Origin of the coleorhiza in cycad seedlings and its structural homology with that of the Poaceae.
Dorety (1908), Chamberlain (1919, 1935), Hooft (1970), Bierhorst (1971), Rao (1971), Pant and Sing (1991), and Foster and Gifford (1974) all agreed that cycad seedlings have a coleorhiza. Stevenson (1990, page 20), however, is of the opinion that a tree coleorhiza is unique to the grasses and that cycads do not have a true coleorhiza. Most of his criticisms about the results of earlier workers, refer to discoloured and/or dried tissue adhering to the radicle. From macroscopic studies on germinating cycad seed kernels, one of us (Grobbelaar 2005) observed that when the primary root emerges from the seed kernel its apex is capped by living tissue which it soon pierces during its longitudinal growth. This coleorhiza-like cap, originally observed on seedlings of several different Encephalartos species, were later also observed on seedlings of Ceratozamia hiMae, Cvcas revohtta. Dioon spinulosum, Macrozamia miquelli, Zamia amblyphyllidia, Zamia furfuracea and Stangeria eriopus. Stevenson (personal communication, 2005) was unable to explain these structures which were new to him and he suggested that their ontogeny should be investigated. This paper reports our findings of an anatomical study of some Encephalartos seeds and seedlings and also compares our results with those obtained by other workers who studied the embryology of members of the Cycadales and Poaceae.
There are also other terms that are used in the descriptions of cycad, seeds and embryos that need clarification such as endotesta, perisperm, epicotyl and hypocotyl. A picture of a longitudinal section of the seed of Encephalartos villosus (before germination) is included to clear up some confusion about the seed and seedling terminology.
Materials and Methods
The soaked seeds of Encephalartos altensteinii and E. villosus containing full-grown embryos were used for the anatomical study. The embryos together with the suspensor were carefully dissected from the female gametophyte (megagametophyte) and fixed in a mixture of 50% ethanol, acetic acid and formaldehyde in the ratio of 9:1:1 by volume. The fixed material was dehydrated in a series of alcohol solutions followed by a series of alcohol/xylol solutions and imbedded in wax. Sections, 10 [micro]m thick were made with a rotary microtome and double stained in safranin and haematoxylin. Seeds of the same species as well as those of E. transvenosus and E. Jerox were germinated and photographs of different stages of germination were taken. Seed germination was done in a germination cabinet, without germination medium, at 25[degrees]C and RH between 90% and 100%. The thin inner layer of the seed coat was carefully removed, cleared in sodium hypochlorite (3.5% m/v) and stained in safranin.
Results and Discussion
Seed Coat Terminology
Before discussing our results, it is necessary to sort out the inconsistent terminology used for describing cycad ovule and seed structure.
Chamberlain (1935, page 103) states that the ovule of the Cycadales has a single, 'massive integument' and in Fig. 101 presents a diagram of a section of a Dioon edule ovule at the stage of pollination, where the principal features are well represented. The diagram shows that the integument (future seed coat or testa) consists of a stony layer with a fleshy layer on its outside as well as on its inside. He further states that, "Later as the female gametophyte grows, it absorbs most of the inner layer, so that it disappears except as a dry, papery membrane closely applied to the megaspore membrane. Usually two strong vascular strands enter the ovule. The outer strands (outside the stony layer) ... extend almost to the micropyle ... The inner strands, after reaching the inner fleshy layer, fork repeatedly ... Usually they end before they reach the nucellus". It is important to note that the 'inner fleshy layer' described above, is part of the integument and not nucellus since the nucellus is generally not vascularised. According to Black et al. (2006), "The typical gymnospermous seed coat comprises three layers: and outer fleshy parenchymatous layer (sarcotesta), a middle sclerenhymatous (stony) layer (sclerotesta) and an innermost parenchymatous layer (endotesta) which generally collapses at maturity and fonns a thin membranous layer". This description conforms to the one by Chamberlain (1935) above, but in Fig. S.26 of Black et al. (2006), the labelling reads as follows: nuc--remains of the nucellus; ifl = remains of the inner fleshy layer (parenchymatous) layer (endotesta); scl = middle sclerenchymatous layer (sclerotesta; off--remains of the outer fleshy (parenchymatous) layer (sarcotesta). The margins between the remains of the nucellus and the 'endotesta' in the lower part of the seed is, however, not clear. In order to get an updated picture of the structures of the seed coat, information of the diagrams of Chamberlain (1935), Fig. 101, Harder et al. (1962), Fig. 711, Foster and Gifford (1974), Fig. 14-4 and