Continued development of the callus involves extensive elongation of the ground meristem c e l l s , not only downward but at an angle to the adjacent protoderm, so that the callus grows downward and obliquely outward (Figs. 117, 118, 119). The proto-dermal c e l l s at the t i p of the call u s remain small, but those higher up on the callus extend l o n g i t u d i n a l l y to keep pace with the expansion of the ground meristem c e l l s . When f u l l y formed, the call u s has a sharp pointed t i p , directed away from the v e r t i c a l axis of the f l o r e t (Figs. 105, 119). The anti-clockwise t i l t i n g of the f l o r e t - r a c h i l l a junction seen i n 0. virescens i s absent i n 0. hymenoides. For ease of comparison, a table i s given of the figures that i l l u s -trate comparable stages i n awn-lemma and callus development i n the two species, using gynoecium development as the standard of reference (Table I ) . 31 TABLE I. Reference table for comparison of stages in awn-lemma and callus development in 0. virescens and 0. hymenoides, using the stage of gynoecium development as a marker. The numbers refer to figures at the pertinent stages. Stage of gynoecium development Initiation of lodicules Initiation of anterior gynoecial wall Growth of anterior gynoecial wall Gynoecial wall appears on posterior side of floret apex Ring-shaped gynoecial wall Prior to integument initiation Initiation of inner integument on upper side Complete inner integument Ovule in orthotropous position 0. virescens 0. hymenoides awn-lemma callus awn-lemma callus 19 20 21 22 23 24 none 25 none 19 26 27 28 none 29 none 30 none 106 107 108 109 110 none 111 none 112 113 114 115 116 117 none 118 none 119 32 Palea Development of the palea in 0. hymenoides is similar to that in 0. virescens (Figs. 120, 121, 122). At maturity the palea differs in the presence of elongate protodermal cells at the distal end (Figs. 123, 123a). Lodicules No visible difference can be seen in the initiation of the posterior lodicule as compared with that in 0. virescens (Figs. 120, 121). However, the fully developed posterior lodicule in 0. hymenoides is relatively thicker and has a blunt distal end (Fig. 124; cf. Fig. 38). This is undoubtedly due to more cell division as development proceeds. The anterior lodicules initiate and develop as in 0. virescens (Fig. 125). At maturity they are uniformly thick structures with an acute abaxial tip (Figs. 126, 127, 128), unlike those in 0. virescens which have an exceptionally thickened anterior portion (cf. Fig. 50). Stamens Development of the stamens is as in 0. virescens, except for the presence of 'bearded anthers' in 0. hymenoides. The anther 'beard' consists of multicellular, uniseriate hairs formed from protodermal cells at the distal end of the anther . An early stage of development of these hairs is seen Figure 129. 33 Gynoecium This structure in i t i a t e s and develops in a similar manner as described for 0. virescens (Fig. 130). Figures 131 and 132 show that that portion of the floret apex which remains after the i n i t i a t i o n of the ring-shaped gynoecial wall continues to grow as the ovule. At anthesis the inner margins of the ovary wall meet but do not fuse, so that the opening between the margins is not completely closed, but remains as the stylar canal (Figs. 113, 294, 295). Around the stylar canal is a region of smaller c e l l s — this is the 'stylar core'. There are two styles, each formed from a lateral portion of the gynoecial wall. The styles are closer here than they are in 0. virescens (Fig. 134; cf. Fig. 76). The vascular supply of each style i s an exten-sion of a lateral vascular bundle of the ovary (Fig. 294). The stigma-toid tissue in each style runs parallel with the vascular bundle in the style. According to Bonnett (1961), the stigmatoid tissue in Avena  sativa initiates in the ovary wall above the f i r s t c e l l layer of the inner surface of the ovary wall and proceeds acropetally into the styles. Initiation of stigmatoid tissue was not studied by the author. Figure 295 shows the position of the stigmatoid tissue In the ovary wall. Early ovule and embryo sac development  Ovule The ovule of 0. hymenoides is similar to that of 0. virescens except for some slight differences in size. The ovule in 0. hymenoides at the inner integument stage is narrower (Fig. 135). At the 8-nucleate stage 34 i t i s also not as wide at the chalazal end (Fig. 143b). Integuments Inception and development of the integuments does not d i f f e r much from that i n 0. virescens (Figs. 135 to 138). Some differences between them are noted. The chalazal portion of the outer integument i n 0. hymenoides d i f f e r s from that i n 0. virescens. In 0. hymenoides the outer integument has only one bump i n the chalazal region, whereas i n 0. virescens two bumps are present. Whereas i n 0. virescens the two bumps are adpressed to appear as one prominent bump which projects into the c l e f t between the s t y l e branches and persists u n t i l a f ter f e r t i l i z a -t i o n , the single bump i n 0. hymenoides i s poorly developed (Fig. 141), and i t st a r t s to f l a t t e n out pr i o r to f e r t i l i z a t i o n (Fig. 142). At the time of f e r t i l i z a t i o n the outer integument has already started to degenerate (Fig. 143b). Nucellus Growth patterns i n the nucellus are s i m i l a r i n both species, except that there are more p e r i c l i n a l divisions i n the nucellar protoderm i n 0. virescens (cf. Figs. 142, 90). Embryo sac As i n 0. virescens there i s one archesporial c e l l i n 0. hymenoides. Megasporogenesis i s the same i n both species (Figs. 139, 140). The two species d i f f e r notably i n the shape of the early embryo sac and i n the d i f f e r e n t i a t i o n of the synergids. In 0. hymenoides the embryo sac i s almost rectangular at the 4-nucleate stage and i n the early 8-nucleate stage. The chalazal s l o t that i s so cha r a c t e r i s t i c of 35 the embryo sac in 0. virescens is absent in 0. hymenoides. At maturity, but before f e r t i l i z a t i o n , the shape of both the embryo sacs becomes ovate (Fig. 143b). Differentiation of the egg v i s i b l y resembles that in 0. virescens (Fig. 142a). The mature unfertilized egg i s highly vacuolate and also lacks starch grains (Fig. 143a). The differentiated synergids, instead of possessing a large vacuole at the lower (= chalazal) end, as is seen in 0. virescens, have a number of scattered smaller vacuoles. A f i l i -form apparatus develops in each of the synergids. Even before f e r t i l i z a -tion the two synergids are different from each other — one degenerates anti stains deeply, while the other becomes highly vacuolate and stains l i g h t l y . The polar nuclei and antipodals appear similar to those of 0. virescens. Fe r t i l i s a t i o n This process appears to be different from that observed in 0. virescens in two respects: the presence of two synergids as opposed to one in 0. virescens, and the discharge of the pollen tube contents into the degenerate synergid. From serial sections of the floret shown in Figures 144a,b,c, chromatin-like bodies are seen in the egg. Also, remnants of the pollen tube can be found in the degenerate synergid (Fig. 145a). The sections examined seem to indicate that the pollen tube contents reach the egg via the degenerate synergid (Figs. 144a,b,c). The polar nuclei are not fused prior to f e r t i l i z a t i o n . 36 Post-fertilization Observations were made up to the 2-cell embryo stage (Fig. 146). The hitherto persistent synergid begins to degenerate. Endosperm formation is initially free-nuclear. DISCUSSION General remarks about the interpretation of hlstogenetic data Histogenetic studies usually follow one or both of these approaches: 1. the vegetative shoot axis and/or the reproductive axis of one or more species are/is studied and the histogenetic data obtained are compared and used to deduce the phyllome or caulome nature of an organ; 2. a broad comparative histogenetic survey is made of an organ-category in as many species as possible (such as the study by Sprotte, 1940). The first method is by far the more popular and is the one used in this study (see Rohweder, 1963, for a critique of the method). It has to be borne in mind that interpretative histogenesis has no vocabulary of its own but is expressed in the language of comparative morphology. Moreover, concepts of organs such as ' l e a f , 'stem', 'lemma', and 'stamen' originate from comparative morphology. An illustration of these two statements is given in these words used to describe the development of the lemma, viz., the lemma is leaf-like. To the author, this simply indicates that the nature of the lemma can best be understood by compar-ing i t to a leaf. Both organs f i t the criteria of a phyllome. At no 37 time i s i t Implied that the lemma had been, at some stage i n the past, a le a f which became modified into a lemma. To v i s u a l i z e the lemma as having been derived from a foliage leaf i s to replace a purely morpho-l o g i c a l concept by a h i s t o r i c a l concept for the v a l i d i t y of which h i s t o -genetic data provide no evidence. Any morphological interpretation of an organ on the basis of i t s histogenesis has to be approached with caution. This i s especially c r u c i a l when is o l a t e d developmental observations are made on highly specialized structures, p a r t i c u l a r l y the stamens and carpels. There i s always the p o s s i b i l i t y of wrong inferences, as the developmental v a r i a -t i o n within these organs i s not w e l l known and has not been observed s u f f i c i e n t l y . The interpretation of the leaf and the stem as being of a phyllome and caulome nature respectively can be made with more c o n f i -dence, especially i n the grasses. Leaf and bud development i n the Gramineae has been studied i n numerous species (Sharman, 1945; Pankow and Guttenberg, 1959). In th i s family the leaves have a shallow s i t e of i n i t i a t i o n ( i n the tunica), and show marginal growth. In comparison, the l a t e r a l shoot axes are established i n the deeper layers of the main axis (in the corpus), do not show marginal growth but approach a r a d i a l symmetry. Also, a s h e l l zone i s present. Generally, leaves have a shallow s i t e of i n i t i a t i o n , but there have been occasional reports i n other angiosperms i n which the leaf i n i t i a t e s i n the deeper layers of the apex. In I r i s germanica (Rudiger, 1939), p e r i c l i n a l divisions which i n i t i a t e leaf formation occur f i r s t i n the two outermost corpus before they occur i n the second tunica layer. In Euphorbia l a t h y r i s (Soma, 38 1959), lea f i n i t i a t i o n involves the two tunica layers as well as the outermost corpus layer. However, i n both cases, no account was given of a x i l l a r y bud i n i t i a t i o n . Later growth stages of the leaf were not described. Rohweder (op. c i t . ) concludes from t h i s and Schnabel's work (Schnabel, 1941, c i t e d by Rohweder) that i n 'diesem F a l l e verhalten sich die Blattprimordien i n sehr jungen Stadien also ganz wie ein Achsen-VP, um so mehr, als s i e anfanglich annahernd radiar-symmetrische Gebilde darstellen'. The writer feels that caution i s as much i n place i n the negation of the phyllome nature of a structure as i n i t s affirmation. F l o r a l morphology  Awn-lemma In i n i t i a t i o n and development the lemmas i n 0. virescens and 0. hymenoides resemble leaves. That the lemma i s l e a f - l i k e i s a descrip-tion that agrees with a l l the grass lemmas that have been described so far (Cannon, 1900; P h i l i p s o n , 1934, 1935; Bonnett, 1953, 1961; Holt, 1954; Barnard, 1955, 1957a; Sharman, 1960a, 1960b; Klaus, 1966; Mehlenbacher, 1970; Maze et a l . 1971, 1972), and i s probably applicable to a l l grass lemmas. As currently interpretated, the lemma represents a bract borne on the spikelet a x i s , specialized for flower protection, and i n the a x i l of which arises the rest of the grass f l o r e t . There have been numerous attempts i n the past to equate the component structures of the awn-lemma with those of the grass l e a f . About a hundred years ago, Van Tieghem propounded the view that when the lemma has a subterminal awn, the awn i s equivalent to the grass blade, and the parts 3 9 of the lemma above and below the Insertion of the awn are the equivalents of the ligule and sheath respectively. In those lemmas where the awn is terminal, the awn is considered as the blade, the whole of the lemma the sheath, and its lateral lobes, i f any, as stipules which by fusion are supposed to give rise to the ligule. Philipson (1934), challenged this interpretation and suggested that in lemmas with abaxial awns, the awn does not represent the whole blade but only a part of the blade which has become separated from the main portion of the blade, and this main portion is represented by the median and terminal portion of the lemma. Recently an interesting parallel was drawn by Maze et a l . (1971), between the growth pattern of the awn-lemma of Stipa tortilis and Oryzopsis miliacea and that of the leaf of Oryza sativa (Kaufman, 1959). Certain growth phenomena in the awn-lemma of S_. tortilis which were used for comparison with Oryza by Maze et a l . are also seen in 0. virescens (viz., multiple periclinal divisions in the adaxial protoderm and ground meristem of the expanded lemma apex). The author concurs with Maze et a l . that the developmental similarity between the Oryza leaf and the awn-lemma of species of Stipa and Oryzopsis is a reflection of the fact that plants are developmentally very simple, and that this similarity does not imply the derivation of one structure from the other. Furthermore, the awn-lemma of the Stipeae is like a grass leaf in that i t has a sheathing base (lemma), and an appendage (awn). It is highly questionable i f anything can be gained from trying to speculate what are the parts taken in the lemma by the various regions of the grass leaf. Whether any significance can be attached to the presence of a 40 multiple protoderm in the lemma apex of 0. virescens and the absence of such in 0. hymenoides cannot be answered in this study. It does, however, result in a fatter apex In 0. virescens. Callus The term callus is usually used for the hardened lower end of the lemma,(Hitchcock, 1951; Pohl, 1968). Developmentally the callus is formed by the downward projection of the base of the lemma. This observation agrees with the earlier observation of Weatherwax (1942), t on 0. hymenoides. Maze et, a l . (1971) were non-commital as to the origin of the callus. In some grasses, such as Heteropogon, Chrysopogon, the base of the spikelet at the point of its articulation with the rachilla forms a sharp projection which simulates the callus formed by the lower end of the lemma in Oryzopsis. The sharp structure at the base of the spikelet is also termed a callus. The development of 'spikelet calluses' has not been studied. It is possible that the development of 'spikelet calluses' and 'lemma calluses' might be different. A sharp callus, whether i t is a structure on the spikelet or the lemma, is an adaptation for dispersal by animals, by adhering to their skin or fur (Stebbins, 1956). In the Gramineae, the development of a similar mechanism for seed dispersal may involve differ-ent original structures. Another example is seen in the modification into bristles for wind dispersal of awns in Aegilops umbellulata and much divided glumes in Sitanion (Stebbins, 1972). 