Planta (BEE.) 84, 199--214 (1969)

Time Relationships of Sporopollenin Synthesis Associated with Tapetum and Microspores in Lilium J. H~SLOP-H~RISO~r a n d I t . G. DICKII~SON Institute of Plant Development, University of Wisconsin, Madison l~eeeived October 14, 1968

Summary. The development of the sporopollenin orbicules (Ubisch bodies) on the tapetal cells of Lilium begins while the spores are still enclosed in the meiotic tetrads. Spherosome-like structures, the pro-orbicular bodies, accumulate in the vicinity of the plasmatemma early in the tetrad period, and are extruded into the space within the degenerating inner walls of the tapetaI ceils. There they acquire a coating of sporopollenin, the accretion continuing until after the release of the spores from the tetrads. Some orbieules remain attached to the plasmalemma by stalks. Synthesis of a material of the general class of sporopollenin begins in the primexine of the young spore in the mid-tetrad period, again outside of the cell membrane, but within the callose tetrad wall. A general scheme for sporopollenin formation in the anther is given. According to this, (a) precursors are synthesised both in the young spores and in the tapetum, and released into the extracellular space; and (b) polymerisation occurs on initiating sites outside of the cell membranes, these sites being the surface of the proorbicular bodies and of the special lamellae concerned in exine growth. Synthesis of sporopollenin in the ant]mr is virtually complete before the main synthesis of the pigmented pollen coat substances (Pollenkitt) begins in the tapetum. It is therefore improbable that the carotenoids produced in the final phase of metabolic activity in the tapetum can be sporopollenin precursors. Introduction I n t h e a n t h e r s of flowering p l a n t s w i t h t h e s e c r e t o r y t y p e of t a p e t u m , bodies with chemical p r o p e r t i e s similar to those of t h e sporopollenin of t h e pollen-grain exine a c c u m u l a t e on t h e inner, loeular faces of t h e t a p e t a l cells d u r i n g t h e l a t t e r p e r i o d of pollen d e v e l o p m e n t . These bodies assume t h e form of flat plates, cups or spheroids, a n d range in size from l 0 ~ in m a x i m u m dimension t o below the resolution l i m i t of t h e l i g h t microscope. T h e y h a v e been n o t e d f r o m t i m e to t i m e b y cytologists from t h e m i d - n i n e t e e n t h c e n t u r y onwards, a n d accounts of t h e m b a s e d on light microscopic o b s e r v a t i o n h a v e been p u b l i s h e d b y Se~NARF (1923), KRJATCESZCKO (1925), U]~ISCH (1927), KOS•ATH (1927) a n d P r (1932). The s t r u c t u r e s h a v e been referred to as " U b i s c h b o d i e s " in some r e c e n t accounts, b u t t h e n a m e is i n a p p r o p r i a t e since U]~IscH was n o t their discoverer. R e c e n t l y BANERJEE (1967) a n d ROWLEY a n d ERDTMAN 14 Planta (Berl.), ]3d. 84

200

J. HESLOP-HAI~RISO~r~nd H. G. DICKINSON:

(1967) have referred to them as "orbicules ". Since this is an unambigu~ ous name, it will be adopted here in place of the term "plaques" used in earlier papers (HESLOI'-I{AI~RISON,1962, 1963a, b). Since the advent of electron microscopy, the tapetal orbicules have been re-investigated in m a n y species, and the chemical identity of the material of their walls with the sporopollenin of the exine has been reasserted from the evidence of response to electron stains and resistance to acetolysis (RowLEu et al., 1959). Early studies of orbicule development as observed at the fine-structural ]eve] were made by ROWLEY (1963) for Poa annua, and by HEsLoI~-HAm~ISO~ (1962, 1963a, b) for Cannabis sativa and Silene pendula. More recently, accounts o# fuller investigations using modern preparation procedures have been published by SKVARLA and LARSO~ for Zea mays (1966), GARN/EL for Oxalis spp. (1967), ROWLEu and ERDTMAN for Salix and Populus spp. (1967) and EGIILIN and GonwI~ for Helleborus ]oetidus (1968 a) while some structural details have been added b y ROI~AND (1967) for Ficaria ranunculoides. I n the study of orbicule development in Cannabis and Silene (H]~sLoPHAI~I~ISO~, 1962, 1963a, b), a difference in the mode of formation was observed. I n Cannabis, spherical or ellipsoidal bodies were detected just within the protoplast, which on extrusion at the plasmalemma acquired electron dense caps over one pole. I t was assumed t h a t the angle of sectioning determined whether these caps presented an " 0 " or a " U " appearance in electron micrographs. I n Silene, the tapetal sporopollenin was found to occur as plates, each with several internal cavities of spherical form. They were preceded during early development of the pollen tetrads by aggregates of particles of similar electron density following permanganate fixation at the surface of, or just within, the protoplasts of the tapetal cells. Since particles of similar electron density were present at this time in tapetal mitochondria, it was suggested t h a t the small superficial particles had their origin in these organelles. The more recent investigations of CAI~NIEI~(1967) and particularly of ROWLEu and ERDTMA2r (1967) and EGHLIN and GoDwI~ (1968a) show t h a t the interpretations offered for Cannabis and Silene have to be revised. In Oxalis and Helleborus, the precursors of the orbicules have been found to be spherical bodies, described by CAI~NIEL as "Lipidtropfen" and EcI~I~IN and GoI)wI~ as " g r e y bodies" or "pro-Ubisch b o d i e s ' , and these have been shown to accumulate initially beneath the plasmalemma of the tapetal cells. These pro-orbicular bodies are extruded into the space below the tapetal cell wall, itself undergoing dissolution at this time, and there progressively acquire a sporopollenin coat. As the layer of sporopollenin increases in thickness, fusions occur between neighbouring bodies to give compound orbicules, present both in Oxalis and Helleborus. As pointed out by EcI~LI~ and GODWIN (1968a),