Sanches-Tinoco and Engleman (2004) Fig. 1 together with our own observations, were used to compile Fig. 1. From the above it is clear that the integument does not cover the whole ovule. Figs. 1 and 2 show a small separation between the micropylar part of the integument and the nucellus. In their description of the gymnosperm ovule, Foster and Gifford (1974) mention that "The lower portion of the ovule, where the integument and nucellus are firmly joined, is termed the chalaza". In the cycad ovule (Fig. 1), the integument is clearly inserted at the top (micropylar) part of the ovule and the female gametophyte lies embedded in a secondary cavity formed by the basal, vascularised (chalazal) part of the ovule, analogous to the pachychalaza found in the ovules of dicotyledonous ovules (Comer 1992; Robbertse et al. (1986). The cycad ovule and seed is therefore clearly pachychalazal. The concept of pachychalaza for gymnosperm seeds was also used by Sanches-Tinoco and Engleman (2004) in their description of the seed coat of Ceratozamia mexicana. The vascularisation of the pachychalaza consists of vascular bundles outside as well as on the inside of the sclerenchymatous zone (Figs. 1, 3 and 4). Figure 4a is a picture of a clear-mount of the thin, inner layer of the seed coat (the so-called endotesta of Black et al. 2006). The vascular bundles branch twice dichotomously and the branches stretch up to a conspicuous rim at the point where the integument (now part of the seed coat) and the nucellus are attached to the pachychalaza. There are no vascular bundles in the rest of the 'endotesta' formed by the integument (Figs. 3 and 4a) or in the nucellar membrane (Fig. 4b).
Corner (1976) introduced an original classification system for dicotyledonous seeds based on the participation of the integuments in the formation of the seed coat and the position of the sclerenchymatous layer. The portion of the seed coat produced by the outer integument, he termed testa and the part produced by the inner integument, the tegmen. Seeds, where most of the seed coat is formed by the testa he termed testal and those where the main part of the seed coat was formed by the tegmen, he termed tegmic. Based on the position of the sclerencymatous tissue in the seed coat, he used the terms exotestal, mesotestal and endotestal for testal seeds and exotegmic, mesotegmic and endotegmic for tegmic seeds. If Corner's (1976) terminology may be applied to gymnspermous seeds, the cycad seed coat should be classified as mesotestal since the sclerenchymatous layer is situated in the middle part of the seed coat. The term "endotesta' should then be discarded and replaced by the term pleurotesta. The terms sarcotesta and sclerotesta are terms used by Comer (1976) for testal seeds with a parenchymatous (juicy) outer testa and a sclerenchymatous inner testa as found in pomegranate seeds. These two terms seem to be appropriate for cycad seeds where there is only one integument.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
The term perisperm is a term used for the remains of the nucellus in dicotyledonous seeds (Fahn 1982) and could be used for the remains of the nucellus covering the micropylar end of the female gametophyte, but not for the pleurotesta.
In summary, the seed coat of the cycad seed therefore mainly derives from the pachychalaza and to a lesser extent from the integument. It consists of three layers, an outer sarcotesta, a middle sclerotesta containing unbranched vascular bundles in the outer layer and a vascularised pleurotesta. In the micropylar end of the seed, there are two separate membranous structures; the one closely adhering to the sclerotesta is the unvascularised part of the pleurotesta, derived from the inner layers of the integument whilst the other one represents the remains of the nucellus (perisperm), firmly attached to the female gametophyte and covering the archegonium chamber.
Seed Germination and Coleorhiza
Figure 2 shows a longitudinal section of an Encephalartos villosus seed ready to germinate. At the micropylar end of the seed, the megagametophyte (sometimes erroneously called endosperm) is covered by the brown, papery left over of the nucellus (perisperm) and forms a cap over the archegonial chamber and radicle (Fig. 4c). The full-grown embryo lies embedded in the megagametophyte with the suspensor attached to the tip of the radicle (Fig. 2). At this stage the radicle has intruded the archegonial chamber and is pushing against the dried, remaining part of the nucellus (perisperm) (Fig. 4c). During germination, the elongating embryo rips a circular segment from the perisperm which can sometimes be seen as a cap on the apex of the 'coleorhiza' as it emerges from the sclerotesta (Fig. 5a). This is the structure which Norstog and Nicholls (1997) refer to as a 'nucellar cap'. Soon thereafter the perisperm cap is sloughed off and the radicle breaks through the suspensor cap or 'coleorhiza' (Fig. 5b). Figures 6a, b and c, 7 and 8 supply evidence that the 'coleorhiza' is part of the suspensor.