41 Palea Like the lemma, the palea initiates and develops in the manner of a phyllome. Most workers interpret the palea as a bracteole borne on the floral axis, which, together with opposing lemma, encloses the grass flower. The lemma, palea, and the grass flower collectively form the grass floret. Schuster (1910), postulated the homology of the palea with two fused sepals, which, together with a hypothetical third sepal, formed the outer perianth series of the grass flower. The example cited by Schuster to support his hypothesis is the extant South Ameri-genus Streptochaeta, in which the palea is bifid almost to the base. On the basis of histogenetic data, the palea seems to f i t best the current interpretation as a bracteole. Its phenetic similarity with a prophyll has been discussed by Arber (1925, 1934), and Philipson (1934). Lodicules Much controversy has centred on, and s t i l l does, the interpretation of the nature of the lodicules. These structures have been variously interpreted as bracts (Hackel, 1881), or modified perianth members (Rowlee, 1898; Schuster, 1910; Arber, 1934; Gould, 1968). Recent histogenetic studies have brought the following interpretations. Barnard (1955, 1957a) describes lodicules as 'appendages with a foliar-like origin developed on the axis of the flower primordium'. Bonnett (1953, 1961) reports that the anterior lodicules in Zea and Avena 42 initiate in the corpus, below the unlseriate tunica, and hence are to be considered as modified stems. Mehlenbacher (1970) found that in Oryzopsis hendersoni the anterior lodicules are partially stem-like and partially leaf-like. Maze et a l . (1971) contend that anterior lodicules are 'de novo' structures and are not comparable with any other plant structures. Moreover, Maze et a l . decry the interpretation of the anterior lodicules as homologues of perianth members, or any other modified structures, on the grounds that such an interpretation would require the assumption that lodicules evolved from perianth or any other structures, and in the process of evolution lost a l l their characteristic features. In view of the varying opinions on the nature of the lodicules, the author would like to discuss the interpretation of the lodicules in some detail. Hackel (1881), considered the anterior lodicules to be lateral halves of a leaf alternating with the palea, the middle part of which rarely develops fully. The posterior lodicule when present was supposed to continue the distichous arrangement of the palea and the anterior lodicules. This interpretation has to depend on a distichous arrange-ment of the lodicules, and cannot be upheld. The trimerous disposition of the lodicules in grasses, especially in the bamboos, has been demonstra-ted, Barnard (1955, 1957a) regards the grass flower as a branch system and categorizes the lodicules as foliar structures which are borne laterally on the main axis of the branch system. His interpretation of the lodicules as foliar structures is rather non-committal. 43 That the lodicules are stem-like i s the conclusion Bonnett (1953, 1961) arrived at after some elegant studies on maize and oat. This interpretation i s based primarily on the deeper s i t e of i n i t i a t i o n of the l o d i c u l e s . Bonnett's text-figures i n this connection are not convincing. For example, i n h i s paper on maize i n 1953, Figure 15D shows a l e a f - l i k e lodicule primordium with p e r i c l i n a l divisions i n the protoderm; and i n his paper on oat i n 1961, Figures 13B and 17E show leaf and l o d i c u l e i n i t i a t i o n respectively, both with i n i t i a t i n g p e r i -c l i n a l divisions i n the c e l l layer beneath the protoderm. Maze et a l . (1971, 1972) contend that the majority of developmental features seen i n the anterior lodicules (= dorsal lodicules i n the papers by Maze et al.) are unique and point to a 'de novo' o r i g i n for these structures. The unique features c i t e d by Maze et a l . are: (1) the anterior lodicules do not i n i t i a t e d i r e c t l y from the f l o r e t apical meristem, but, instead, from the base of the developing stamens; (2) they i n i t i a t e i n an area of some c e l l d i f f e r e n t i a t i o n ; (3) i n i -t i a t i o n i s spread over a considerable area as seen i n longitudinal section; (4) early growth i n the lodicules i s the resu l t of mit o t i c a c t i v i t i e s over a considerable portion of the organ ; (5) the pattern of marginal growth i n the anterior lodicules has not been described i n any other organ. The author disagrees with the extreme Interpretation of Maze et a l . and w i l l review i n d e t a i l each of t h e i r above arguments. (1) That the anterior lodicules do not arise d i r e c t l y from the f l o r e t a p i c a l meristem i s a consequence of the time of t h e i r i n i t i a t i o n — they ar i s e l a t e r than 44 the stamens. In other words, their inception i s centrifugal. While a centripetal (= acropetal) sequence of organ inception occurs i n many of the f l o r a l apices studied so far, a centrifugal sequence has been report-ed i n several plant taxa (see, for example, Corner, 1946). Cheung and Sattler (1967) report that in Lythrum sa l i c a r i a centrifugal inception occurs in the formation of not only one but three kinds of appendages, sepals, petals and stamens. Sattler (1967) cites a few genera in which the petal primordium i s initiated on a common stamen-petal complex, so that a petal primordium does not always have to i n i t i a t e directly on the f l o r a l apex. (2) This feature is a direct corollary of (1) and should not be considered. (3) This argument is based on a very limited number of observations (Stipa t o r t i l i s and Oryzopsis miliacea, Maze et a l . 1971; Stipa lemmoni, Maze et a l . , 1972), and is not upheld by other histogenetic studies, including the author's (cf. Figs. 40, 41, for Oryzopsis virescens and Fig. 120 for 0. hymenoides). In Bambusa  arundinacea (Barnard, 1957a, Figs. 3, 5), and Avena sativa (Bonnett, 1961, Figs. 17 E,F), i n i t i a t i o n of the lodicules involves only two to three cells in the vertical direction. (4) While this i s a v a l i d point with respect to anterior lodicule growth i n those species of Stipa and Oryzopsis that have been investigated by Maze ejt a l . and the author, there are no data on other grasses for comparison. Moreover, the limited observations available, as mentioned above, are on closely related taxa, so that any generalization of the uniqueness of this feature is open to suspicion. (5) The same criticisms leveled at (4) also apply to (5). 45 In the author's opinion, some of the 'unique' features that Maze et a l . describe are not so unique after a l l , and others are based on too l i m i t e d a sample. Furthermore, the interpretation of Maze et a l . i s that the posterior l o d i c u l e i s dif f e r e n t from the anterior l o d i c u l e . These workers offer three possible Interpretations of the posterior lodicule on the basis of t h e i r histogenetic studies, v i z . , (1) as a 'de novo' structure; (2) as a f o l i a r structure which i s i n the process of being l o s t ; (3) as a f o l i a r structure which i s i n the process of conversion to an organ s i m i l a r to the anterior lodicules. Again, available data on posterior l o d i c u l e growth are too scanty to permit any generalization of t h i s nature. Moreover, i n the bamboos, such as Bambusa nutans(Arber, 1925), Arundinaria f a l c a t a (Rowlee, 1898), the anterior lodicules and and the posterior l o d i c u l e are a l i k e i n structure, and i t i s not un l i k e l y that the posterior lodicule i s histogenetically s i m i l a r to the anterior l o d i c u l e s . I t seems more l o g i c a l to the author to consider a l l lodicules as one whorl of f l o r a l appendages of one class. There are perianth-like features i n the anterior l o d i c u l e s . These are: (1) the i n i t i a t i o n of the anterior lodicules involves p e r i c l i n a l divisions i n the protoderm; (2) the anterior lodicules are determinate organs; (3) development of the anterior lodicules involves marginal growth. Histogenetic data do not preclude the interpretation of lodicules as modified perianth structures, and the author prefers this view. On the other hand, there are no histogenetic data at present that unequivocally support or negate this view. The amount of information on grass-floret 46 development i s very scanty. The difference i n form between the anterior and posterior lodicules i s perhaps related to thei r function. At anthesis, the anterior lodicules become turgid and force the lemma outwards. The posterior lodicule i s relegated to an i n s i g n i f i c a n t r o l e . I t Is interesting that protogyny occurs i n certain grass species, such as Anthoxanthum odoratum L. and Alopecurus pratensis L., i n which there are no lodicules. Arber (1926, 1927, 1928, 1934) described structures intermediate between lodicules and stamens i n cultivated plants of Cephalostachyum  virgatum and i n w i l d plants of Schizostachyum l a t i f o l i u m . She labeled them as stamen-lodicules and commented that although the existence of lodicul a r stamens may not i n i t s e l f prove the perianth nature of the lo d i c u l e s , i t lends p r o b a b i l i t y to this view, as i t i s not unusual to f i n d perianth members associated with stamens. Of course these unusual structures could also be dismissed as t e t r a l o g i c a l organs of no importance. Although both Arber and Schuster interpreted the lodicules as members of an inner perianth s e r i e s , they d i f f e r e d i n their interpretation of the outer perianth series. Arber's view i s that the outer perianth series has been l o s t i n the evolution of the grass flower. The reason she postulated a b i s e r i a t e l o d i c u l a r series was to bring the grass flower closer to the t y p i c a l monocotyledonous f l o r a l diagram. She acknowledged that her f l o r a l diagram of the grasses 'merely formed a hypothetical framework upon which to arrange her ideas'. Schuster's views are extreme-l y i n t e r e s t i n g and deserve more than the scant attention that i s paid to 47 them. Schuster suggested that the palea was homologous to two outer perianth members and that the third outer perianth member probably disappeared during the evolution of grasses from their ancestral forms. The lodicules formed the inner perianth. In the evolution of the grass flower the lodicules became reduced in structure as a result of the enveloping of the flower by the lemma. The formation of the lodicules as thickened structures was a later adaptation, brought about by the fusion of the originally two separate paleas. As a result, the flower became completely enclosed. Because of an altered situation, the lodicules had adapted to another function, and this was to open the opposing lemma and palea by swelling. However interesting these two hypotheses are, i t has to be kept in mind that they are speculative reconstructions of ancestral forms and have no factual existence. Stamens In the two species of Oryzopsis studied, the histogenesis of the =- r stamens differs from that of a f o l i a r structure. The stamens have a deeper site of i n i t i a t i o n , and in early stages of growth assume a globose form. Also, in stamen i n i t i a t i o n the protoderm undergoes antic l i n a l divisions only. Similar developmental features have been reported in stamen histogenesis in other grasses by Bonnett (1953, 1961), Holt (1954), Barnard (1955, 1957a), Mehlenbacher (1970), and Maze e£ a l . (1971, 1972). Maze et a l . also report the presence of a shell zone in 48 stamen formation in Stipa tortilis and Oryzopsis miliacea. These workers consider the stamens to be stem-like in grasses. Stem-like stamens have also been reported by Satina and Blakeslee in Datura (1943), by Barnard in Carex. Scirpus and Cyperus (1957b), in Juncus and Luzula (1958), and in Stypandra and Bulbine in the Liliaceae (1960). Various other workers, among whom may be mentioned McCoy (1940), Boke (1949), Tepfer (1953), Tucker (1959), Cheung and Sattler (1967), and Singh and Sattler (1972), consider stamens to be leaf-like. The stamens in question were found to initiate in the same layers as perianth members and carpels. Stamens in Downingia bacigalupii (Kaplan, 1968), and Hordeum distichon (Klaus, 1966), initiate beneath the superficial layers of the floral meristem, but have been interpreted as morphologically of a phyllome nature. The exact significance of these contradictory findings is not clear. One could as easily argue for leaf-like stamens as for stem-like stamens. It is interesting to note thatMerxmulIer and Leins (1967) have shown that in Sisymbrium strictissimum members of the same set of appendages (in this case stamens), may be initiated in different cell layers. Histo-genetic data for stamen development in the grasses studied so far support a cauline interpretation for grass stamens. Gynoecium The unilocular ovary of the grass flower has often been interpretated as either unicarpellate or tricarpellate. Among those who have upheld 49 the tricarpellate condition may be mentioned Schuster (1910), Arber (1934), Hitchcock (1951), and Gould (1968). Among adherents of the opposite view are Hackel (1889), Bews (1929), and Pilger (1954). Proponents of the tricarpellate condition consider the vasculature of the gynoecium and the presence of three styles in some grass flowers as evidence of three fused carpels. In most grass gynoecia four vascu-lar bundles are present, one posterior, one anterior and two lateral. The anterior and the lateral bundles each represents the midrib of one carpel supposedly. The posterior bundle supplies the ovule and is interpreted as the fused lateral bundles of the lateral carpels. Such a tricarpellate gynoecium would f i t in 'nicely' with the basic trimerous plan of the monocotyledonous flower. Developmental studies have not upheld this interpretation. The gynoecial wall in those grasses studied by Bonnett (1953, 1961), Holt (1954), Sharman (1960b), Klaus (1966), Mehlenbacher (1970) and Maze et a l . (1971, 1972) initiates as a single leaf-like structure. The author's observations in Oryzopsis virescens and 0^ hymenoides agree with the interpretation of these workers. Barnard (1957a) has suggested that the grass gynoecium consists of four carpels, one anterior, two lateral, and one posterior. It is Barnard*e Barnard's opinion that different parts of the gynoecial wall initiate at different levels and at different times, and that the different parts of the gynoecial wall represent different carpels. The morphologically lowest is the anterior carpel, the second is the posterior carpel, and the two 50 lateral carpels are the most distal. Barnard's claim is not substantiated by evidence and has been rejected by later workers. With the exception of Barnard, a l l students of grass-floret histogenesis interpret the gynoecial wall as a single, continuous, leaf-like structure. But the relation between the gynoecial wall and the ovule is a hotly debated issue. The question is: Are grass ovules terminal (stachyosporous), or are they borne on the gynoecial wall (phyllosporous)? In Oryzopsis virescens and 0. hymenoides the formation of the gynoecial wall does not use up the whole floret apex. The residual floret apex is gradually transformed into an ovule. These observations correspond with the observations of Holt (1954), Bonnett (1953, 1961), Sharman (1960b), Pankow (1962), Mehlenbacher (1970), and Maze et a l . (1971, 1972), a l l of whom interpret the grass ovule to be terminal on the floret axis. A corollary of this interpretation is that the concept of 'carpel' no longer applies to the grass gynoecium, since a carpel is defined as a phyllome that bears ovule(s). The point at issue is the criterion by which an organ may be judged to be terminal. This question is discussed in some detail in connection with the author's review of Klaus' work, as i t concerns Klaus' interpretation of the grass gynoecium. According to Klaus, in Hordeum distichon L. the gynoecial wall arises as a peltate carpel and the ovule is borne on the 'cross-zone' ('querzone') of the carpel (see Tro l l , 1939, for terminology). The floral apex is nearly used up in the formation of the carpel, that i s , the carpel is nearly terminal. The grass gynoecium is phyllosporous. 