Time Relationships of Sporopollenin Synthesis in Lilium

201

the earlier o b s e r v a t i o n s on Cannabis are r e a d i l y i n t e r p r e t e d in t e r m s of t h e process described b y t h e m for Helleborus. I t is r e a s o n a b l e to a s s u m e t h a t t h e spherical bodies seen w i t h i n t h e p r o t o p l a s t of t h e Cannabis t a p e t u m correspond to the " p r o - U b i s e h " bodies of ECttLIN a n d GODWIn, a n d t h e s o m e w h a t different a p p e a r a n c e a t m a t u r i t y is e v i d e n t l y due to a less vigorous deposition of sporopollenin, giving a limited, cups h a p e d e n s h e a t h m e n t , a n d r e s t r i c t i n g t h e possibilities of fusion to give c o m p o u n d orbieules. T h e m a t u r e sporopollenin plaques of Nilene are, in contrast, u n d o u b t e d l y c o m p o u n d , a n d correspond closely in s t r u c t u r e to those described for Oxalis a n d Helleborus. I n view of t h e ontogenetic o b s e r v a t i o n s on these two genera, t h e suggestion t h a t t h e plaques in Silene originate from m i t o c h o n d r i a l particles can no longer be sustained. A t present, no suggestion can be m a d e as to t h e n a t u r e a n d f u n c t i o n of t h e m i t o c h o n d r i a l particles, b u t i t m a y be n o t e d t h a t t h e y h a v e been o b s e r v e d s u b s e q u e n t l y in t h e m i t o c h o n d r i a of t h e t a p e t u m of o t h e r species following p e r m a n g a n a t e fixation, a n d have been i l l u s t r a t e d also b y EDWAt~DSON (1962) in t h e t a p e t a l cells of Zea ~weys. I n a c u r r e n t s t u d y of pollen d e v e l o p m e n t in Lilium, t h e o n t o g e n y of t h e t a p e t a l orbieules has been followed, a n d a n a t t e m p t has been m a d e to w o r k o u t t h e t i m i n g of t h e i r f o r m a t i o n in r e l a t i o n to o t h e r e v e n t s in the spores a n d t a p e t u m , a n d to trace t h e i r fate u p to t h e dehiscence period. The p r e s e n t p a p e r gives a n a c c o u n t of this work, a n d a t t e m p t s to set o u t a general scheme showing t h e r e l a t i o n s h i p of spore a n d t a p e t a l sporopollenin synthesis. Materials and Methods Observations were made on greenhouse-grown plants of Lilium Iongi/lorum T~u~B. and L. henryi BAKEm Selected anthers were segmented, the developmental stage established from an acetic-orcein smear of one fragment, and the remainder fixed for 4 hr at 2--4 ~ in 3% glutaraldehyde buffered at pH 6.8 in 0.1 M phosphate buffer. The fixed segments were washed in buffer, and post-fixed for 3 hr at 2--4 ~ in 2% OsO4 buffered at pH 6.8 in 0.1 N phosphate buffer. After washing, the material was dehydrated through an alcohol series and embedded in an ArMditeEpon mix. Sections were post-stained in saturated uranyl acetate. For scamping electron microscopy, several preparation procedures were tested: a) Fresh anthers were segmented with a razor blade and slices cut manually to expose the required surfaces. These were dried out at room temperature both with and without prior alcohol or ether washing, and the slices mounted on the object holder with Dueo cement. b) Selected fresh anther segments were encased in Lipshaw embedding matrix M1 and cooled rapidly to --20 ~ in the quicMreeze block of an IEC model CTD cryostat. Thick sections with the desired orientation were then cut in the cryostat and brought to room temperature. The embedding matrix was then washed away and the sections dried out at room temperature before mounting. c) GlutarMdehyde-fixed material was washed, and treated as in a). Gold coating was carried out in the standard manner. The inner anther wall is a very satisfactory object for scanning electron microscopy, and satisfactory preparations were obtained with all techniques tried. 14 ~