[FIGURE 4 OMITTED]
A section of the embryo (Fig. 7) shows that the plumule is separated from the base of the radicle only by a plate of vascular tissue, called the cotyledonary,, plate by Rao (1971). During germination the stem axis below the apical meristem does not elongate and therefore there is no epicotyl as stated by Rao (1971). Anatomically the switch from the exarch position of the primary xylem of the radicle to the endarch position in the cotyledons takes place in the very basal part of the radicle (Rao 1971). Therefore, there is no true hypocotyl unless the very short part of the radicle between the cotyledonary plate and the attachment of the 'coleorhiza' to the radicle (Fig. 7) is referred to as the hypocotyl. In Encephalartos embryos, the two cotyledons are completely fused for most of their length, but at the base they are fused only along their margins to form a cylindrical structure around the plumule. Rao (1971) erroneously calls this part the epicotyl in Cycas circinalis.
In a longitudinal section of the radicle (Fig. 7), it is clear that the apical meristem conforms to a typical Gymnosperm root tip structure (Fahn 1982 Fig. 38.1) where the root cap is formed by an elongated columella deriving from the columella mother cells below the permanent initials. The columella may also be easily misinterpreted as a 'coleorhiza'.
[FIGURE 5 OMITTED]
As can be seen in Figs. 7 and 8, the suspensor, which is part of the pro-embryo, remains firmly attached to the proper embryo with the epidermis of the suspensor continuous with that of the embryo proper. The suspensor is a hollow tube and when the embryo starts to enlarge and the radicle is pushed backwards in the direction of the micropyle, the tip of the radicle grows into the, now funnel-shaped, distal part of the suspensor. The folded/spiralized middle part of the suspensor unfolds and forms a loop (Figs. 6 and 7) with the proximal end still attached to the archegonium wall and the distal end attached to the radicle (Figs. 6 and 8). The cells of the proximal part of the suspensor may shrivel, but those around the radicle tip remain intact and form a sheath of live cells around the radicle. It is this part of the suspensor that was described by Foster and Gittbrd (1974) as "... a sheath of tissue which encloses the root tip" and designated as a coleorhiza. Figures 5b and 6 show the radicle breaking through the suspensor cap (coleorhiza) during germination. Rao (1971) supplies a diagram (Figure 12) in which a part of the embryo of Cycas circinalis shows the split 'coleorhiza'. This structure corresponds to the part of the suspensor described above, meaning that there is indeed a cap-like structure, derived from the distal part of the suspensor, covering the radicle tip of the cycad seedling. The question now arises whether this structure should be called a coleorhiza or something else.
[FIGURE 6 OMITTED]
[FIGURE 7 OMITTED]
Homology of the Coleorhiza of Cycad and Grass Embryos
A coleorhiza is found in the embryos of the grass family (Poaceae) but its origin and homology is disputed (Black et al. 2006). In his monumental work on "The Structure and Reproduction of Corn" (Zea mays L.), Kieselbach (1949) supplies excellent pictures of sections of the developing embryo and shows where the initials of the primary root begin to differentiate "... just above the Suspensor". Van Lammeren (1987) also states that "The root meristem is established at the inside of the embryo proper and in line with the suspensor" and further: "The cause of the precise location of root meristem formation on the transition region of the embryo proper and suspensor remains to be elucidated". In many other textbooks where longitudinal sections of the maize and wheat seedlings are presented (Black et al. 2006, page 218, Esau 1977, pages 477, 485 and 486; Fahn 1982, page 493 and Raven et al. 1992, page 443), it is clear that the suspensor in grasses is a relatively 'massive' structure. The radicle initiates close to the attachment of the suspensor to the embryo proper and the developing radicle actually grows into the broad, multi-cellular distal part of the suspensor. Mauseth (1988) presents a longitudinal section of the embryo of a grass species (Panicum maximum) where, similar to the embryos of corn and wheat, all the cells composing the coleorhiza are vacuolated cells whereas the cells of the radicle, plumule and even the coleoptile are meristematic. In the figures presented by Fahn (1982) and Black et al. (2006) the tip of the coleorhiza is even labelled as "remains of suspensor", but no separation between the latter and the coleorhiza is indicated. If these pictures are studied properly it becomes clear that the whole coleorhiza in grass embryos indeed represents the remaining distal part of the suspensor as in the cycad embryo.