51 In view of the controversial reports of phyllospory versus stachyospory not only in the Gramineae but also in other Angiosperms (see, for example, Barnard, 1957b; Schultze-MoteL, 1959; Eckardt, 1957; and Pankow, 1962), the author will discuss Klaus' work in some detail and attempt to resolve the contorversy. Two points are involved: Where the gynoecial wall arises and whether a floral apex remains after gynoecial wall initiation. Klaus' statement that the floret apex is nearly used up in the formation of the gynoecial wall is very vague. The fate of the residual apex is not mentioned. His illustrations contradict his interpretation (see Klaus, 1966, Abb. 37, 38, 39). In the formation of a terminal, or nearly terminal, gynoecial wall, the corpus of the floret apex shows increased mitotic activity, and many cell divisions are seen in the apical surface initials (Brooks, 1940; Tucker and Gifford, 1966a, 1966b). Both features are not manifested in Hordeum distichon. A floret apex remains after gynoecial wall formation, and i t is gradually transformed into an ovule. If one follows Buder's criterion (1928) that for an organ to be considered terminal, i t must develop from apical i n i t i a l cells, then the grass ovule is terminal. The presence of a cross-zone ('querzone') in the grass gynoecial wall is highly questionable. According to continental European literature, the development of the Angiosperm carpel (sensu lato) is similar to the development of a peltate leaf, and where the marginal meristems of the carpel meet, they form a transverse meristem, the cross zone. Whether the 'querzone' is something new, or an extension of the marginal meristems, i t is not identifiable in the two grasses 52 studied by the author. An analysis of the development of the gynoecial wall w i l l explain why. The gynoecial w a l l i s actually i n i t i a t e d by an e n c i r c l i n g row of i n i t i a l s on the f l o r a l apex. These i n i t i a t i n g d ivisions s t a r t on the anterior side and progress to the posterior side. Growth of the gynoecial wall outwards as a l a t e r a l appendage occurs very early and i s greatest at the point where the i n i t i a t i n g divisions f i r s t appear, that i s , on the anterior side. This growth, which i s mainly through apical a c t i v i t y , occurs simultaneously with the spread of the e n c i r c l i n g i n i t i a t i n g divisions around the flanks of the f l o r a l apex. When the i n i t i a t i n g divisions reach the posterior side of the f l o r a l apex, the shape of the gynoecial wall may be likened to a cylinder that has been cut diagonally i n h a l f . From then on, the posterior rim shows active apical a c t i v i t y and the gynoecial w a l l grows upwards as a tube. In this sequence of events, i s there anything which can be said to fuse? The grass gynoecium i s interpreted by the author as a unit structure which develops from a single gynoecial primordium and which encloses a terminal ovule. I t i s i n fact acarpellate ( S a t t l e r , personal communication). Embryo sac development The ovule i s hemianatropous, bitegmic and pseudocrassinucellar. The embryo sac i s of the monosporic, 8-nucleate type. The antipodals p r o l i f e r a t e soon after the mature 8-nucleate embryo sac i s formed. These observations i n Oryzopsis virescens and 0. hymenoides are s i m i l a r to those described by Brown (1949), i n Stipa l e u t o t r i c h a , and Mehlenbacher 53 and Maze ejt a l . in species of Stipa and Oryzopsis. This seems to be the usual type of embryo sac development in the Gramineae (Davis, 1966). The only grasses that do not follow this pattern of embryo sac develop-ment are those of the Bouteloua curtipendula (Michx.) Torr. complex, described by Mohamed and Gould (1966). In these grasses, embryo sac development is of the Adoxa type, and the antipodals do not proliferate. The number of bumps in the outer integument is interesting. The presence of one bump in the outer integument in the chalazal region has been reported in Stipa hendersoni (formerly known as Oryzopsis  hendersoni Vasey) by Mehlenbacher in 1970, in tortilis (Maze et a l . , 1970), S. lemmonii (Maze et a l . , 1972) and elmeri (Maze and Bohm, 1973). In 0. hymenoides, one bump is present. In 0. virescens there are two bumps, and in the only other species of Oryzopsis studied, 0. miliacea, there are also two bumps. Fertilization in the two species appears to be different. In 0. virescens no filiform apparatus is seen. The synergids degenerate in before fertilization, but in those florets examed, one synergid persists in a degenerate form until fertilization. The site of pollen tube enbry seems to be between the persistent synergid and the embryo sac wall. In 0. hymenoides, in which each synergid has a distinct filiform apparatus, the behavior of the synergids prior to fer t i l i z a -tion is different. One synergid decreases in size, becomes densely-staining and degenerates, while the other becomes highly vacuolate and shows no visible signs of degenerating. Both synergids persist until the 2-cell proembryo stage. The pollen tube enters the embryo sac 54 via the degenerate synergid. The entry and discharge of the pollen tube into one of the synergids has been demonstrated for Vallisneria (Wylie, 1941), Zea (Diboll, 1968), Gossypium (Jensen and Fisher, 1968) and species of Stipa. Of these examples, a l l but Vallisneria have synergids in each of which a filiform apparatus is present. On the other hand, in Cardiospermum halicacabum (Kadry, 1946), each synergid has a distinct filiform apparatus but the pollen tube enters the embryo sac between the synergids and the sac wall. CONCLUSION *n Oryzopsis virescens and 0. hymenoides histogenesis of the lemma, palea, posterior lodicule and the gynoecial wall is similar, and indicates their foliar nature. The anterior lodicules differ from them in having a deeper initiation site. The interpretation of the anterior and posterior lodicules as reduced perianth structures rather than as structures 'de novo' is preferred. Developmentally the stamens are stem-like. The gynoecium consists of a unit gynoecial wall surrounding a terminal ovule. There are two styles, each of which develops from the lateral portions of the gynoecial wall. The grass gynoecium may be considered acarpellate. Embryo sac development is of the monosporic, 8-nucleate type. 55 PART II Comparative developmental studies of the floret and embryo sac in Oryzopsis virescens, 0. hymenoides, 0. micrantha, 0. kingii, and 0. asperifolia. INTRODUCTION The problems of circumscribing the genus Oryzopsis have been discussed in the general introduction. The systematic disposition of the five species studied is briefly reviewed here. Oryzopsis virescens (n = 12; Avdulov, 1928, as cited by Darlington and Wylie, 1955; Johnson, 1945a), is an Eurasian species and belongs to the Old World section Piptatherum. Its floret characters, which are described in PART I, (Figs. 1, 2, 3), f i t in well with those features which are generally employed to set Oryzopsis apart from Stipa. Oryzopsis hymenoides (n = 24, Johnson, 1945a) is in the North American section Eriocoma. It was originally known as Stipa hymenoides Roem. and Schult. (see Syst. 2: 339, 1817, for description). The generic transfer to Oryzopsis was proposed by Ricker, and was formally presented by Piper in 1906 (see volume II, page 109, U.S. Natl. Herb. Contrib., 1906). This species is widespread in arid areas throughout western North America. It resembles 0. virescens in its open panicle and its indurate lemma, but i t is distinctively set apart from 0. virescens by its sharp callus and densely villous lemma (Figs. 4, 5, 6). A sharp callus and a villous lemma are features that are used to dis-tinguish Stipa from most species of Oryzopsis. Moreover, 0. hymenoides 56 crosses spontaneously with various species of Stipa. To date, such hybrids involing eleven different species of Stipa have been reported (Johnson, 1972, and the references therein). No other species of Oryzopsis is known to hybridize with Stipa. One other supposed case of intergeneric hybridization, Oryzopsis hendersoni x Stipa lemmoni, reported hy Spellenberg and Mehlenbacher (1971), is in fact, not such. Oryzopsis hendersoni Vasey is more closely a l l i e d to Stipa than to Oryzopsis, (Mehlenbacher, 1970). In the joint publication of 1971, Mehlenbacher made a formal transfer of the species to Stipa. According to Sheerer and Johnson (1968), 0. contracta (B.L.Johnson) Shechter may have evolved through hybridization between 0. hymenoides and 0. micrantha. No other interspecific crosses have been reported in Oryzopsis. Oryzopsis micrantha, 0. k i n g i i and 0. asperifolia are North American species and are placed in the New World section Oryzopsis. The f i r s t two species are diploids (n = 11; Johnson, 1945a), while the third one is a polyploid (n = 23; Johnson, 1945a). Bowden (1960) reported a chromosome count of 2n = 48 for 0. asperifolia. Oryzopsis  micrantha is known to occur in British Columbia, Alberta, Montana, North Dakota, south to Nebraska, Oklahoma, New Mexico, Arizona and Nevada. Oryzopsis k i n g i i has a very restricted distribution. It is endemic to the meadows at high altitudes in the central Sierra Nevada mountains in California. Oryzopsis asperifolia occurs from Newfoundland to British Columbia, Maine to the Dakotas, including the Great Lakes region, and in the Rocky Mountains from Montana to New Mexico (Hitchcock, 1969). 57 The floret of 0. micrantha has an obovate glabrous lemma which i s f a i r l y indurate, an indistinct callus and a weak, untwisted and deciduous awn (Figs. 147, 148, 149). With respect to these features 0. micrantha approaches the section Piptatherum. However, the major a f f i n i t y of this species, as discussed by Johnson (1945a), is to the section Oryzopsis. Oryzopsis k i n g i i may be described as a 'borderline species' between Oryzopsis and Stipa. As indicated by the Synonymy in Hitchcock (1951), 0. k i n g i i was originally described by Bolander in 1872 as a Stipa, and was later transferred to Oryzopsis by Beal in 1896. It is so closely a l l i e d to Stipa on most characters that i t s assignment to either Oryzopsis or Stipa is largely arbitrary. Its narrow non-indurate lemma, with a sharp callus and a strong, twisted geniculate awn (Figs. 150, 151), and i t s narrow panicle and slender involute leaves are strongly suggestive of Stipa. Oryzopsis  micrantha and Oryzopsis k i n g i i represent the morphological extremes of the section Oryzopsis. At one end, 0. micrantha presents some Piptatherum features, chiefly through i t s resemblance to 0. miliacea. At the other end, 0. ki n g i i grades into Stipa. Not surprisingly, this definition of the section Oryzopsis has caused some uncertainty about the sectional and generic identity of certain species. Oryzopsis asperifolia has the largest floret as compared to the other four species. The floret has a fusiform lemma which is convolute, a swollen, blunt, pubescent callus and a straight deciduous awn (Figs. 152, 153, 154). It is reputed to be an allotetraploid, and i t s combination of Piptatherum characters (weak and flexuous awn, many-nerved glumes, broad f l a t leaves, f a i r l y indurate lemma) and Oryzopsis 58 features (glumes equal in length to lemma, f i r s t glume shorter than second glume) is used as an indication of i t s hybrid origin (Johnson, 1945a). Developmental data have not been incorporated into the systematic studies of the genus Oryzopsis. Recent work by Mehlenbacher (1970) on Stipa hendersoni (Vasey) Mehlenbacher, and by Maze et a l . (1971, 1972) on 0. miliacea, S_. t o r t i l i s , and S. lemmoni has showen that developmental data can add another facet of evidence towards a better understanding of Stipa and Oryzopsis. It is hoped that data from the present study, together with such developmental data on other species of Stipa and Oryzopsis as are available, can be used effectively as an adjunct to other characteristics in delimiting and working out the taxonomy of Oryzopsis. There are certain advantages to using develop-mental data. Studies of f l o r a l histogenesis lead to a better under-standing of the morphological features of the mature floret, for the mature form is the result of developmental processes. The analysis of growth patterns, in the early stages of development of a structure, provide more features for comparison. In mature structures early growth patterns, which are due to the frequency and plane of c e l l division, are usually obliterated by the c e l l enlargement phase of growth. MATERIALS AND METHODS The preparation of materials for study is as described in PART I for 0. virescens and 0. hymenoides. Materials of 0. micrantha and 0. asperifolia were collected from transplants growing in the experimental 5 9 plots at the University of B r i t i s h Columbia. Plants of 0. micrantha and 0.^asperifolia were o r i g i n a l l y from Marble Canyon Pr o v i n c i a l Park and Williams Lake respectively; both locations are i n B r i t i s h Columbia. Florets of 0. k i n g i i were collected from the high meadows i n Yosemite National Park, C a l i f o r n i a . Some l i v e plants are maintained i n the afore-mentioned experimental plots. Voucher specimens of a l l f i v e species are deposited i n the Vascular Plant Herbarium of the Botany department, University of B r i t i s h Columbia. Orientation of the sections follows that used i n PART I. OBSERVATIONS Oryzopsis micrantha (Trin. and Rupr.) Thurb. Floret organogenesis The awn-lemma primordium i s the f i r s t structure to appear on the f l o r e t apical meristem (Figs. 155, ,56). This i s followed by the palea and the stamens, which i n i t i a t e at the same time. I n i t i a t i o n of the lodicules occurs next (Fig. 157). The callus and the gynoecium are the l a s t structures to appear (Figs. 158, 159). The sequence of i n i t i a t i o n of the f l o r a l parts p a r a l l e l s that i n 0. virescens. Floret Apical Meristem The f l o r e t apex has a one-layered tunica (Fig. 155). As i n 0. virescens and 0. hymenoides, the f l o r e t apex i n 0. micrantha undergoes a displacement and re-orientation i n the course of f l o r e t development. Early i n f l o r e t development, the f l o r e t apex i s oriented more or less p a r a l l e l with the f l o r e t axis (Figs. 155, 156). With the i n i t i a t i o n 60 of the anterior stamen, the floret apex is deflected away from the axis to the posterior side of the floret (Fig. 157). The floret apex continues to be posteriorly directed during early growth of the anterior portion of the gynoecial wall (Figs. 158, 159). After i n i t i a t i n g division of the gynoecial wall have encircled the floret o apex (Figs. 160, 181), the f l r e t apex continues growth as the ovule (Figs. 161, 182), but i t becomes re-oriented to a v e r t i c a l , and then to an anteriorly-directed position. Awn-lemma Initiation and early development of the awn-lemma primordium is - similar to that in 0. virescens (Figs. 155, 156). Differentiation of the awn-lemma junction is f i r s t indicated by a constriction which is the result of greater c e l l expansion in the awn base than in the lemma apex (Figs. 158, 159). Periclinal divisions occur in the ground meristem on both the abaxial and adaxial sides of the lemma apex, forming an expanded lemma apex (Figs. 163, 164, 165). The proportion of the lemma to the rest of the floret (excluding the awn) changes drastically as the floret develops. Prior to, and up t i l l , the in i t i a t i o n of the anterior portion of the gynoecial wall, the lemma is relatively small and leaves the distal half of the floret exposed (Figs. 158, 159). By the time the ring-shaped gynoecial wall i s formed, the lemma has extended to just above the top of the stamens (Fig. 160). Further growth of the floret sees extensive elongation of the lemma, as i t overtops, (Fig. 161), and encloses, the rest of the floret (Fig. 162). These observations are similar to those 6 1 described for 0. virescens and 0. hymenoides. The upper portions of the free margins of the lemma increase in height more than the expanded lemma apex, and form two 'ears' in front of the awn (Figs. 161, 166). They are, however, smaller than the free 'ears' in 