202

J. HESLOP-HAI~RISONand H. G. DICKINSON: Observations

On the basis of the appearance of the spore walls, the tetrad period in lily can be divided into three intervals: (a) Y o u n g tetrad, from the cleavage of the d y a d to the first evidence of the primexine; (b) early p a t t e r n e d period, when probacula first become apparent; a n d (c) late patterned period, when all the major features of primexine p a t t e r n are established, b u t before the beginning of break-up of the tetrads (H~sLorHARmSON, 1968c). The interval between the release of the spores f r o m the tetrads and anthesis can be divided into four further periods using reasonably conspicuous criteria, as follows: (d) Y o u n g spore, from Table. Deposition o/ sporopollenin in association with tapetum and spores o/Lilium longi/lorum Sequence of events, based upon data from cultivars "Croft" and "Nellie Dean" grown under glass during spring and early summer, 1967. Period

Approximate bud-length range (mm)

23.5--24.5 a) Dyad cleavage to early primexine formation

Tapetum

Spores

First appearance of pro-orbicular bodies; wall dissolution in progress

Primexine matrix (cellulose) appearing

b) Early patterning

24.0--25.0

Extrusion of proorbicular bodies; first sporopollenin deposition

Lamellae generated in probacula

e) Late patterning

24.5--25.5

Extrusion complete; active sporopollenin deposition

Sporopollenin-like material deposited in probacula; nexine 1 forming

d) Spore release to vacuolation

25--34

Continued growth of orbicules; carotenoid synthesis begins in spherosomes

Nexine 2 forming; beginning of intine depositition and sexine thickening

e) Vaeuolation to beginning of dissolution of tapetum f) Tapetal dissolution to pollen mitosis g) Pollen mitosis to anthesis

3244

Orbicule growth ends; carotenoidcontaining spherosomes coalesce Pollenkitt released

Sexine growth completed; starch and other reserves accumulating Deposition of pigmented Pollenkitt in exine cavities Differentiation of vegetative and generative cells completed

42--54

52--138

Orbieules become apposed to inner anther wall

Time I~elationships of Sporopollenin Synthesis in Lilium

203

Fig. 1. Tapetal orbicules shortly after extrusion of the pro-orbicular bodies during the interval b) of the Table. The cores (C) bear a thin film of sporopollenin. The orbieules lie within the inner, loeular wall of the cell (W), which is undergoing dissolution. • ca. 9,800 Fig. 2. Orbieules during the interval e) of the Table. The cores (C) show a somewhat higher electron density than that characteristic of the pro-orbicular bodies. Two reveal stalks (arrows) connecting them with the outer zone of the cytoplasm, now markedly Mveolate. • ca. 9,500 tetrad break up to the beginning of vacuolation; (e) mid-spore, from vaeuolation to beginning of t a p e t u m dissolution; (f) old spore, from tapetal dissolution to pollen mitosis, and (g) gametophyte, from pollen mitosis to anthesis. The sequence of events in the formation of the orbieules is summarised in the Table in relation to this reference scale and other developmental changes in the spores. Taken with the illustrations of Figs. 1 - - 6 this table conveys most of the facts, b u t some points are amplified in the following notes.

a) Origin o[ the Pro-Orbicular Bodies. I n lily, these appears first in the outer layers of the cytoplasm of the tapetal cells as globuli with only moderate affinity for osmium tetroxide with the present preparation technique. A t the time when t h e y are being formed, the plasmalemma

204

J.H~sLoP-Ha_aRIso~ and H. G. DICKINSON:

Fig. 3. Surface of a tapetal cell in slightly oblique section, at a stage slightly later

than Fig. 1. The plasmalemma is deeply invaginated, and in some processes proorbicular bodies (PO) can be seen. A population of lipid globuli (S) is developing deeper within t~he cell. These are unpigmented at this time, but later become eentres for carotenoid accumulation. • ca.. 10,200

shows convolutions which foreshadow the development of the alveolate zone (I-II~SI~OI'-HAI~I~ISON,1968b) seen in Figs. 2, 3 and 4. The bodies do not appear to be membrane-invested, and no association with ribosomal endoplasmic reticulum of the kind described by EcI~nI~ and GoDwin (1968a) in Helleborzvs has been observed although it is not excluded that such a relationship m a y exist during developmental stages not encountered. I n electron-microscopic appearance, the proorbicular bodies are not distinguishable by any consistent criteria from spherosomes as these have been typified by Ft~EY-WYssLI~G et al. (1963) and JACKS et al. (1967).

b) Extrusion. There is no evidence of the passage of th e pro-orbicular bodies through prcsistent gaps in the plasmalemma. They appear first to be pressed out into evaginations from the cell surface (Fig. 3), later becoming free. Many sections show a persistent stalk linking the mature

Time l~dla~iollships of Sporopollenin Synthesis in .bilium.

205

Fig. 4. Orbieules during the early vacuotated spore period. Orbicule growt,h is essentially complete a~ this time, and eamtenoids are detectable in t.he lipid globuli of ghe tapetum. • ca. 11,700

orbicule with the surface of the originating cells (Figs. 2 and 4). Such a stalk would be left shouId the link ~.~ith the plasmalemma fail to break following evagination of the pocket containing the pro-orbicular body. An orion of this kind would seem to predicate the existence of a unit membrane around tahe extruded body, representing a, fragment of the plasmalemma. Yet investing membranes cannot be distinguished in lily (ef. Fig. 1), and it is apparent from the account given by ECnLIN and GODWIN (1968a) for Helleborus that no investing membranes were observed in their study once the pro-orbicular bodies left the cell. In contrast, I~OWL.EY and Et~DT~IAN (1967) give evidence to suggest that the corresponding body in Populus is enclosed in a membrane. Possibly in lily the membrane is withdrawn from the distal part of the proorbicular body after its extrusion, while continuing to adhere to the proximal part. Images like that of the orbicule towards the Nght in Fig. 2 show that sometimes only one hemisphere becomes exposed to receive a sporopollenin coat. This is evidently the normal behaviour

206

J. H~sLoP-HA~RlSO~and H. G. DmK~SO~r

Fig. 5. Ste1~osean electron micrograph of the inner surface of the ta,petum shortly before anther dehiscence, The contraction of the anther wall has brought many orbieules into contact. • ca. 15,000 in Cannabis, where thin sections usually show C- or U-shaped profiles (H~sLoP-HA~msoN, 1962). c) Sporopollenin Deposition. There is no accretion of sporopollenin while the pro-orbicular bodies lie within the cytoplasm of the tapetM cells, confirming the point firmly made by E c m ~ ~ and GODWI~~ (1968a) for HelIeborus. However, accumulation begins immediately after extrusion, continuing thereafter throughout the period of sporopoltenin synthesis in the anther. The product is an irregular sphere, showing the pro-orbicular body as a core when seen in section (Figs. 2, 4, 5 and 6). Compound bodies comparable with those formed in Oxalis (CAI~IEr,, 1967) and Silene (H]~sLOP-ItARRISON, 1963a) are not formed in Lilium. In Helleborus, ECHLIN and GODWIN (1968a) noted that in the later period of orbicule growth lamellae originating near the plasmalemma became apposed; this has not been observed in the present study. d) Relationship between Pro-Orbicules and Other Lipid Globuli. Shortly after the formation of the pro-orbieules in the outer zone of the cytoplasm of the tapetal cells, a second population of globuli begins to develop (Fig. 3). These are the bodies concerned with the synthesis of the " P o l l e n k i t t " applied to the exine during the later stages of pollen matm~ and their characteristics and fate halve been described elsewhere (H~SLO]~-IIAx~SON, 1968b). These globuli differ from their first

Time Relationships of Sporopollenin Synthesis in Lilium

207

Fig. 6. Stereoscan electron micrograph of the inner surface of a dehisced anther, showing pollen grains lying on the carpet of orbieules. • ca. 4,050 emergence in electron density following osmium tetroxide fixation, being always appreciably lighter than the pro-orbicules. Initially they are unpigmented, but carotenoids (in Lilium lonig/lorum, principally ~.-earotene-5,6-epoxide) begin to accumulate in them shortly after the release of the spores from the tetrads. In view of the very interesting suggestion by BnOOKS and SHAW (1968) that sporopollenin arises from carotenoid precursors, some significance attaches to the timing of Wo-orbieule formation and sporo-