0. virescens. The mature lemma apex resembles that of 0. virescens in its double-convex shape and the presence of periclinal divisions in the abaxial and adaxial ground meristem, but the resemblance is somewhat superficial. In 0. micrantha (Fig. 167), the abaxial convexity is the result of ground meristem cell expansion rather than considerable periclinal divisions. In the adaxial ground meristem, periclinal divisions are not as extensive as in 0. virescens. Unlike 0. virescens, the adaxial protoderm has only a few isolated periclinal divisions (Figs. 165, 166, 167). At the point of attachment of the awn to the lemma, the cells are much smaller than the cells in the awn base and the lemma apex (Fig. 167). Callus Initiation of the callus is through periclinal divisions and cell enlargement in the ground meristem at the base of the lemma on the anterior side (Fig. 168). As in 0. virescens, ground meristem tissue forms the bulk of the callus (Fig. 173). Early growth of the callus involves the expansion and vacuolation of the ground meristem cells, in a direction perpendicular to the adjacent protoderm, forming a rounded protuberance (Figs. 169, 170). This is accompanied by cell expansion in the adjacent protoderm (Fig. 170). At the time of initiation of the inner integument, the ground meristem cells in the 62 middle section of the callus on the anterior side begin to expand obliquely, that i s , inward and at an angle to the adjacent protoderm (Fig. 171). The protodermal and ground meristem cells at the distal end of the callus, near the axil of the lemma, expand in a longitudinal direction, pushing the callus downward (Fig. 172). At the same time, the protodermal and ground meristem cells in the basal portion of the callus continue to expand obliquely, toward the axis of the floret (Figs. 172, 173). Consequently, the callus grows downward and inward, forming a rounded tip (Figs. 172, 173). As in 0. virescens the flo r e t - r a c h i l l a t i l t s in an anti-clockwise direction (compare Figs. 168, 169 and 171). The t i l t i n g is not as extreme as in 0. virescens so that the rounded tip of the callus in 0. micrantha is laterally directed and does not l i e on the vertical axis of the floret (Fig. 162). Two growth feactures seen in 0. micrantha but not in 0. virescens involve the protoderm of the callus and the ra c h i l l a . Some of the protodermal cells of the callus on the anterior side undergo random pe r i c l i n a l , oblique and anticlinal divisions at a time when the floret undergoes extensive elongation growth (Figs. 171, 172, 173). The protoderm of the ra c h i l l a on the posterior side elongates ant i c l i n a l l y to produce a slight projection (Fig. 173). Palea Development of the palea in 0. micrantha is similar to that in 0. virescens, except that in 0. micrantha the palea does not have a biseriate distal portion (Figs. 174, 175). 63 Lodicules Initiation and growth of the lodicules is similar to that in 0. virescens. Stages in lodicule development are shown in Figures 176, 177, 178 and 179. Stamens The development of the stamens in 0. micrantha and in 0. virescens is remarkably similar. A description of stamen development in 0. micrantha would be repetitious. Gynoecium Early stages in the development of the gynoecium are shown in Figures 180, 181. The ovule is terminal, and the gynoecial wall grows upward as a tube (Figs. 181, 182). The lateral sides of the gynoecial wall give rise to a style branch each (Fig. 183). The style branches are closer together in 0. micrantha than they are in 0. virescens, (Fig. 185; cf. Fig. 76). Proliferation of tissues ('stylar core') on the inner margins of the ovary wall brings the margins together (Figs. 184, 184a). The inner margins come in contact but do not fuse, so that the opening between the inner margins is not completely closed but remains as the stylar canal (Fig. 184). Ovule and embryo sac development  Integuments The inner integument is first seen on the upper side (Fig. 183). At this stage, the archesporial cell is not distinguishable. When the 64 inner integument appears on the lower side, the single-celled archespo-rium is distinguished by its larger size and its denser cytoplasm (Fig. 186). The outer integument is also first visible on the upper side (Fig. 187). In the chalazal region of the outer integument, a single prominent bump is formed (Figs. 189, 191). The outer integument starts to degenerate prior to fertilization. Nucellus and embryo sac development Growth patterns in the nucellus leading to increase in size of the ovule and its orientation to a hemianatropous position are similar to those observed in the previous species (Figs. 188, 191, 192). Oryzopsis  micrantha is like 0. hymenoides in that there are few periclinal divisions in the nucellar protoderm. In megasporogenesis, a linear tetrad is more commonly found than a T-shaped one. In the linear tetrad the first megaspore to abort is the micropylar. In the T-shaped tetrad, both the micropylar megaspores degenerate at the same time and are the first ones to do so. Inter-mediate stages in megagametogenesis are seen in Figures 190 and 191. The synergids have a filiform apparatus each (Fig. 193). They remain densely cytoplasmic throughout their growth (Figs. 193, 194). Starch granules are seen in the egg before fertilization (Fig. 193a). Crystals are present in the nucleoli of the polar nuclei (Fig. 193). The polar nuclei fuse before fertilization. Fertilization It was very difficult to trace the path of the pollen tube into the embryo sac. Both the synergids appear to degenerate at 65 f e r t i l i z a t i o n . One decreases in size more than the other. Chromatin-like bodies are seen in the larger degenerate synergid (Figs. 195, 195a). This seems to indicate that the pollen tube contents enter the egg via one synergid. Post-fertilization I n i t i a l l y the endosperm is also nuclear. Oryzopsis k i n g i i (Boland.) Beal Floret organogenesis The floret apical meristem, with a one-layered tunica, f i r s t i nitiates the awn-lemma (Fig. 197). This is followed by the appearance of the palea and the stamens (Fig. 198). The lodicules are the next structures to develop, and during their early gorwth (Fig. 199), but before the gynoecium i n i t i a t e s , the callus and the awn-lemma junction become evident (Figs. 205, 212). The gynoecium is the last structure to i n i t i a t e (Fig. 200). Changes in the shape of the floret during growth are shown in Figures 201 to 204. As in the other three species described previously, In 0. kingi i the floret apical meristem undergoes a displacement and re-orientation in the course of floret development (cf. Figs. 200, 201, 203, 204). Concurrently, there is a change in the porportion of the lemma to the rest of the floret (excluding the awn). 66 Awn-lemma The awn-lemma primordium in its initiation and early growth is similar to that in the previous species. Differentiation of the awn-lemma junction starts early and is indicated by periclinal divisions and cell expansion in the adaxial ground meristem of the lemma apex (Figs. 205, 206, 207, 208). On the abaxial side the protoderm is one-layered, while some cells in the ground meristem may divide peri-clinally once (Fig. 209). The mature lemma apex is quite different from that of 0. virescens, 0. hymenoides and 0. micrantha. Instead of a biconex shape as present in the other three species, the lemma apex of 0. kingii has a convex adaxial surface and a staight abaxial surface (Fig. 211). The adaxial convexity results from a combination of the expansion of the protodermal cells in an anticlinal direction, and periclinal divisions in the ground meristem (Fig. 211). Periclinal divisions are also seen in the protoderm (Figs. 210, 211). The upper free margins of the lemma do not extend above the expanded lemma apex to form 'ears'. The point of attachment of the awn to the lemma is not marked by a constriction. The awn base is approximately as wide as the rest of the awn. The absence of a group of smaller cells at the junction of the awn and lemma (Fig. 211) probably accounts for the presistence of the awn. Callus Initiation and early development of the callus is similar to that of 0. virescens, 0. hymenoides and 0. micrantha. In a l l four species 67 periclinal divisions in the ground meristem at the base of the lemma indicate the beginning of the callus (Figs. 212, 213). Also, the ground meristem forms the bulk of the tissue of the callus. In the early stages of growth the young callus does not have the hemispherical shape that is so characteristic of the callus of 0. virescens and 0. micrantha. Instead, at the time of initiation of the posterior portion of the gynoecial wall, the callus has a downward-directed, broadly acute apex, similar in shape to that of 0. hymenoides (Fig. 215) . The calluses in 0. hymenoides show some Interesting differences in their development. In 0. hymenoides elongation of ground meristem cells downward and obliquely outward is responsible for the pointed shape of the callus (Fig. 119). The protodermal cells at the tip of the callus remain small (Fig. 119). In 0. kingii ground meristem cells in the distal and central portions of the callus on the anterior side expand longitudinally downward, parallel with the adjacent protoderm (Fig. 216). The sharps point of the callus is mainly the result of oblique expansion of the protodermal cells at the tip, in a direction outward and away from the floret axis (Fig. 216). Two other developmental features are of interest. Where the callus is attached to the rachilla there are regular cell files (Fig. 217), described as tabloid cells by Maze et a l . (1971) for Stipa t o r t i l i s . On the posterior side of the rachilla adjoining the base of the floret there is a projection (Figs. 204, 218). This projection differs from a similar one in 0. micrantha in that i t is formed primarily by the 68 anticlinal elongation of ground meristem cells. Palea The palea in 0. kingii develops in the same manner as in the other three species (Figs. 219, 220, 221). At maturity, i t has a narrow base and a rather long biseriate distal portion (Fig. 222). Stamens The anther 'beards' are of protodermal origin. Elongation of the protodermal cells begins when the anterior portion of the gynoecial wall is initiated. Lodicules Initiation and growth of the lodicules is as in the other three species. In 0. kingii the lodicules are smaller. Development of the posterior lodicule i s shown in Figures 220, 221, and 222. Stages in the growth of the anterior lodicules are seen in Figures 223 and 224. Gynoecium This structure is similar to the gynoecia of the other three species. However, the initiating divisions appear on the posterior side of the floret apex at a lower level, leaving a more distinct residual apex (Fig. 226). When the posterior side of the gynoecial wall begins to grow upward, the residual apex develops directly into the ovule (Fig. 227). As in the other species, the lateral sides of 69 the gynoecial wall develop as the styles (Figs. 228, 229). The styles are situated a short distance apart from each other (Fig. 231), as ^ 9.' virescens. At anthesis, the ovarian cavity is 'closed' by the proliferation of tissues on the inner margins of the ovary wall (Figs. 230, 230a). Ovule and embryo sac development  Integuments Both the inner and outer integuments are first seen on the upper side (Figs. 232, 233, 234). The archesporial cell is distinguished early in integument development (Fig. 232). This was also noticed in 0. hymenoides. There is only one bump in the chalazal region of the outer integument. Nucellus Growth of the nucellus is as in the other three species. There are not as many periclinal divisions in the nucellar protoderm as there are in 0. virescens. Embryo sac A hypodermal cell functions as the megaspore mother cell (Fig. 234). The megaspores are commonly arranged in a linear tetrad (Fig. 235). The chalazal megaspore is the functional one (Fig. 236). Stages in megagametogenesis are seen in Figures 237, 238, 239 and 240. Figure 238 shows a frequently encountered situation in which the abortion of the three non-functional megaspores is delayed until the 2-nucleate embryo sac stage. 70 The embryo sac before the differentiation of the egg and the synergids is spindle-shaped (Fig. 240). The micropylar and chalazal ends broaden out later in growth (Fig. 241). Visible signs of differentiation in each of the synergids are the development of small, scattered vacuoles and a filiform apparatus (Fig. 241). The egg becomes highly vacuolate (Figs. 240a, 241a). The polar nuclei fuse before fertilization. Fertilization The pollen tube seems to discharge its contents into one of the synergids. Inall the florets examined the two synergids degenerate prior to fertilization. In some florets one of the synergids dis-integrates completely. Remnants of the pollen tube contents are seen in the persistent degenerate synergid (Figs. 242, 242a, 242b). In other florets, both the degenerate synergids persist, though one of them becomes appreciably smaller. Figure 243 shows the first zygotic division of the fertilized egg. Chromatin-like bodies are seen in the larger, degenerate synergid. Post-fertilization The outer integument is s t i l l prominent at the 2-cell embryo stage (Figs. 244, 244b). The endosperm is nuclear i n i t i a l l y . The antipodals form a considerable mass at this stage. The synergids are s t i l l visible after the first zygotic division (Figs. 244a, 244b). 71 Oryzopsis asperifolia Michx. Floret organogenesis This species lacks a posterior lodicule. The initiation of the floret appendages follows this sequence: lemma, palea and stamens, anterior lodicules, gynoecium and callus. This is illistrated in Figures 245, 246, 247 and 248. The slow rate of development of the lemma, with respect to the rest of the floret, is unique here. At the time initiating divisions of the gynoecium form on the posterior side of the floret apex (Fig. 248), the lemma leaves the stamens and the gynoecium exposed. Up t i l l the time of the 4-nucleate stage in embryo sac development, the lemma does not overtop the stamens and styles, but leaves half their length exposed (Figs. 249, 250, 251). Only at the 8-nucleate embryo sac stage does thelemma extend above the rest of the floret. Awn-lemma The awn-lemma primordium is the first structure to develop from the floret apical meristem (Fig. 245). The floret apex is striking in its large size. Like the other four species, the floret apex has a one-layered tunica and undergoes displacement and re-orientation in the course of floret development. Figures 247 and 252 show the awn-lemma primordium of a floret at the time of gynoecium initiation. The cells in the distal portion are larger and more vacuolate than the cells in the lower portion of the primordium. This distal portion is the future awn, and i t elongates rapidly (Fig. 248). When the posterior portion 1 72 of the gynoecial wall is initiated, the awn-lemma junction is barely perceptible, and the lemma is more or less equally wide in sagittal section (Fig. 253). Ground meristem cells on both the adaxial and abaxial sides of the lemma apex start to divide periclinally. Just before the inception of the integuments in ovule development, the lemma apex becomes expanded, (Fig. 254), a result of the previously mentioned periclinal divisions plus some cell enlargement. There are fewer periclinal divisions in the abaxial ground meristem than there are in the adaxial ground meristem. At the same time, some adaxial protodermal cells have divided periclinally once. At this stage the awn base is marked out from the lemma apex by a constriction, as a result of fewer cell divisions. The expanded portion of the lemma is spread over almost half the total length of the lemma (Figs. 249, 250, 254, 255). The convex adaxial side of the lemma apex is mainly the result of cell division in the ground meristem, forming a tissue of considerable size (Fig. 256). Just before the megaspore mother cell undergoes meiosis, the patterns of cell division on this side of the lemma apex are obliterated, as a result