208

J. HESLOP-HAI~RISOI, r and H. G. DICKINSON:

pollenin deposition in relation to the first synthesis of carotenoid pigments in the tapetum. As the Table (p. 202) shows, the pro-orbicular bodies arc effectively all extruded from the t a p e t n m before the end of the tetrad period, by when sporopollenin deposition is actively in progress. At this time, the carotenoid concentration in the anther is low and accounted for principally by the contribution of chloroplasts in the wall (I-IEsLOP-~IARRISON,1968b, d). Significant amounts of tapetumsynthesised carotenoid can be detected first when the buds are of a length of 28--30 mm. The orbicules then appear as in Fig. 4; they are in fact approaching their final size. Exine growth is also well advanced. The tapetal carotenoids are not released until the final dissolution of the tapetum, some considerable time after both orbicules and exine have completed growth. At no earlier period are pigments detectable in the thecal fluid. The carotcnoids appearing in the t a p e t u m during its final phase of activity cannot therefore be concerned with sporopollenin synthesis. I t is not excluded, of course, that colourless precursors of a similar character are the true sporopollenin precursors, and indeed this interpretation has attractive features. Discussion The observations recorded here for Lilium longi/lorum taken with those on other species mentioned in the Introduction permit a number of generalisations to be made about the synthesis of sporopollenin in the flowering-plant anther: 1. Compounds of the general class of sporopo]lenin are not normally formed withifi cells but only on surfaces external to the plasmalemma, or just possibly on special cvaginations of it. This is true both for the spores themselves and the cells of the tapetum. 2. Sporopollenin m a y be deposited on the surface of globuli (the pro-orbicular bodies) or on specialised lamellae (RowLgY and SOUTItWORTH, 1967). Lamellae m a y be concerned in the development of the tapetal orbicu]es in some species (EcHLINand GODWIN,1968a; I~OWLEY and EUDTMAN, 1967), and it seems t h a t in lily at ]east they are involved in sporopollenin deposition in all parts of the exine - - sexine, nexine 1 and nexine 2 (Dickinson and ItESLoP-HARP~ISON,1968). 3. Sporopol]enin precursors must be present in the thecal fluid, and must also be released into the extracellnlar space between the spore p]asmalemma and the callose tretad wall. This conclusion arises from the fact t h a t deposition occurs in two sites, at the spore surfaces and at the tapetal surface, at a time when the spores are isolated from the tapetum by the callose tetrad wall, a wall known to have limited permeability (HssLoP-HAnuISON and MACKENZIE,1966).

Time Relationships of Sporopollenin Synthesis in L i l i u m

I Spore cytoplasm Secreted sporopollenin precursors

in Ev the tetrad

Lamellae ] formed at the plasmalemma

209

I Tapetal cytoplasm Secreted sporopollenin precursors in the theeal fluid

Pro-orbicular bodies extruded at the plasmalemma

Extraeellular I p~176 ~, pollenin tOprimexinegiVe sporoof the I

Extraeellular 1 / polymerisation k / to give mature exine

Events in the thecal fluid

Fig. 7. Scheme for deposition of sporopollenin in the orbicules and exine, based upon the evidence from L i l i u m and other genera with a tapetum of the parietal, secretory type 4. There is no evidence t h a t sporopollenin is ever re-mobilised metabolically once deposited in any site within the anther. This indicates t h a t flowering plants lack enzymes capable of degrading this class of wall polymers. A scheme can now be drawn up for the various processes involved in sporopollenin deposition in the anther, taking into account events both in spores and t a p e t u m ; this is given in Fig. 7. This scheme appears to be compatible with all of the principal accounts of spore wall and orbieule development of recent date, notably those of SKVARLA and LA~SO~ (1966), GoDwIN et al. (1966), CARN~L (1967), ROLAXD (1967), ROWLEY and ERI)TMAX (1967) and EC~LTN and GODWIN (1968a, b). I t embodies the important principle t h a t so far as the morphogenetic aspect of exine growth is concerned, control mainly resides in the spores, not in the t a p e t u m (for recent discussions of this matter, see GoDwI>-, 1968; and I-I~sLoP-HAt~mSON, 1968a, 1969). This conclusion has been affirmed b y all recent workers, and is inherent in EeDTMaN'S entertMning concept (1966) of an "endo-eontractor" responsible for launching the initial phase of exine construction and an " e x o - c o n t r a c t o r " concerned with the supply of materials for later stages of construction, particularly for the thickening and reinforcement of the sexine.