of cell expansion in a horizontal plane (Fig. 256). At the same time, on the abaxial side of the lemma apex, the ground meristem cells expand, parallel with the adjacent protoderm, and the growth patterns due to periclinal divisions are s t i l l obvious. The abaxial ground meristem cells at the lemma apex form a narrow strip of tissue in sagittal section as compared with the adaxial ground meristem (Fig. 256). Later, greater cell expansion in the abaxial ground meristem produces 73 a tissue as wide as the adaxial ground meristem (Fig. 257). Callus Like the awn-lemma, the callus is initiated at a later stage in floret development in comparison with the other four species of Oryzopsis that have been described. This structure is not obvious until the gynoecial wall has appeared on the posterior side of the floral apex (Figs. 258, 259). Initiation and growth of the callus is mainly through cell division in the ground meristem (Figs. 258, 259, 260, 261, 263). Early in its development the callus on the anterior side forms a rounded protuberance (Fig. 260). Some protoderm cells on the same side divide periclinally once (Figs. 259, 260, 261, 263). Continued cell divisions in the ground meristem increase the size of the rounded callus (Fig. 261). In a mature floret, the callus forms an open ring around the base of the floret, with the edges meeting on the median posterior side (Fig. 154). In off-set sections parallel with the sagittal plane, the posterior callus can be seen as a slight bulge, the result of periclinal divisions in the protoderm and the ground meristem (Figs. 262, 265). Figures 261 and 262 are of a floret at the stage when integument formation has been completed in ovule development. The initially rounded hump of the callus begins to be directed slightly downward at the megaspore mother cell stage in ovule development (Fig. 263). Some of the protodermal cells in the upper part of the callus elongate to form hairs. The protodermal cells at the base of the callus divide once or twice to form a small patch of 74 multiple protoderm (Fig. 263). Further distension downward produces an obtusely-angled callus tip (Fig. 264). The massive callus is mainly the result of the large number of cells rather than cell expansion. The basal patch of multiple protoderm on the anterior side is no longer recognisable at maturity. On the posterior side, the callus is also distended downward slightly, and a multiple protoderm is seen clearly in the rounded tip (Fig. 265). For ease of comparison, a table is presented (TABLE II) of comparable stages in awn-lemma and callus development in the five species of Oryzopsis studied, using the stage of gynoecium development as the reference. TABLE II. Reference table for comparison of stages in awn-lemma and callus development in 0. virescens, 0. hymenoides, 0. micrantha,, 0. kingii, and 0. asperifolia, using the stage of gynoecium development as a marker. The numbers refer to figures at the pertinent stages. 0. 0. 0. 0. 0. Stage of gynoecium virescens hymenoides micrantha kingii asperifolia awn- awn- awn- awn- awn-development lemma callus lemma callus lemma callus lemma callus lemma callus Initiation of lodicules 19 19 106 113 none none none none 252 258 Initiation of anterior gynoecial wall 20 26 107 114 158 168 206 213 252 258 Growth of anterior gynoecial wall 21 27 108 115 163 none 207 214 none none Gynoecial wall appears on posterior side of floret apex 22 28 109 116 164 169 208 215 253 259 Ring-shaped gynoecial wall 23 none 110 117 165 170 209 none none none Prior to integument initiation 24 29 none none none none none none 254 260 Initiation of inner integument on upper side none none 111 118 166 171 210 216 none none Complete inner integument 25 30 none none 167 172 211 217 255 261 & Ovule in orthotropous 263 position none none 112 119 none none none 218 256 76 Palea Growth of the palea is similar to that in the other four species except that in 0. asperifolia cell division ceases at a comparatively later stage (Figs. 266, 267, 268, 269). Stamens The stamens develop in the same manner as in the other species. They are hirsute-tipped at maturity. Periclinal divisions of the antherprotoderm occur at the time the anterior portion of the gynoecial wall Initiates (Fig. 266). The hairs are elongate protodermal hairs. i Lodicules Only the anterior lodicules are present in this species. Their initiation and growth is the same as in the other four species (Figs. 270, 271). At maturity they are distinctly two-nerved (Fig. 272). Gynoecium Development of the gynoecium is illustrated in Figures 266, 273, 274, 275, 276 and 277. The process is very much like that described in the other four species of Oryzopsis. There are two styles (Fig. 279). At anthesis, the ovarian locule is 'closed' by the adpression of the inner margins of the ovary wall (Figs. 278, 278a). 77 Ovule and embryo sac development  Integuments Both the inner and outer integuments initiate on the upper side first (Figs. 280, 281). The outer integument has two bumps in the chalazal region (Figs. 282, 283). Nucellus and embryo sac development Stages in nucellus and embryo sac development are shown in Figures 284, 285 and 286. Like the other species of Oryzopsis development of the embryo sac is also of the monosporic 8-nucleate type. Fertilization and post-fertilization At fertilization only one synergid was seen. The pollen tube seems to enter the embryo sac via the persistent synergid, from which dense, chromatin-like bodies appear to be extruded (Fig. 287). At the 2-cell proembryo stage, remnants of pollen tubes could s t i l l be seen. Where the pollen tube had entered the synergid, a densely-staining material was deposited (Figs. 288a, b). DISCUSSION Floret development The sequence of organ initiation in the floret of the five species of Oryzopsis studied differs only in the time of initiation of the callus. In 0. virescens, 0. micrantha and 0. asperifolia the sequence of organogeny in the floret is: (1) lemma; (2) palea and stamens almost 78 simultaneously; (3) lodicules; and (4) gynoecium and callus almost at the same time. In 0. hymenoides and 0. kingii the callus is precocious and appears before the gynoecium initiates. In those species of Stipa that have been studied developmentally, the callus also appears before the gynoecium initiates. Growth of the callus in 0. virescens, 0. hymenoides, 0. micrantha and 0. kingii is primarily due to cell enlargement. In 0. asperifolia the conspicuous collar-like callus is mainly the result of cell division. With the exception of the callus, the young stages of the developing florets prior to gynoecium development are remarkably similar in the five species of Oryzopsis studied. Comparable stages in species of Stipa and Oryzopsis studied by Maze et a l . (1971, 1972), also bear a striking resemblance to those species of Oryzopsis studied by the author. Developmental differences that are diagnostic for the species begin to appear when the gynoecial wall is initiated on the posterior side of the floret apex. The probable basis for the similarity of the early stages in floret development is this: The mature form of an organism is the result of developmental processes. Genes produce effects on the visible morphological characters of the mature organism only through their influence on development. Similarly, mutations directly alter developmental processes, and only indirectly characters. Mutations which act relatively late in ontogeny are less likely to disorganise the whole process of development, with consequential deleterious effects, than those which alter the earlier stages (Stebbins, 1950). The mutations established, therefore, will be 79 those that affect development at the latest stage for the modification of the mature structure. That part of the ontogeny which is not affected by the mutations will exhibit similarity. As a rule, embryological characters are constant within a genus, (Cave, 1953). Oryzopsis is no exception. Inter-relationships of 0. virescens, 0. hymenoides, 0. micrantha, 0. kingii, 0. asperifolia, 0. miliacea, S. lemmoni, S. hendersoni, S. t o r t i l i s , and S. richardsoni. Table III presents characters derived from developmental features in 6 species of Oryzopsis and 4 species of Stipa. Since the taxonomic problem presented by Oryzopsis involves not only the nature of the inter-relationships between its species but also its intricate relation-ship with Stipa, a comparison of Oryzopsis and Stipa species rather than a comparison of Oryzopsis species alone is more meaningful. Data on 0. virescens, 0. hymenoides, 0. micrantha, 0. kingii and 0. asperifolia are from the author's own studies; data on 0. miliacea, S_. lemmoni, S_. hendersoni and S. tortilis are from a publication of Maze et a l . (1972) . In this paper by Maze et a l . the data on S_. hendersoni were taken from Mehlenbacher (1970) and modified by them. Data on S. richardsoni are also from Maze, but unpublished. Some of the characters analyzed are similar to those analyzed by Maze et a l . (1972), but the author has deleted some of the characters in the afore-mentioned publication, and has added some new ones. For each measurable character, only one measurement was taken. The author is 80 f u l l y aware that there i s no provision for a range of variation for each measurement. However, the time-consuming nature of developmental studies mitigates against multiple measurements. For each character, data for a l l ten species were taken and the mean was calculated. A two-state coding of characters was used: those greater than the mean were treated as plus (+); those less than the mean minus (-). The percentage similarity among species was calculated using the simple matching coefficient of Sokal and Sneath r' (1963), S - x 100, sm n ' S where S g m = percentage of similarity, m = the number of character matches (positive and negative), n = the number of characters. The coefficients of similarity obtained are presented in the form of a similarity matrix in Table IV. The coefficients of similarity of 0. miliacea and S_. t o r t i l i s , and of 0. micrantha and S_. t o r t i l i s are the lowest. This i s interesting because these two species of Oryzopsis are considered to be clear-cut Oryzopsis species, and S_. t o r t i l i s is a 'typically Stipoid' species. The other similarity coefficients seem to f a l l into a pattern of continuous variation. 81 TABLE I I I . Comparison of developmental featured of Oryzopala v i r e s c e n s , 0. hymcnotdeg. 0. micrantha, 0. k t n f t l l , 0. a s p e r i f o l i a , 5* P l l l a c c a , Stlp» lemmonll, S. hendersoni!, S. t o r t i l i s and S. r i c h a r d s o n i 1. Character 0. v l r . 0. hym. 0. mic. 0. k i n . 0. asp. 0. m i l . S. 1cm*. S. hen. S. t o r . S. r l c . 1. Thickness l n number of c e l l s , of the ground meristem of the lemma apex. 19 18 12 8 14 6 19 20 21 13 2. Angle of growth, i n degrees, r e l a t i v e t o the l o n g i t u d i n a l a x i s o f the avn, i n the ground meristem on the a d a x i a l s i d e of the lemma apex. 88 125 128 88 91 65 74 85 117 83 3. Angle of growth, i n degrees, r e l a t i v e to the l o n g i t u d i n a l a x i s of the avn, l n the ground meristem on the a b a x i a l s i d e of the lemma apex. 10 56 22 80 75 68 90 27 66 73 4. Sclerenchyma mother c e l l s l n awn ground meristem. - - - - - - + - + + 5. Number of protoderm l a y e r s on the a d a x i a l s i d e of the lemma apex 4 1 2 2 2 1 5 6 7 1 6. Angle o f growth, i n degrees, r e l a t i v e to the l o n g i t u d i n a l a x i s o f the avn, In the m u l t i p l e protoderm on the a d a x i a l s i d e o f the lemma apex. 71 NA 82 77 88 NA 36 63 95 NA 7. Number of protoderm l a y e r s on the a b a x i a l s i d e of the lemma apex. 1 1 1 1 1 1 1 1 2 1 8. Number o f protoderm l a y e r s on the a d a x i a l s i d e of the awn base. 1 1 1 1 1 1 1 1 5 1 9. Number of protoderm l a y e r s on the a b a x i a l s i d e of the awn base. 1 1 1 1 1 1 1 1 2 1 10. R a t i o awn wi d t h at base/width of awn-lemma J u n c t i o n i n s a g g l t a l plane. 1.1 2.2 1.21 0.93 1.1 1.27 0.82 1.23 1.00 0.97 11. R a t i o lemma wi d t h at apex/vidth of awn-lemma j u n c t i o n l n s a g g l t a l plane. 1.8 2.0 2.0 1.2 1.1 1.45 1.39 1.73 1.0 1.08 12. R a t i o , longest c e l l ( a n t i c l i n a l plane) In protoderm on a d a x i a l 6lde of lemma apex/longest c e l l ( a n t i c l i n a l plane) i n protoderm on a d a x i a l s i d e o f lemma below apex. 0.73 1.2 1.5 3.3 1.2 1.33 3.5 3.5 0.63 1.0 13. T o t a l number of protoderm c e l l s l n m u l t i p l e protoderm at lemma apex. 24 NA 4 3 6 NA 16 31 47 NA 14. Height ( i n mm.) of a p i c a l p o r t i o n of the fre e margin of the lemma above the awn-lemma j u n c t i o n when the magaspore mother c e l l i s h o r i z o n t a l . 0.46 0.27 -0.04 0.016 0.017 0.30 0.50 0.08 0.001 0.02 15. Ridge i n lemma apex I n f r o n t a l plane. - - - - - - - + -16. Angle, i n degrees, formed by the c a l l u s t i p . 100 57 138 68 127 92 48 66 21 42 17. R a t i o area of l a r g e s t protoderm c e l l / a r e a o f l a r g e s t ground meristem c e l l i n c a l l u s . 0.43 0.19 0.48 0.17 0.65 2.46 0.73 1.25 0.39 0.54 18. Thickness l n number of c e l l s i n the ground meristem of the c a l l u s along a l i n e p e r p e n d i c u l a r to the l o n g i t u d i n a l a x i s of the c a l l u s . 6 7 4 4 12 4 9 8 7 6 19. T a b l o i d c e l l s at the attachment of the f l o r e t t o the s p l k e l e t . - - - - - - + + + + 20. Protoderm peg on the p o s t e r i o r s i d e o f the s p l k e l e t a x i s . - - • + + - + - - - -21. Protoderm peg on the p o s t e r i o r s i d e of the f l o r e t base. - - - - - + - - - -22. P e r i c l i n a l d i v i s i o n s i n c a l l u s protoderm. - - + - + + - + - -23. R a t i o palea l c n g h t / l e n g t h of a n t e r i o r g y n o e c i a l w a l l when the p o s t e r i o r w a l l of the gynoecium i s i n i t i a t e d . 1.8 2.5 2.5 1.4 1.0 2.5 1.7 1.5 0.5 1.1 24. Thickness In number of c e l l s at the p a l e a base. 15 12 8 5 7 4 12 5 4 4 25. R e l g h t , l n u, above the attachment of the a b a x i a l g y n o e c i a l w a l l , of the s i t e of i n i t i a t i o n of the p o s t e r i o r g y n o e c i a l w a l l . 0 3 3 rlO 0 100 -100 -2 -1 -1 26. Angle, i n degrees, of divergence of the ovule from the l o n g i t u d i n a l a x i s of the f l o r e t when the integuments are i n i t i a t e d . 42 41 37 40 37 85 30 40 30 42 27. Number of bumps l n the outer Integument.' 2 1 1 1 2 2 1 1 1 1 28. R a t i o l e n g t h of the ovule attachemcnt t o the p l a c e n t a / l e n g t h of the ovule c a . f e r t i l i z a t i o n . 0.59 0.85 0.76 0.83 0.93 0.40 0.84 0.61 0.63 0.67 29. F i l i f o r m apparatus. - + + + + + • + - + + 30. R a t i o l e n g t h / v l d t h embryo sac ca. f e r t i l i z a t i o n . 2.7 1.9 2.2 2.6 1.7 3.0 2.4 2.7 3.6 2.3 31. Egg s t a r c h . - - + • - + + • - - -TABLE IV. Similarity matrix ( % ) of Oryzopsis virescens, 0. hymenoides, 0. micrantha, 0. kingii, 0. miliacea, Stipa lemmoni, S. hendersoni, S_. t o r t i l i s , and S. richardsoni. 0. viresc-ens 0. hymenoi-des 0. micran-tha 0. kingii 0. asperi-folia 0. miliacea s. lemmoni s. hender-soni s. tortilis 0. hymenoides 68.96 0. micrantha 53.33 75.00 0- kingii 45.16 55.17 60.00 0. asperifolia 51.61 58.62 66.66 68.75 0. miliacea 55.17 44.82 60.71 51.72 65.51 s. lemmoni 58.06 68.96 50.00 67.74 54.83 34.48 s. hendersoni 58.06 58.62 46.66 54.83 48.38 37.93 61.29 s. tortilis 41.93 51.72 26.66 51.61 45.16 24.13 . 64.51 51.61 s. richardsoni 58.62 55.17 46.42 82.75 68.96 55.17, 65.51 55.17 62.06 CO to 83 Inter-relationships of the 10 species were analyzed using a 2-dimensional ordination. The ordination technique used is that below, attempts to extract from a matrix of similarity coefficients a spatial pattern in which the distance between two species is directly related to their degree of similarity. In other words, a high degree of similarity w i l l be represented by a low spatial separation. In this technique, the inverse of the similarity coefficient between two species is equated with linear distance. The inversions are accomplished by subtracting each coefficient of similarity from a maximum similarity value of 100. The two species which have the lowest coefficient of similarity ( = the highest inverse coefficient) are chosen as the end points of the X-axis. These two species are 0. miliacea and S_. t o r t i l i s , which are separated by a linear distance of 75.9 units. The position of any species on the X-axis is calculated by Beal's equation (cited by Gauch and Whittaker, 1972): used by Bray and Curtis (1957). This technique, which is outlined X = L + D 2 L where X = the position of the species on the X-axis, L = distance between end points of the X-axis, distance of species from the f i r s t end point (= inverse coefficient of species and 0. miliacea), 84 distance of species from the second end point (= inverse coefficient of species and S. to r t i l i s ) . To construct the Y-axis, two new end points are selected which are in close proximity on the X-axis, but which are nevertheless separated by a high degree of dissimilarity. It is important that the end points are constructed of species which are most dissimilar, otherwise the axes will not cover the total range of variability expressed by the inverse coefficients. The results of this ordination technique are presented in Figure 289. With regard to its correct generic disposition, 0. hymenoides has been somewhat of an enigma. Johnson (1972) comments that its affinities are puzzling, because morphologically, (according to Johnson, 1945a), i t resembles 0. virescens in some respects (viz., an indurate lemma and an open panicle), while its distribution and hybridization with Stipa suggest some relationship with Stipa. Developmentally the affinities of 0. hymenoides are with Stipa (Fig. 289) . Morphologically, 0. hymenoides has been deemed to be an Oryzopsis on the basis of its diffuse panicle, deciduous awn, and indurate lemma. However, these three characters are not unknown in species of Stipa. An open panicle is present in S_. richardsoni L., S_. porteri Rydb. (= Ptilagrostis porteri (Rydb) Weber) , S_. lepida Hitchc., S_. cernua Stebbins and Love: S. neomexicana (Thurb.) Scribn., S. webberi (Thurb.) D2 = Y 50 + 40--x S_. richardsoni :( 0. miliacea y 10 0. asperifolia 0. kingii x x 0. micrantha 30 -• 20 Oo hymenoides X S_.: tortilis X S_. lemmoni x X So hendersoni 0. virescens 1 1 K H 1 1 1 1 1 10 20 30 40 50 60 70 80 X FIG. 289. Ordination results to show inter-relationships of Oryzopsis virescens, 0. hymenoides, 0. micrantha, 0. kingii, 0. asperifolia, 0. miliacea, Stipa  lemmoni, S_. hendersoni, S_. t o r t i l i s , and S_. richardsoni. oo 86 Johnson, and S_. ichu Hara have deciduous awns (see Maze et a l . , 1966); S. viridula Trin., S_. lemmoni (Vasey) Scribn.r," and S_. hendersoni (Vasey) Mehlenbacher have indurate lemmas. On the other hand, 0. hymenoides possesses features that are strongly suggestive of a Stipa, viz., a sharp callus, pilose lemma, and involute leaves. Besides the 1stipoid' morphological features present in 0. hymenoides its crossibility with species of Stipa provide further positive evidence of its affinity with Stipa. Hybridization between 0. hymenoides and eleven different species of Stipa have been documented by Weber (1957), Johnson and Rogler (1943), and by Johnson (1945b, 1960, 1962, 1963). A fertile amphiploid from a cross between 0. hymenoides and S_. viridula was recovered in nature by Nielson and Rogler (1952). Johnson suggests that the chromosome number of n = 24 in 0. hymenoides could indicate its origin by amphiploidy between the 12-chromosome line of Stipa and the 12-chromosome line of Oryzopsis. Even so, its chromosome number and its postulated amphiploid origin do not preclude it from being a Stipa. A count of n = 24 in S_. splendens Trin. was reported by Love and Myers in 1947 (cited by Darlington and Wylie, 1955). Evidence from developmental, morphological and hybridization studies suggest that the affinity of 0. hymenoides is with Stipa. Oryzopsis miliacea and 0. virescens belong to section Piptatherum, which consists of a coherent group of species marked by specialization in characters that distinguish them from Stipa (reduction of callus, induration, dorsiventral flattening and increase in size of the lemma, 87 increase in length and nervation of the glumes). Morphological studies by Johnson indicate that in this section, 0. virescens forms a closely-knit group with 0. paradoxa, 0. coerulescens and 0. holciformis, while 0. miliacea is set apart from them. The results in this study show that 0. miliacea is closer to 0. micrantha than i t is to 0. virescens. There are no developmental data on other Eurasian species of Oryzopsis for comparison, but i t is not unreasonable to predict a high degree of similarity between 0. virescens, 0. paradoxa, 0. coerulescens and 0. holciformis on the basis of their development. The relationship between 0. miliacea and 0. micrantha will be discussed after a consideration of the section Oryzopsis, to which 0. micrantha belongs. The North American species of 0. micrantha, 0. pungens, 0. exigua, 0. canadensis and 0. kingii constitute diploidsof the section Oryzopsis — they form a group marked by specialization in characters (differen-tiation of the callus; development of a twisted, persistent awn; indurate lemma) which merge with the genus Stipa. At one end, 0. micrantha is detached from the other species of the same section by its resemblance to 0. miliacea; at the other end 0. kingii intergrades completely with Stipa. The three species in between reportedly form a gradated series towards Stipa. The sectional disposition of 0. micrantha was questioned by Elias (1942), who placed i t in Piptatherum. On the basis of morphological and developmental data, 0. miliacea should be removed from Piptatherum, and perhaps placed in section Oryzopsis. However, evidence from other sources must be considered 88 before such a re-arrangement is attempted. According to Johnson (1945a, 1972), the Eurasian species of Oryzopsis are diploids with a count of n = 12, and the North American diploids have a count of n = 11. However, in Oryzopsis and Stipa, i t is difficult to relate the geographic distribution with chromosome number. In genera in which polyploidy is rampant, such as Oryzopsis and Stipa, chromosome number loses its value as a guide to generic and sectional delimitation (Rollins, 1953). Furthermore, a count of n = 12 in 0. pungens has been reported by Bowden (1960). Besides the five n = 11 diploids in section Oryzopsis, there are two other species, 0. asperifolia (n = 23) and 0. swallenii (n = 17; Hitchcock and Spellenberg, 1968). The latter species i s , like 0. kingii, a borderline species between Oryzopsis and Stipa (Hitchcock and Spellenberg, 1968). Until Mehlenbacher (1970) did a thorough study of S_. hendersoni, (n = 17; Spellenberg, 1968), this species too was placed in the scetion Oryzopsis as Oryzopsis hendersoni (Vasey) . The relationships between 0. micrantha, 0. kingii and 0. asperifolia, and between them and the other seven species, compared developmentally, are expressed in Figure 289. As they stand, 0. virescens seems to be isolated from the other nine species, and there does not seem to be a discontinuity between the following nine species: 0. miliacea, 0. micrantha, 0. asperifolia, 0. kingii, S_. richardsoni, 0. hymenoides (Stipa), 0. hendersoni, S_. lemmoni and £5. t o r t i l i s . In this connection, i t is interesting to note that Hoover (1966), 89 considered 0. miliacea (L.) Bentham and Hooker to be Stipa miliacea (L.) Hoover; and to quote Hoover, "the separation of a group of species as the 'genus' Oryzopsis does violence to the relationships of the plants". CONCLUSION Data from morphology, distribution, and hybridization studies suggest that Oryzopsis hymenoides belongs to the genus Stipa. Developmentally i t also shows a high degree of similarity to Stipa. This further supports the contention that 0. hymenoides be transferred to Stipa. Morphologically, 0. miliacea is set apart from 0. virescens and other members of the section Piptatherum. Developmental studies have upheld morphological data. It is suggested that 0- miliacea be removed from Piptatherum. With the removal of 0. miliacea, 0. virescens 0. paradoxa, 0. coerulescens and 0. holciformis would form a more well-defined section. Further studies on other members in the Oryzopsis are needed before the inter-relationships between 0. micrantha, 0. asperifolia and 0. kingii can be understood. It would seem that more comprehensive studies of the genus Oryzopsis will either lead to its mergence with Stipa or at least to a redefinition of the sections of Oryzopsis. 90 LITERATURE CITED Arber, A. 1925. Monocotyledons: a morphological study. Cambridge Botanical Handbooks. Cambridge. 1926. Studies in the Gramineae. I. The flowers of certain Bambuseae. Ann. Bot. 40: 447-469. i 1927. Studies in the Gramineae. II. Abnormalities in Cephalostachyum virgatum Kurz, and their bearing on the interpretation of the bamboo flower. Ann. Bot. 41: 47-74. 1928. Studies in the Gramineael IV. I. The sterile spikelets of Cynosurus and Lamarkia. 2. Stamen-lodicules in Schizostachyum. 3. The terminal leaf of Gigantochloa. Ann. Bot. 42: 173-187. 1934. The Gramineae: a study of cereal, bamboo and grass. Cambridge Univ. Press. London. Barnard, C. 1955. Histogenesis of the inflorescence and flower of Triticum aestivum L. Australian J. Bot. 3: 1-20. 1957a. Floral histogenesis in the monocotyledons. I. The Gramineae. Australian J. Bot. 5: 1-20. 1957b. Floral histogenesis in the monocotyledons. II. The Cyperaceae. Australian J. Bot. 5: 115-128. 1958. Floral histogenesis in the monocotyledons. III. The Juncaceae. Australian J. Bot. 6: 185-198. • 1960. Floral histogenesis in the monocotyledons. IV. The Liliaceae. Australian J. Bot. 8: 213-225. Bews, J.W. 1929. The World's Grasses. Their differentiation, distribution, economics and ecology. Longmans, Green and Co., London. Boke, N.H. 1947. Development of the adult shoot apex and floral initiation in Vinca rosea L. Amer. J. Bot. 34: 433-439. ________ 1949. Development of the stamens and carpels in Vinca rosea L. Amer. J. Bot. 36: 535-547. Bonnett, O.T. 1953. Developmental morphology of the vegetative and floral shoots of maize. Univ. Illinois, Agr. Exp. Sta. Bull. 568. 91 1961. The oat plant: its histology and development. Univ. Illinois, Agr. Exp. Sta. Bull. 672. Bowden, W.M. 1960. Chromosome numbers and taxonomic notes on northern grasses. III. Twenty-five genera. Can. J. Bot. 38: 541-577. . . . Bray, J.R. and J.T. Curtis. 1957. An ordination of the upland forest communities of southern Wisconsin. Ecol. Monogr. 27: 325-349. Brooks, R.M. 1940. Comparative histogenesis of vegetative and floral apices in Amygdalus communis with special references to the carpel. Hilgardia 13: 249-306. Brown, W.V. 1949. A cytological study of cleistogamous Stipa  leucotricha. Madrono 10 (3): 65-96. Buder, J. 1928. Der Bau des phanerogamen.Vegetationspunktes und seine Bedeutung fiir die Chimarentheorie. Ber. deutsch. bot. Ges. 46: 20-21. Cannon, W.A. 1900. A morphological study of the flower and embryo of the wild oat, Avena fatua L. California Acad. Sci., Proc. 3 (Botany) 1: 329-364. Cave, M.S. 1953. Cytology and embryology in the delimitation of genera. Chronica Botanica, 14 (3): 140-153. Cheung, M. and R. Sattler. 1967. Early floral development of Lythrum  salicaria. Can. J. Bot. 45: 1609-1618. Corner, E.J.H. 1946. Centrifugal stamens. J. Arnold Arboretum, 27: 423-437. Cronquist, A. 1968. The evolution and classification of flowering plants. Houghton Mifflin, Boston. Darlington, CD. and A.P. Wylie. 1953. Chromosome atlas of flowering plants. London. Davis, G.L. 1966. Systematic embryology of the Angiosperms. John Wiley and Sons, Inc., New York, London, Sydney. Davis, P.H. and V.H. Heywood. 1965. Principles of Angiosperm taxonomy. Oliver and Boyd, Edinburgh and London. Diboll, A.G. 1968. Fine structure development of the megagametophyte of Zea mays following fertilization. Amer. J. Bot. 55: 787-806. 92 Eckardt, T. 1957. Vergleichende Studien iiber die Morphologischen Beziehungen zwischen Fruchblatt, Samenanlage und Blutenachse bei einigen Angiospermen. Neue Hefte zur Morphologie, 3: 1-91. Weimar. Elias, M.K. 1942. Tertiary Prairie grasses and other herbs from the high plains. Geol. Soc. of America special papers 41: 1-176. Fagerlind, F. 1958. Is the gynoecium of the Angiosperms built up in accordance with the phyllosporous or the stachyosporous scheme? Svensk Bot. Tidskr. 52: 421-425. Feder, N. and T.P. O'Brien. 1968. Plant microtechnique: some principles and new methods. Amer. J. Bot. 55: 123-142. Gauch, H.G. and R.H. Whittaker. 1972. Comparison of ordination techniques. Ecology. 53: 868-875. Gould, F.W. 1968. Grass systematics. McGraw-Hill, New York. Guttenberg, H.V. 1960. Grundziige der Histogenese hoheren Pflanzen. I. Die Angiospermen. Handbuch der Pflanzenanatomie. 2e Aufl. ed. Zimmerman-Ozenda. Band VIII, Teil 3. Berlin-Nikolassee. Hackel, E. 1881. Untersuchungen iiber die Lodiculae der Graser. Bot. Jahrb. 1: 336-361. 1889. Gramineae. In Engler and Prantl, Die Naturlichen Pflanzenfamilien, II. 2: 1-97. Hitchock, A.S. 1951. Manual of the grasses of the United States. 2nd ed. U.S. Dep. Agr. Misc. Publ. 200. Hitchcock, C.L. and R. Spellenberg. 1968. A new Oryzopsis from Idaho. Brittonia 20: 162-165. Hitchcock, C.L., A. Cronquist, M. Ownbey, and J.W. Thompson. 1969. Vascular plants of the Pacific North-West, I. University of Washington Press, Seattle and London. Holt, I.V. 1954. Initiation and development of the inflorescences of Phalaris arundinacea L. and Dactylis glomerata L. Iowa State Coll. J. Sci. 28: 603-621. Hoover, R.F. 1966. Miscellaneous new names for California plants. Leaflets of Western Botany, 10 (16): 337-350. Jensen, W.A. and D.B. Fisher. 1968. Cotton embryogenesis: The entrance and discharge of the pollen tube in the embryo sac. Planta 78: 158-183. 93 Johansen, D.A. 1940. Plant microtechnique. McGraw-Hill, New York. Johnson, B.L. 1945a. Cytotaxonomic studies in Oryzopsis. Bot. Gaz. 107: 1-31. 1945b. Natural hybrids between Oryzopsis hymenoides and several species of Stipa. Amer. J. Bot. 32: 599-608. 1960. Natural hybrids between Oryzopsis and Stipa. I. Oryzopsis hymenoides x Stipa speciosa. Amer. J. Bot. 47: 736-742. 1962. Natural hybrids between Oryzopsis and Stipa. II. Oryzopsis hymenoides x Stipa nevadensis. Amer. J. Bot. 49: 540-546. 1963. Natural hybrids between Oryzopsis and Stipa. III. Oryzopsis hymenoides x Stipa pinetorum. Amer. J. Bot. 50: 228-234. 1972. Polyploidy as a factor in the evolution and distribution of grasses. In The biology and utilization of grasses, ed. V.B. Younger and C.M. McKell (a volume of Physiological Ecology). Academic Press, New York and London. Johnson, B.L. and G.A. Rogler. 1943. A cyto-taxonomic study of an intergeneric hybrid between Oryzopsis hymenoides and Stipa  viridula. Amer. J. Bot. 30: 49-56. Kadry, A.E.R. 1946. Embryology of Cardiospermum halicacabum L. Svensk Bot. Tidskr. '40: 111-126. Kaplan, D.R. 1968. Histogenesis of the androecium and gynoecium in Downihgia bacigalupii. Amer. J. Bot. 55: 933-950. Kaufman, P.B. 1959. Development of the shoot of Oryza sativa L. II. Leaf histogenesis. Phytomorphology 9: 277-311. Kaussman, B. 1941. Verleichende Untersuchungen iiber die Blattnatur der Kelch-, Blumen- und Staub- Blatter. Bot. Arch.. 42: 503-572. Klaus, H. 1966. Ontogenetische und histogenetische Untersuchungen an der Gerst (Hordeum distichon L.) Bot. Jahrb. 85: 45-79. Maheshwari, P. 1950. An introduction to the embryology of Angiosperms. McGraw-Hill Book Co., Inc., New York, Toronto and London. 1961. Embryology in relation to taxonomy. In Recent advances in botany, IX th International Botanical Congress, Montreal, 1959. 2 vol. Univ. of Toronto Press, pp 679-682. 94 Maze, J. and L.R. Bohm. 1973. Comparative embryology of Stipa elmeri (Gramineae). Can. J. Bot. 51: 235-247. Maze, J., N.G. Dengler, W.R. Hildreth, and R. Myatt. 