210

J. HESLoP-HAlu~ISONand I-I. G. DleXlNSON: Radially oriented i lamellae, localised in probaeula I and derivatives ,

> Sexine

Tangentially oriented lamellae, distributed over the spore surface

>~exine I

Tangentially oriented lamellae, general, but with --~ Nexine 2 some concentration near the colpus Fig. 8. Scheme for extracellular lamella formation in the developing exine of Lilium. The lamellae provide the sites for sporopollenin deposition. The strata on the right are initiated by the production of lamellae formed at or near the plasmalemma in the sequence shown on the left of the diagram

W i t h regard to the early - - and obviously critical - - morphogenetic events themselves, it would certainly represent a considerable step forward could it be shown to be generally true t h a t lamellae of the type described b y ROWLEY and SOUTHWOI~TI~ (1967) and GODWIN et al. (1967) for the nexine 2 are concerned in the early definition of all layers of the exine, as t h e y a p p a r e n t l y are in lily (DIcKI~SOl~ and HESI~OPI-I~RISO~, 1968). Such a generalisation would make it possible to formulate a t h e o r y for exine morphogenesis placing the whole onus of pattern-establishment u p o n the plasmalemma of the spore and the cytoplasmic structures associated with it, including elements of the endoplasmic reticulum, where these are implicated in the localisation of features such as the probacula (HESLOP-HAI~I~ISO~, 1963 a, b; SKVAI~I,A and LA~so~, 1966). The role of the plasmalemma could be envisaged as being primarily to generate lamellae, the potential surfaces for sporopollenin accretion, according to a p r o g r a m m e worked out in the outer zone of the spore cytoplasm. P a t t e r n would then be determined basically b y the sites of lamella formation, the orientation given to the lamellae, and the timing of their production over different parts of the spore surface. I n lily, the sequence would be as in Fig. 8. Current objections to a scheme of this nature arise from the possibility t h a t the lamellae observed in the early lily probacula m a y not be directly homologous with those participating in the growth of the nexine layers, and from observations like those of ROWLEY and DUNBAl~ (1967) which have been interpreted as indicating the origin of sporo-

Time Relationships of Sporopollenin Synthesis in Lilium

211

pollenin-bearing membranes de novo0 at some distance from the plasmalemma. A further proposition embodied in the scheme of Fig. 7 is t h a t sporopollenin precursors are synthesised both in the spores and the t a p e t u m . For some species (e.g., Helleborus /oetidus: ECItLIN and GODWIN, 1968 b), the evidence suggests t h a t the synthesis in the two localities begins essentially simultaneously, since the materials of the orbicules and the probacula retain the same electron density throughout. I n others, including Silene pendula and Cannabis sativa (HEsLoP-HAI~mSON, 1963a, b) and Oxalis spp. (CAR~IEL, 1967), the accretion of sporopollcnin in the patterned component of the primexine begins later t h a n it does a r o u n d the pro-orbicules, if the electron density of the material deposited at the two sites can be accepted as an indicator. Yet another implication of the scheme of Fig. 7 is t h a t sporopollenin deposition is not necessarily dependent on the presence of fresh lamellae, or on special qualities of the surfaces of pro-orbicular bodies, b u t t h a t it m a y take place on pre-existing sporopollenin surfaces. This is apparent from the facts t h a t the orbicules grow b y accretion of new materials on their outer faces, and t h a t the thickening of sexine after release from the tetrads occurs without accession of new lamcllae. This being so, it follows t h a t the special role of the lamellae, and presumably the surfaces of the pro-orbicular bodies, is to provide initiating loci. Polymerisation begins there, and subsequently continues without arrest while substrate remains available. I t is possible t h a t this initiation is enzymatic; but this explanation does not have to be invoked, since the property involved m a y simply be the capacity to capture and hold monomers while chain-building begins. Obviously this capacity is not shared by all surfaces within the anther, since deposition does not occur over the whole area of the tapetal plasmalemma exposed to the thecal fluid, nor over the degenerating remains of the tapetal walls. A very good analogy for these aspects of sporopollenin synthesis is available in the extracellular synthesis of the very different wall polymer, cellulose, by the bacterium Acetobacter xylinum. BEN-HAYYI~ and O ~ I ) (1965) have shown that what is released from the bacterial cell is a diffusible, "prefibrous" form of cellulose; this is seemingly synthesised by enzymes anchored at the cell membrane, the glucose residues being added there from sugar nucleotides. The cellulose microfibrils arise by polymerisation remote from the cell, and through a process akin to crystallisation not involving enzymatic activity. The addition of sodium carboxymethylcellulose to the medium has important effects. It is incorporated into the microfibrils by a process of co-crystallisation, and its presence leads to an orientation of the microfibrils, an effect due, according to BEN-ttAYYI~ and CHAD,to the pattern of charge distribution it induces. It is easy to sketch a corresponding picture for sporopollenin, based upon the proposition that monomers or oligomers are released into the extracellular spaces in the anther, diffusing to preferred assembly sites where the final polymerisation occurs.

212

J. HESnoP-HA~RISONand H. G. DICKINSON:

Turning finally to the role of the tapetal orbicules, it seems that little can yet be said of any general applicability. At the time of anther dehiscence, they form a carpet over the inner surface of the anther wall, and in lily the contraction of the wall that accompanies dehiscence brings them into close contact with each other (Fig. 5). This carpet is both hydro- and lipophobie (HI, sLot-HARrison, 1968b), and even although considerable amounts of Pollenkitt are present in the lily anther, the pollen grains do not become cemented to the wall, and can be shaken free quite readily. This perhaps suggests a function in pollen dispersal (H]~sLo~-I-IA]cRISON, 1968a). The remarkable observations of BANEI~JEE (1967) on the sporopollenin tapetal membranes of grasses point to the same possibility. Here the orbieules are connected in a network which must provide an unwettable surface from which the pollen grains can easily be detached. Zusammenfassung Die Entwieklung der Sporopollenin-Orbicula (Ubisch-K6rper) in den Tapetenzellen yon Lilium beginnt wenn die Sporen noch in den meiotischen Tetraden zusammen gesehlossen sind. Spharosomenahnliehe Gebilde, die Pro-OrbieularkSrper, werden im friihen Tetradenstadium in der NiChe des Plasmalemmas angehi~uft und dann in den Raum innerhalb der degenerierenden Innenwi~nde der Tapetenzellen ausgesto~en. Dort werden sie mit einer Hiille aus Sporopollenin versehen; dieser Vorgang dauert noch an, wenn die Sporen aus dem Tetradenverband freigesetzt worden sind. Einige der Orbicula bleiben mit dem Plasmalemma mittels Stielchen verbunden. Im mittleren Tetradenstadium setzt in den Primexinen der jungen Sporen die Synthese eines Materials ein, welches der allgemeinen Klasse der Sporopollenine angeh6rt; auch dieser Vorgang verli~uft auBerhalb der Zellmembran, aber innerhalb der aus Callose bestehenden Tetradenwand. Ein allgemeines Schema ffir Sporenpollenblldung in der Anthere wird vorgeschlagen. Danach werden a) Vorstufen sowohl in der jungen Spore als auch im Tapetum synthetisiert und in den extracellularen l%aum ausgesehieden, und b) finder Polymerisation an spezifischen Stellen aul~erhalb der Zellmembran, und zwar den Oberfli~ehen der ProOrbiculark6rper und der fiir das Wachstum der Exine maBgeblichen, speziellen Lamellen, start. Die Synthese des Sporopollenins in der Anthere ist praktisch beendet bevor die Synthese der pigmentierten Substanzen der Po]lenhaut (Pollenkitt) im Tapetum einsetzt. Es ist daher unwahrseheinlich, dal~ die Carotinoide, die in den Endstadien der Stoffweehselt~itigkeit der Tapetenzellen gebildet werden, Vorstufen des Sporopollenins sein kSnnen.

Time Relationships of Sporopollenin Synthesis in Lilium

213

We are indebted to the Science Research Council and to the Graduate School of the University of Wisconsin for support of this work. We wish also to thank Engis Equipment Co., Morton Grove, Illinois, for the use of a Cambridge Instruments Stereoscan Mk. I I a electron microscope, and Mr. F. RossI of Engis Equipment for his skilled collaboration. References