1966. The reduced awns of some members of the genus Stipa and their possible relationship to dispersal mechanisms. Amer. J. Bot. 53: 632. Maze, J., L.R. Bohm, and L.E. Mehlenbacher, Jr. 1970. Embryo sac and early ovule development in Oryzopsis miliacea and Stipa  t o r t i l i s . Can. J. Bot. 48: 27-41. Maze, J., N.G. Dengler, and L.R. Bohm. 1971. Comparative floret development in Stipa tortilis and Oryzopsis miliacea (Gramineae). Bot. Gaz. 132: 273-298. Maze, J., L.R. Bohm, and C.E. Beil. 1972. Studies on the relationships and evolution of supraspecific taxa utilizing developmental data. I. Stipa lemmoni (Gramineae). Can. J. Bot. 50: 2327-2352. McCoy, R.W. 1940. Floral organogenesis in Frasera carolinensis. Amer. J. Bot. 27: 600-609. Mehlenbacher, L.E. 1970. Floret development, embryology, and systematic position of Oryzopsis hendersoni (Gramineae). Can. J. Bot. 48: 1741-1759. Merxmuller, H. and P. Leins. 1967. Die Verwandtschaftsbeziehungen der Kreuzblutler und Mohngewachse. Bot. Jahrb. 86: 113-129. Michaux, A. 1803. Flora Boreali- Americana Vol. 1. Paris, p. 51. Mohamed, A.H., and F.W. Gould. 1966. Biosystematic studies in the Bouteloua curtipendula complex. II. Megasporogenesis and embryo sac development. Amer. J. Bot. 53: 166-169. Moeliono, B.M. 1970. Cauline or carpellary placentation among the dicotyledons. Koninklijke van Gorcum, N.V. Assen. Nielson, E.L., and G.A. Rogler. 1952. An amphiploid of x Stiporyzop-sis. Amer. J. Bot. 39: 343-348. Pankow, H. 1959. Histogenetische Untersuchungen an der Plazenta der Primllaceen. Ber. deutsch. bot. Ges. 72: 111-122. 1962. Histogenetische Studien an der Blviten einiger Phanerogamen. Bot. Stud. Heft. 13: 1-106. Pankow, H. and H.V. Guttenberg. 1959. Studien uber die Anlage der Achselknospen und Blattprimordien bei Graminnen. Planta 52: 629-643. 95 Philipson, W.R. 1935. The development of the spikelet in Agrostis  canina L. New Phytol. 34: 421-436. 1934. The morphology of the lemma. New Phytol. 33: 359-371. Pilger, R. 1954. Das System der Gramineae. Bot. Jahrb. 76: 281-384. Pohl, R.W. 1968. How to know the grasses. W.C. Brown Co., Iowa. Rohweder, 0. 1963. Anatomische und histogenetische Untersuchungem an Laubsproszen und Bluten der Commelinaceen. Bot. Jahrb. 82: 1-99. Rollins, R.C. 1953. Cytogenetical approaches to the study of genera. Chronica Botanica 14: 133-139. Roth, I. 1959. Histogenese und Morphologische Deutung der Plazenta von Primula. Flora 148: 129-152. Rowlee, W.W. 1898. The morphological significance of the lodicules of grasses. Bot. Gat. 25: 199-203. Rudiger, W. 1939. Die Sproszvegtationspunkte einiger Monocotylen. Beitr. Biol. Pflanz. 26: 401-443. Satina, S. and A.F. Blakeslee. 1943. Periclinal chimeras in Datura in relation to the development of the carpel. Amer. J. Bet. 30: 453-462. Sattler, R. 1966. Towards a more adequate approach to comparative morphology. Phytomorphology 16: 417-429. _________ 1967. Petal inception and problem of pattern detection. J. Theoretical Bat. 17: 31-39. Schultze-Motel, W. 1959. Entwicklungsgeschichtliche und vergleichend-morphologische Untersuchungen im Bliitenbereich der Cyperaceae. Bot. Jahrb. 78: 129-170. Schuster, J. 1910. Ueber die Morphologie der Grasbliite Flora 100: 213-266. Sharman, B.C. 1943. Tannic acid and iron alum with safranin and orange G in studies of the shoot apex. Stain Technol. 18: 105-111. 1945. Leaf and bud initiation in the Gramineae. Bot. Gat. 106: 269-289. 96 1960a. Development of the inflorescence and spikelets of Anthoxanthum odoratum L. New Phytol. 59: 60-64. 1960b. Developmental anatomy of the stamen and carpel primordia in Anthoxanthum odoratum L. Bot. Gaz. 121: 192-198. Shechter, Y. and B.L. Johnson. 1968. The probable origin of Oryzopsis contracta. Amer. J. Bot. 55: 611-618. Singh, V. and R. Sattler. 1972. Floral development of Alisma triviale. Can. J. Bot. 50: 619-627. Sokal, R.R. and P.H.A. Sneath. 1963. Principles of numerical taxonomy. W.H. Freeman, San Francisco. Soma, K. 1958. Morphogenesis in the shoot apex of Euphorbia lathyris. J. Fac. Sci. Tokyo Univ. Sect. III. Bot. 7: 199-256. Spellenberg, R.W.. 1968. Notes on Oryzopsis hendersoni (Gramineae). Madrono, 19: 283-286. Spellenberg, R.W. and L.E. Mehlenbacher. 1971. Anatomical and cytological studies of an intergeneric hybrid, Oryzopsis  hendersoni x Stipa lemmoni (Gramineae). Can. J. Bot. 49: 1565-1574. Sprotte, V.K. 1940. Untersuchungen liber Wachstum und Nervatur der Fruchbliitter. Bot. Arch. 40: 463-506. Stebbins, G.L. 1950. Variation and evolution in plants. Columbia Univ. Press. _________ 1956. Cytogenetics and evolution of the grass family. Amer. J. Bot. 43: 890-905. 1972. The evolution of the grass family. In The biology and utilization of grasses, ed. Younger, V.B. and CM. McKell. (a volume of Physiological Ecology). Academic Press, New York, London. Tepfer, S.A. 1953. Floral anatomy and ontogeny in Aquilegia formosa var. truncata and Ranunculus repens. Univ. California Pub. Bot. 25: 513-648. Thompson, J.M. 1937. On the place of ontogeny in floral enquiry. Publ. Hart. Bot. Lab. 17: 3-20. Troll, W. 1939. Die morphologische Natur der Karpelle. Chronica Botanica 5: 38-41. 97 True, R.H. 1893. On the development of the caryopsis. Bot. Gaz. 18: 212-226. Tucker, S.C. 1959. Ontogeny of the inflorescence and the flower in Drimys winteri var. chilensis. Univ. California Pub. Bot. 30: 257-336. Tucker, S.C. and E.M. Gifford. 1966a. Organogenesis in the carpellate flower of Drimys lanceolata. Amer. J. Bot. 53: 433-442. 1966b. Carpek development in Drimys lanceolata. Amer. J. Bot. 53: 671-678. Verzar-Petri, G. and G. Baranyai-Szentpetery. 1960. Beitrage zur Entwicklungsgeschichte des Gynazeums von Datura stramonium L. Acta Biol. Ac. Sci. Hung. 11: 155-174. Weatherwax, P. 1942. Morphology of the spikelet of Oryzopsis hymen hymenoides. Amer. J. Bot. 29 (Suppl): 195. Weber, W.A. 1957. A new intergeneric natural hybrid involving Oryzopsis and Stipa (Gramineae). Rhodora 59: 273-277. Weier, CE. and H.M. Dale. 1960. A development study of wild rice, Zizania aquatica. Can. J. Bot. 38: 719-739. Wylie, R.B. 1941. Some aspects of fertilization in Vallisneria. Amer, J. Bot. 28: 169-174. 98 APPENDIX For ease of reference, the figures which are from the same floret are grouped together: 0. virescens (Trin.) Beck. 8, 15; 9, 18, 31, 51; 10, 19, 32, 40; 11, 20, 26, 33, 57; 12, 21, 27, 34, 41, 58; 13, 24, 29, 56, 67, 68, 80; 22, 28, 35, 42, 62, 63, 207; 25, 36a,b,c, 38, 44; 30, 37, 43, 81; 39, 48, 55; 46, 52; 47, 54; 60, 61; 64, 65, 66, 79; 69, 70, 71, 72; 73, 82. j). hymenoides (Roem. & Schult.) Ricker 101, 106, 113, 120; 99 APPENDIX (continued) 102, 107, 114, 121; 103, 109, 116, 122, 125, 131; 104, 110, 117, 132; 108, 115, 130; 111, 118, 123, 123a, 124, 126, 129; 127, 128. 0. micrantha (Trin. & Rupr.) Thurb. 158, 159, 168; 160, 164, 169, 177, 181; 161, 165, 170, 175, 178, 182; 163, 174; 166, 171, 183; 167, 172; 176, 180, 0. kingii (Boland.) Beal. 199, 205, 212, 219; 200, 206, 213, 220; 201, 215, 221, 223; 202, 209, 227, 228; 203, 210, 216, 222, 224, 229; 207, 214; 101 FIGS. 1-6. Floral parts. Figs. 1, 2, 3, Oryzopsis virescens; f i g . 1, spikelet; f i g . 2, floret side view; f i g . 3, floret posterior view. Figs. 4, 5, 6, 0. hymenoides; f i g . 4, spikelet; f i g . 5, floret side view; f i g . 6, floret posterior view. Line represents ltnm.; a, awn; c, callus; e, ears at apex of lemma; 1, lemma; p, palea. 100 APPENDIX (continued) 208, 226; 204, 218; 211, 217. asperifolia Michx. 247, 252, 258, 266; 248, 253, 259, 267, 270; 249, 254, 260, 268, 271; 251, 257; 255, 261, 262; 256, 263, 269; 264, 265; 273, 274. 102 1D*> FIGS. 8 - 14. Stages In floret development in 0. virescens. Fig. 8, awn-lemma initiation (unlabeled arrow). Fig. 9, palea and stamen initiation. Fig. 10, lodicule initiation. Fig. 11, gynoecium initiation. Fig. 12, spikelet during growth of anterior portion of gynoecial wall (unlabeled arrow shows awn-lemma junct on). Fig. 13, floret prior to integument initiation. Fig. 13 a, b, c, cross-sections of floret at levels shown in fi g . 13. Fig. 14, floret at megaspore mother cell stage. Lines represent 0.05 mm.; a, awn; ag, anterior gynoecial wall; a l , anterior lodicule; c, callus; 1, lemma; p, palea; p i , posterior lodicule; s, stamen. 104 FIGS. 15 - 20. Stages in floret development in 0. virescens. Fig. 15, awn-lemma initiation (unlabeled arrow). Fig. 16, early awn-lemma growth. Fig. 16a, young spikelet at approximately the same stage as fig . 16. Fig. 17, awn-lemma at later stage. Fig. 18, palea and stamen initiation. Fig. 19, lodicule initiation. Fig. 20, initiation of gynoecium. Lines represent 0.01 mm.; a, awn; ag, anterior gynoecial wall; agl, anterior glume; a l , anterior lodicule; c, callus; 1, lemma; p, palea; pc, procambium; pgl, posterior glume; p i , posterior lodicule; s, stamen; t-, outer tunica layer. 106 no, FIGS. 21 - 25. Stages in awn-lemma development in 0. virescens. Lines represent 0.01 mm.; unlabeled arrows indicate junction between awn and lemma; e, ears at apex of lemma; pc, procambium; s, stamen; vt, vascular tissue. 108 109 FIGS. 26 - 30. Stages in callus development in 0. virescens. Lines represent 0.01 mm.; unlabeled arrows indicate axils of lemma; pc, procambium. 110 FIGS. 31 - 39. Stages in floret development in 0. virescens. Figs. 31, 32, palea i n i t i a t i o n and early growth. Fig. 33, posterior lodicule i n i t i a t i o n and developing palea. Fig, 34, young flower and palea. Fig. 35, young flower and palea at i n i t i a t i o n of posterior gynoecial wall. Fig. 36, palea and posterior lodicule. Fig. 36 a, b, c, portions of palea at levels indicated in f i g . 36. Figs. 37, 38, developing posterior lodicule. Fig. 39, cross-section of posterior lodicule. Line in f i g . 36 represents 0.05 mm., a l l other lines represent 0.01 mm.; ag, anterior gynoecial wall; a l , anterior lodicule; p, palea; pc, procambium; pg, posterior gynoecial wall; p i , posterior lodicule; s, stamen. 112 113. FIGS. 40 -56. Stages in stamen and anterior lodicule development in 0. virescens. Figs. 40 - 44, anterior lodicule. Figs. 45-49, cross-sections of anterior lodicule; f i g . 45, initiation of anterior lodicules (unlabeled arrows); f i g . 46, spread of initiation in posterior direction (unlabeled arrow); f i g . 47, anterior lodicule with distinct posterior margin; f i g . 48, posterior and anterior margins distinct; f i g . 49, mature anterior lodicule. Fig. 50, whole structure of anterior lodicules, adaxial view. Fig. 51, stamen initiation (unlabeled arrow). Figs. 52 - 55, cross-sections of stamens, stippling indicates sporogenous cells. Fig. 56, lateral stamen, tangential section. Line in f i g . 50 represents 0.1 mm., a l l other lines represent 0.01 mm.; circle in figs. 46 - 49 represents anterior axis of floret; a l , anterior lodicule; am, anterior margin; pc, procambium; pm, posterior margin. 114 FIGS. 57 - 72. Stages in gynoecium development in 0. virescens. Fig. 57, initiation of anterior gynoecial wall. Fig. 58, early growth of anterior gynoecial wall. Fig. 59, frontal section of young gynoecium Figs. 60, 61, cross-sections of young gynoecium, 14 ja apart; f i g . 61 is at a higher level than f i g . 60. Figs 62, 63, adjacent sections 14 ju apart. Figs. 64, 65, 66, adjacent serial sections 14 M apart, showing a distinct posterior gynoecial wall ('querzone'), and the beginning of style branches. Figs. 69, 70, 71, 72, serial cross-sections of young gynoecium, 14 ju. apart, and of increasing height from left to right. Lines represent 0.01 mm.; ag, anterior gynoecial wall; fm, floret apical meristem; l g , lateral gynoecial wall; pg, posterior gynoecial wall; p i , posterior lodicule; sb, style branch. 116 >I7 FIGS. 73 - 77. Stages in gynoecium development in 0. virescens. Fig. 73, style branch with developing stigmatic hairs (unlabeled arrows). Figs. 74, 74a show the same cross-section; fig. 74a, floret. Figs. 75, 75a are from the same cross-section; f i g . 75, 'stylar core'; fi g . 75a, ovary; circle in both figs, represents the anterior axis of the floret. Fig. 76, ovary and anterior lodicules, abaxial view. Fig. 77, a series of transverse sections from the base upwards of a young floret at the megaspore stage; the sections are 14 ja apart. Lines in figs. 74a, 75a, 76 represent 0.1 mm., a l l other lines represent 0.01 mm.; agt, anterior gynoecial trace; alt, anterior lodicule trace; c l , cleft; lgt, lateral gynoecial trace; pgt, posterior gynoecial trace; s, style; sh, stigmatic hair; sr, stylar core region; st, stamen; stt, stamen trace. Stippling shows procambium. 118 119 FIGS. 78 - 84. Early stages in ovule and embryo sac development in 0. virescens. Figs. 78, 79, development of posterior portion of gynoecial wall (unlabeled arrows); stippling indicates densely cytoplasmic c e l l s . Fig. 80, young gynoecium. Fig. 81, i n i t i a t i o n of inner integument. Figs. 82, 82a, same section* i n i t i a t i o n of outer integument. Figs. 83, 83a, same section, megaspore mother c e l l stage. Fig. 84, meiosis II. Lines in figs. 82a, 83a represent 0.05 mm.; a l l other lines represent 0.01 mm.; double-headed arrows indicate growth patterns; agt, anterior gynoecial trace; fm, floret apical meristem; i i , inner integument; mme, megaspore mother c e l l ; o i , outer integument; ov, ovule trace (= posterior gynoecial trace). 120 FIGS. 85 - 90. Stages i n ovule and embryo sac development i n 0. virescens. Figs. 85, 85a, megaspore stage, from the same section; f i g . 85, ovule; f i g . 85a, gynoecium. Fig. 86, development of second bump i n outer integument at megaspore stage. Fig. 87, T-shaped tetrad, with one degenerated megaspore (indicated by arrow and s o l i d black). Figs. 88, 88a, same section at functional megaspore stage; f i g . 88, ovule; f i g . 88a, whole gynoecium. Fig. 89, 2-nucleate stage. Figs. 90, 90a, same section at 4-nucleate stage; f i g . 90, ovule; 90a, whole gynoecium. Lines i n f i g s . 85a, 88a, 90a represent 0.05 mm., a l l other l i n e s represent 0.01 mm.; double-headed arrows indicate growth patterns; b^, f i r s t bump; b^, second bump; i i , inner integument; o i , outer integument. 122 FIGS. 91 - 98. Stages in embryo sac development in 0. virescens. Figs. 91, 91a, same section at 8-nucleate stage; f i g . 91, ovule; f i g . 91a, whole gynoecium. Figs 92, 93, same section at differentiated 8-nucleate stage; f i g . 92, embryo sac; f i g . 93, egg and synergids. Figs. 94, 95, same embryo sac prior to f e r t i l i z a t i o n ; f i g . 94, embryo sac; f i g . 95, egg and synergids. Figs. 96, 97, same section at f e r t i l i z a t i o n ; f i g . 96, egg apparatus and surrounding nucellus and integuments; f i g . 97, ovule. Fig. 98, embryo sac with proembryo and nuclear endosperm. Line in f i g . 91a represents 0.05 mm., a l l other lines represent 0.01 mm.; double-headed arrows in f i g . 91 indicate growth patterns; c, chromatin-like bodies; e, egg; h, homogenous dense staining material; i i , inner integument; o i , outer integument; pe, proembryo. 124 FIGS. 99 - 105. Stages in floret development in 0. hymenoides. Fig. 99, lemma initiation (unlabeled arrow). Fig. 100, palea initiation (unlabeled arrow). Figs 101 - 105, outline diagrams of entire florets; f i g . 101, at lodicule initiation (unlabeled arrow indicates site of posterior lodicule initiation); f i g . 102, at initiation of anterior portion of gynoecial wall (unlabeled arrow); fig . 103, at initiation of posterior portion of gynoecial wall (unlabeled arrow); f i g . 104, prior to integument initiation; f i g . 105, megaspore mother cell stage. Lines in figs. 99, 100 represent 0.01 mm., a l l other lines represent 0. 05 mm.; a, awn; a l j , awn-lemma junction; c, callus; g l , glume; 1, lemma; p, palea. 1X1 FIGS. 106 - 112. Stages i n awn-lemma development i n 0. hymenoides. Lines represent 0.01 mm.; arrows i n d i c a t e j u n c t i o n between awn and lemma; pc, procambium; s, stamen. 128 /49 FIGS. 113 - 119. Stages in callus development in 0. hymenoides. Lines represent 0.01 mm.; unlabeled arrows indicate axils of lemma; agl, anterior glume; pgl, posterior glume. IZI FIGS. 120 - 128. Stages in floret development in 0. hymenoides. Figs. 120, 121, flower and palea. Fig. 122, posterior lodicule and palea. Fig. 123, posterior lodicule and palea. Fig. 123a, tip of palea shown in fi g . 