BA~-EI%JEE,V. C.: Ultrastructure of the tapetal membranes of grasses. Grana Palynol. 7, 365--377 (1967). B:s~-HAYYIM, G., and I. OHAD: Synthesis of cellulose by Acetobacter xylinum. VIII. On the formation and orientation of bacterial cellulose fibrils in the presence of acid polysaccharides. J. Cell Biol. 25 (2), 191--207 (1965). B~ooKs, J., and G. S~Aw: Chemical structure of the exine of pollen walls and a new function for the earotenoids in nature. Nature (Lond.) 219, 532--533 (1968). CAI%I~IEL,K. : Licht- und elektronenmikroskopische Untersuchungen der Ubischk6rperentwicklung in der Gattung Oxalis. 0st. bot. Z. 114, 490--501 (1967). DICKInSOn, H. G., and J. HESLOP-HA~RISOlV: A common mode of deposition for the sporopollenin of sexine and nexine. Nature (Lond.) 220, 926--927 (1968). ECItLIN, P., and H. GODWIN: The ultrastructure and ontogeny of pollen in Hellehorus/oetidus L. I. The development of the tapetum and Ubisch bodies. J. Cell Sci. 3, 161--174 (1968a). - - - - The ultrastructure and ontogeny of pollen in Helleborus/oetidus L. II. Pollen grain development through the cMlose special wall stage. J. Cell Sei. 3, 175--186 (19685). EDWARDSON,J. R. : Cytoplasmic differences in T-type male sterile corn and its maintainer. Amer. J. Bot. 49, 184--187 (1962). ERDTMAN,G. : Sporoderm morphology and morphogenesis. A collocation of data and supposition. Grana Palynol. 6, 319--323 (1966). FI%EY-WYssLING,A., E. GRIESHABER, and K. MUttLETHALER: Origin of spherosomes in plant cells. J. Ultrastruct. Res. 8, 506--516 (1963). GODWIN, I-I. : The origin of exine. New Phytologist 67, 667--676 (1968). - - P. ECttLIlV, and B. CHAPMAN: The development of the pollen grain wall in Ipomoea purpurea (L.) Roth. Rev. Palaeobotan. Palynol. 3, 181--195 (1967). HESLOI~-HARRISON,J.: Origin of exine. Nature (Lond.) 195, 1069--1071 (1962). - - Ultrastructural aspects of differentiation in sporogenous tissue. Syrup. Soc. exp. Biol. 17, 315--340 (1963a). - - An ultrastructurM study of pollen wall ontogeny in Silene ]gendula. Grana Palynol. 4, 7--24 (1963b). -Pollen wall development. Science 161, 230--237 (1968a). TapetM origin of pollen coat substances in Lilium. New Phytologist 67, 779--786 (19685). - - Wall development within the microspore tetrad of Lilium longi/lorum. Canad. J. Bot. 46, 1185--1192 (1968c). - - Anther carotenoids and the synthesis of sporopollenin. Nature (Lond.) 220, 605 (1968d). - - The emergence of pattern in the cell walls of higher plants. In: The emergence of order in developing systems (M. LOCKE, ed.) (in press). New York: Acad. Press 1969. - - , and A. MACKENZIE: Autoradiography of (2-14C)-thymidine derivatives during meiosis and microsporogenesis in Lilium anthers. J. Celt Sci. ~', 3 8 7 ~ 0 0 (1967).

214

HESLOI~-HAt~RISON and DIcI~INso~: Time Relationships

JAGXS, J. T., L . Y . YAwsu, and A. M. ALTSGHUL: Isolation and characterisation of peanut spherosomes. Plant. Physiol. 42, 585--597 (1967). KOSMATH, L. : Studien fiber das Antherentapetum. 0st. bot. Z. 76, 234--241 (1927). KRJATGHERrKO, M. D. D. : De l'activit6 des chondriosomes pendent le d6veloppement des grains de pollen et des cellules nourrici~rcs du pollen dans Lilium croceum CI~AIX. Rev. g6n. Bot. 37, 193--211 (1925). PY, G.: l~echerches cytologique sat l'assis nourriei~re des mierospores et les microspores des plantes vasculaires, l~ev. g6n. Bot. 44, 3 1 6 ~ 7 3 (1932). ROLAND, F. : Differentiation du sporoderme chez Ficaria ranunculoides MOENGH. Observation et 6volution de "Corps d'Ubisch". Pollen Spores 9, 4 1 5 ~ 2 5 (1967). ROWLEu J. R. : Ubisch body development in Poa annua. Grana Palynol. 4, 25--36 (1963). - - , and A. DUNBAR." Sources of membranes for exine formation. Svensk bot. T. G1, 4 9 ~ 6 4 (1967). - - , and G. ERDTigAN: Sporoderm in Populus and Salix. Grana Palynol. 7, 518--567 (1967). - - K. MUHLETttALER,and A. FREY-WYSSLIIqG: A route for the transfer of materials through the pollen wall. J. biophys, biochem. Cytol. 6, 537--538 (1959). - - , and D. SOUTHWO~T~: Deposition of sporopollenin on lamellae of unit membrane dimensions. Nature (Lond.) 213, 703--704 (1967). SCHNARF, K.: Kleine Beitrgge zur Entwicklungsgeschichte der Angiospermen. IV. Uber das Verhalten des Antherentapetums einiger Pflanzen. Ost. bot. Z. 7~, 242--245 (1923). SKVA~LA, J. J., and D. A. LARSON: Fine structural studies of Zea mays pollen. I. Cell membranes and exine ontogeny. Amer. J. Bot. 53, 1112--1125 (1966). U~ISG~, G. v. : Zur Entwicklungsgeschichte der Antheren. Planta (Berl.) 3, 490-495 (1927). Professor J. HESLOP-HAI~RISON Institute of Plant Development, University of Wisconsin Birge Hall 430 Lincoln Avenue Madison, Wisconsin 53706, USA