123. Fig 124, posterior lodicule. Figs. 125, 126, development of anterior lodicule. Fig. 127, whole structure of anterior lodicules, adaxial view. Fig. 128, same lodicules as f i g . 127, but abaxial veiw and with ovary attached. Line in f i g . 123a represents 0.05 mm., lines in figs. 127 and 128 represent 1.0 mm., all other lines represent 0.01 mm.; ag, anterior portion of gynoecial wall; p, palea; pc, procambium; p i , posterior lodicule; s, stamen. 132 FIGS. 129 - 134. Stages in floret development in 0. hymenoides. Fig. 129, portion of stamen, anther 'beard' i n i t i a t i o n . Fig. 130, growth of anterior portion of gynoecial wall. Fig. 131, gynoecial wall appears on posterior side (unlabeled arrow). Fig. 132, young gynoecium. Fig. 133, cross-section of top of ovary to show stylar core tissue, heavy line (arrow) indicates 'closure' of locule. Fig. 133a, cross-section of gynoecium; figs. 133 and 133a are of the same section, c i r c l e in both indicates anterior axis of floret. Fig. 134, whole structure of ovary. Line in f i g . 133a represents 0.1 mm., i n f i g . 134, 0.05 mm., a l l other lines represent 0.01 mm.; ag, anterior portion of gynoecial wall; fm, floret apical meristem; pg, posterior portion of gynoecial wall. 134 FIGS. 135 - 140. Ovule and early embryo sac development in 0. hymenoide hymenoides. Figs. 135, 135a, same section at inner integument i n i t i a t i o i n i t i a t i o n on upper side (unlabeled arrow in f i g . 135); f i g . 135, ovule; f i g . 135a, gynoecium. Fig. 136, ovule at i n i t i a t i o n of outer integument on upper side (unlabeled arrow). Figs. 137, 137a, same section at i n i t i a t i o n of outer integument on lower side (unlabeled arrow); f i g . 137, ovule; f i g . 137a, gynoecium. Figs. 138, 138a, same section at megaspore mother c e l l stage; f i g . 138, ovule; f i g . 138a, gynoecium. Figs 139, 139a, same section at megaspore stage; f i g . 139, ovule; f i g . 139a, gynoecium. Fig. 140, functional megaspore. Lines in figs. 135a, 137a, 138a, 139a represent 0.05 mm., a l l other lines represent 0.01 mm.; double-headed arrows indicate growth patterns; solid black in f i g . 140 represents aborted megaspores; i i , inner integument; o i , inner integument. 136 t$7 FIGS. 141 - 143. Stages in embryo sac development in 0. hymenoides. Fig. 141, 4-nucleate stage. Figs. 142, 142a, from same same embryo sac at 8-nucleate stage; f i g . 142, ovule; f i g . 142a, egg. Figs. 143, 143a, 143b, from same embryo sac prior to f e r t i l i z a t i o n ; f i g . 143, embryo sac; f i g . 143a, egg and one synergid; f i g . 143b, ovule. Lines represent 0.01 mm.; dsy, degenerating synergid; e, egg; fa, fi l i f o r m apparatus; i i , inner integument; nv, nucleolar vacuole; o i , outer integument; p, polar nucleus; psy, persistent synergid; sy, synergid. 138 /39 FIGS. 144 - 146. Fe r t i l i z a t i o n and early post-fertilization stages in 0. hymenoides. Fig. 144, adjacent serial section, 7 ju. apart, of egg apparatus at f e r t i l i z a t i o n ; f i g . 144a, egg and surrounding nucellus and inner integument; f i g . 144b, egg and two synergids; fig s . 144c, 144d, two synergids. Fig. 145, adjacent ser i a l sections, 7 ju apart, of egg apparatus at f e r t i l i z a t i o n ; f i g . 145a, egg, degenerat degenerate synergid and surrounding nucellus and integuments; figs. 145b, 145c, egg and synergids. Fig. 146, 2-cell proembryo stage. Lines represent 0.01 mm.; dsy, degenerate synergid; h, homogenous dense staining material; i i , inner integument; o i , outer integument; psy, persistent synergid; pt, pollen tube; sn, chromatin-like bodies; s?, sperm nuclei?. 140 f FIGS. 147 - 154. Floral parts. Figs 147, 148, 149, Oryzopsis  micrantha; f i g . 147, spikelet; f i g . 148, floret side view; f i g . 149, floret posterior view. Figs. 150, 151, 0. k i n g i i ; f i g . 150, spikelet; f i g . 151, floret side view. Figs. 152, 153, 154, 0. asperifolia; f i g . 152, spikelet; f i g . 153, lemma side view; f i g . 154, lemma posterior view. Line represent 1.0 mm., a l l drawings at the same magnification; a, awn; c, callus; 1, lemma; p, palea. 142 FIGS. 155 - 162. Stages in floret development in 0. micrantha. Fig. 155, awn-lemma initiation (unlabeled arrow). Fig. 156, early awn-lemma growth. Fig. 157, Initiation of posterior lodicule (unlabeled arrow). Fig. 158, initiation of anterior portion of gynoecial wall. Fig. 159, spikelet same section as fig . 158. Fig. 160, spikelet at initiation of posterior gynoecial wall (unlabeled arrow). Fig. 161, floret prior to integument initiation. Fig. 162, floret at megaspore mother cell stage. Lines in figs. 159, 160, 161, 162,represent 0.05 mm., a l l other lines represent 0.01 mm.; a, awn; a l , anterior lodicule; c, callus; g, glume; 1, lemma; pc, procambium; p, palea; p i , posterior lodicule; s, stamen. 144 FIGS. 163 - 167. Stages in awn-lemma development in 0. micrantha. Lines represent 0.01 mm.; e, ears; pc, procambium; s, stamen; vt, vascular tissue. 146 141 FIGS. 168 - 173. Stages in callus development in 0. micrantha. Unlabeled arrows indicate axils of lemma; lines represent 0.01 mm.; vt, vascular tissue. 148 FIGS. 174 - 18.5. Stages in floret development in 0. micrantha. Figs. 174, 175, palea and posterior lodicule. Figs. 176 - 178, anterior lodicule. Fig. 179, anterior lodicules, whole structures, adaxial view. Figs. 180, 181, development of gynoecial wall. Figs. 182, 183, later stages in gynoecial wall development. Figs. 184, 184a, same section of top of ovary; f i g . 184, 'stylar core' region, heavy black line (arrow) indicates adpressed inner margins; f i g . 184a, ovary; c i r c l e in both diagrams indicates anterior axis of floret. Fig. 185, ovary and anterior lodicules, abaxial view. Line in f i g . 184 represents 0.1 mm., lines in figs. 179 and 185 represent 0.5 mm., a l l other lines represent 0.01 mm.; ag, anterior gynoecial wall; fm, floret apical meristem; p, palea; pg, posterior gynoecial wall; p i , posterior lodicule; sb, style branch; sr, 'stylar core' region. 150 1*1 FIGS. 186 - 191. Ovule and embryo sac development in 0. micrantha. Fig. 186, inner integument on upper and lower side. Fig. 187, outer integument see on upper side. Fig. 188, megaspore mother cell stage. Fig. 189, megaspore stage. Fig. 190, 2-nucleate embryo sac and surrounding nucellus. Fig. 191, 4-nucleate. All lines represent 0.01 mm.; i i , inner integument; mme, megaspore mother cell; o i , outer integument. 152 /5"3 FIGS. 192 - 196. Stages in embryo sac development in 0. micrantha. Fig. 192, 8-nucleate stage, before differentiation of egg and synergids. Fig. 193, 8-nucleate stage with differentiated egg and synergids. Fig. 193a, egg from same embryo sac as Fig. 193. Fig. 194, egg apparatus, polar nuclei about to fuse, surrounding nucellus and integuments prior to fertilization. Fig. 195, post-fertilization. Fig. 195a, fertilized (?) egg and synergids; Figs. 195 and 195a are from the same embryo sac. Fig. 196, 2-cell proembryo stage. All lines represent 0.01 mm.; e, egg; es, starch granules in egg; pt, pollen tube; sn, chromatin-like bodies. 154 FIGS. 197 - 2Q4. Stages in floret development in 0. kingii. Fig. 197, spikelet at early avm-lemma growth (unlabeled arrow). Fig. 198, spikelet at palea and stamen initiation. Figs. 199 - 204, outline diagrams of whole florets; fig. 199, early growth of lodicules; f i g . 200, at initiation of anterior gynoecial wall (unlabeled arrow); fi g . 201, at initiation of posterior portion of gynoecial wall (unlabeled arrow); fig . 202, prior to integument initiation; f i g . 203, at inner integument initiation, unlabeled arrow indicates awn-lemma junction; fi g . 204, megaspore mother cell stage. Lines in figs. 197, 198 represent 0.01 mm., a l l other lines represent 0.05 mm.; a, awn; a l , anterior lodicule; c, callus; g l , glume; 1, lemma; p, palea; p i , posterior lodicule; pz, site of palea initiation; sz, site of stamen initiation. 156 157-FIGS. 205 - 211. Stages in awn-lemma development in 0. kingii. Lines represent 0.01 mm.; unlabeled arrows indicate junction between awn and lemma; pc, procambium; st, stamen; vt, vascular tissue. I 158 FIGS. 212 - 218. Stages in callus development in 0. kingii. Figs. 212 - 217.are drawings of calluses, f i g . 218 is a drawing of the posterior side of the rachilla immediately below the attachment of the floret. Lines represent 0.01 mm.; unlabeled arrows Indicate axils of lemma; t, tabloid cells; vt, vascular tissue. FIGS. 219 - 231. Stages in floret development in 0. kingii. Fig. 219, young flower and palea. Fig. 220, young flower and palea at initiation of anterior gynoecial wall (unlabeled arrow). Figs. 221, 222, posterior lodicule and palea. Figs. 223, 224, anterior lodicule; unlabeled arrow In fi g . 223 indicates axil of anterior stamen. Figs. 227, 228 are adjacent serial sections; fig. 227, growth of posterior gynoecial wall; fig. 228, arrow indicates growth of lateral portion of gynoecial wall into a style branch. Fig. 229, young gynoecium. Figs. 230, 230a show the same cross-section; f i g . 230, 'stylar core' region at top of ovary, heavy line (indicated by unmarked arrow) shows 'closure' of ovarian locule; f i g . 230a, ovary; circle in both figs, represent anterior axis of floret. Fig. 231, anterior lodicules and ovary, abaxial view. Line in f i g . 230a represents 0.1 mm.; in fi g . 231, 0.05 mm.; all other lines represent 0.01 mm.; a l , anterior lodicule; fm, floret apical meristem; i , integument; p, palea; pg, posterior gynoecial wall; p i , posterior lodicule; s, stamen; sb, style branch; sr, 'stylar core' region. 162 S FIGS. 232 - 239. Ovule and early embryo sac development in 0. kingii. Fig. 232, inner integument initiation on upper side (unlabeled arrow). Fig 233, inner integument initiation on lower side. Fig. 234, outer integument on both upper and lower sides. Fig. 235, megaspore stage. Fig. 236, functional megaspore, solid black indicates aborted mega-spores. Fig. 237, 2-nucleate. Fig. 238, 2-nucleate stage with 3 persistent megaspores, embryo sac outlined with heavy black line. Fig. 239, 4-nucleate. Lines represent 0.01 mm.; i i , inner integument; o i , outer, integument; pm, persistent megaspores. 164 I (,6 FIGS. 240 - 244. Stages in embryo sac development in 0. k i n g i i . Fig. 240, ovule. Fig. 240a, egg, from the same embryo sac shown in f i g . 240. Fig. 241, embryo sac with fused polar nuclei. Fig. 241a, egg, from same embryo sac shown in f i g . 241. Figs. 242, 242a, 242b, adjacent serial sections 7 jn apart, at f e r t i l i z a t i o n ; f i g . 242, embryo sac; figs. 242a, 242b, egg, one synergid, and surrounding nucellus and inner integument; heavy line (arrow) indicates boundary between nucellar protoderm and inner integument. Fig. 243, zygotic division and two synergids. Figs. 244, 244a, 244b, adjacent ser i a l sections 7 ;u apart at 2-cell proembryo stage. Lines indicate 0.01 mm.; e, egg; fa, f i l i f o r m apparatus;- i i , inner integument; nv, nucleolar vacuole; o i , outer integument; pt, pollen tube; sn, nuclear-like material; sy, synergid; v, vacuole. 166 FIGS. 245 - 251. Stages in floret development in 0. asperifolia. Fig. 245, awn-lemma initiation (unlabeled arrow). Fig. 246, palea and stamen initiation. Figs. 247 - 251, outline drawings of entire florets; f i g . 247, at initiation of anterior gynoecial wall and anterior lodicule; f i g . 248, at initiation of posterior gynoecial wall,,unlabeled arrow indicates awn-lemma junction; f i g . 249, prior to integument initiation; f i g . 250, at megaspore mother cell stage, unlabeled arrow indicates awn-lemma junction; f i g . 251, 4-nucleate stage. Lines in figs. 245, 246 represent 0.01 mm., a l l other lines represent 0.05 mm.; a, awn; agz, initiation site of anterior gynoecial wall; alz, site of anterior lodicule initiation; c, callus; 1, lemma; p, palea; pgz, initiation site of posterior gynoecial wall; pz, site of palea initiation; sz, site of stamen initiation. 168 U>9 FIGS. 252 - 257. Stages in awn-lemma development in 0. asperifolia. Lines represent 0.01 mm.; unlabeled arrows mark junction of awn and lemma; pc, procambium; s, stamen; vt, vascular tissue. 170 ni FIGS. 258 - 265. Stages in callus development in 0. asperifolia. Figs. 262 and 265 are the posterior sides of the callus shown in figs. 261 and 264 respectively. Lines represent 0.01 mm.; unlabeled arrows denote axils of lemma. /7S FIGS. 266 - 279. Stages in floret development in 0. asperifolia. Fig. 266, flower and palea at initiation of anterior gynoecial wall (unlabeled arrow). Figs. 267, 268, 269, palea development. Figs. 270, 271, anterior lodicule development. Fig 272, anterior lodicules, whole structures, abaxial view. Fgis. 273, 274, adjacent serial sections of young gynoecium. Fig. 275, frontal section of young gynoecium. Fig. 276, initiation of posterior gynoecial wall (unlabeled arrow). Fig. 277, growth of posterior gynoecial wall. Figs. 278, 278a, same cross-section of top of ovary; f i g . 278, 'stylar core' region, contact of inner margins indicated by heavy line (arrow); fi g . 278a, ovary; circle in both figs, indicates anterior axis of floret. Fig. 279, ovary. Line fig . 278a represents 0.1 mm., lines in figs. 272, 279 represent 0.5 mm., a l l other lines represent 0.01 mm.; ag, anterior gynoecial wall; a l , anterior lodicule; fm, floral apex; l g , lateral gynoecial wall; p, palea; pg, posterior gynoecial wall; s, stamen; sr, 'stylar core' region. 174 ^278 17$ FIGS. 280 - 284. Ovule and early embryo sac development in 0. asperifolia. Figs. 280, 280a are of the same section; f i g . 280, ovule; f i g . 280a, gynoecium. Fig. 281, megaspore mother cell stage. Fig. 282, megaspore stage. Fig. 283, functional megaspore; solid black indicates aborted megaspores. Fig 284, 2-nucleate stage. Line in f i g . 280a represents 0.05 mm., a l l other lines represent 0.01 mm.; b^, first bump; b^i second bump; i i , inner integument; oi , outer integument. 176 177 FIGS. 285 - 288. Stages in embryo sac development and fertilization in 0. asperiflolia. Fig. 285, 8-nucleate stage. Fig. 285a, egg, from the same embryo sac as fig . 285. Figs. 286, 286a, 286b, 8-nucleate stage, prior to fertilization, a l l from the same embryo sac; f i g . 286, ovule; f i g . 286a, embryo sac; f i g . 286b, egg. Fig. 287, egg at fertilization. Figs. 288, 288a, 288b, 2-cell proembryo stage, from the same embryo sac; f i g . 288, ovule; figs. 288a, 288b, 2-cell proembryo, synergid and surrounding nucellus and integuments. Line in f i g . 288 represents 0.05 mm., a l l other lines represent 0.01 mm.; an, antipodals; e, egg; es, starch granules in egg; h, homogenous dense staining material; i i , inner integument; o i , outer integument; pe, proembryo; pt, pollen tube; sn, nuclear-like material; sy, synergid. 178 FIG. 290. Scanning electronmicrograph of young florets of Oryzopsis virescens, during early growth of the awn-lemma. 260 x. FIG. 291. Scanning electronmicrograph of young florets of 0. virescens, prior to initiation of the gynoecium. 125 x. Abbreviations for both figures: a, awn; a l , awn-lemma; fa, floret apical meristem; gl 1 , first glume; g l _ , second glume; 1, lemma. 180 FIG. 292. Scanning electronmicrograph of young florets of Oryzopsis virescens, during growth of the anterior portion of the gynoecial wall. 260 x. Abbreviations: a, awn; ag, anterior portion of gynoecial wall; as, anterior stamen; 1, lemma; p, palea; p i , posterior lodicule. 182 FIG. 293. Sagittal section of part of a young floret of Oryzopsis  virescens, during growth of the anterior portion of the gynoecial wall. 440 x. Abbreviations: ag, anterior portion of gynoecial wall; fa, floret apex; p, palea; p i , posterior lodicule. 184 FIG. 294. Transverse section at top of ovary in Oryzopsis hymenoides at anthesis. 225 x. FIG. 295. Frontal longitudinal section of ovary of 0. hymenoides at anthesis. 65 x. Abbreviations for both figures: sc, 'stylar core1 region; sr, stigmatoid tissue; vb, vascular bundle. 186 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0101004/manifest

Comment

Related Items