[FIGURE 8 OMITTED]
In dicotyledonous plant embryos where there is no coleorhiza, the suspensor is either absent as in the Euphorbiaceae or mostly consists of a single strand of cells attached to the initials of the radicle of the embryo proper (Raven et al. (1992) and it is impossible for the radicle to grow into the suspensor as in the case of grasses and cycads. Van Lammeren (1987), suggesting that the 'radicle' in grass embryos is a seminal root, concludes his chapter on "Embryogenesis in Zea mays L." by stating: "The coleoptile and coleorhiza are not considered to be parts generated by the eventual shoot and root meristems. The former is considered to be a part of the scutellum, the latter probably the rudimentary radicula".
More research on the origin of the coleorhiza is required, but based on the available information, we are postulating that the coleorhiza of the grass embryo as well as the coleorhiza of the cycad embryo is the remaining distal part of the suspensor and both structures are therefore structurally homologous and should be referred to as a coleorhiza.
Bierhorst, D. 1971. Morphology of vascular plants. Macmillan, New York.
Black, M., J. D. Bewley & P. Halmer. 2006. The encyclopedia of seeds: science, technology and uses. Cromwell, Towridge.
Chamberlain, C. 1919. The living cycads. University of Chicago, Chicago.
--1935. Gymnosperms: Structure and evolution. University of Chicago, Chicago.
Corner, E..I.H. 1976. Seeds of the Dicotyledons. Cambridge University Press.
--1992. The pachychalaza in Dicotyledons:primitive or advanced? Journal of the Linnaean Sociey 108:15 19.
Dorety, H. 1908. The seedling of Ceratozamia. Botanical Gazette 46: 203-220.
Esau, K. 1977. Anatomy of seed plants, ed. 2nd. Wily, New York.
Fahn, A. 1982. Plant anatomy, 3rd edn. Pergamon.
Foster, A. S. & E. M. Gifford. 1974. Comparative morphology of vascular plants, ed. 2nd. Freeman, New York.
Grobbelaar, N. 2005. Does a cycad seedling have a coleorhiza? Encephalartos 83:28-29.
Harder, R., F. Firbas, W. Schnmaeher & D. Von Denver. 1962. Lehrbuch der Botanik Fur Hochschulen. Gustav Fisher Verlag, Stuttgart.
Hooft, J. 1970. Zamia from seed. Carolina Tips 33:21-22.
Kieselbaeh, T. A. 1949. The structure and reproduction of corn. Research Bulletin 161, University of Nebraska College of Agriculture, Agricultural Experiment Station.
Mauseth, J. D. 1988. Plant anatomy. The Benjamin/Cummings Publishing Company Inc, Menlo Park California.
Norstog, K. J. & T. J. Nicholls. 1997. The biology of the Cycads. Cornell University Press, Ithaca.
Pant, D. D. & R. Sing. 1991. Unusual orthotropous germination in seeds of Cycas rumhii Miq. And the morphological nature of cycad coleorhiza. Plant Science Bulletin 2:9-14.
Rao, L. N. 1971. Life-history of Cvcas circinalis L. Part V. Seedling Anatomy. Proceedings of the Indian academy of sciences. Section B, 76:47-57
Raven, P. H., R. F. Evert & S. E. Eichhorn. 1992. Biology of plants, ed. 5th. Worth, New York.
Robbertse, P. J., I. Von Teiehman & H. J. van Rensburg. 1986. A re-evaluation of the structure of the mango ovule in comparison with those of a few other Anacardiaceae species. South African Journal of Botany 52: 17-24.
Sanehes-Tinoco, M. Y. & M. Engleman. 2004. Seed coat anatomy of Ceratozamia mexicana (Cycadales). The Botanical Review 70:24-38.
Stevenson, D. W. 1990. Morphology and Systematics of the Cycadales. Memoirs of the New York Botanical Garden 57:8-55.
Van Lammeren, A. A. M. 1987. Embryogenesis in Zea mars L.: a structural approach to maize caryopsis development in vivo and in vitro. Proefschrift, Wageningen.
Hannes P. J. Robbertse (1,3) * Nat Grobbelaar (2) * Elsa du Toit (1)
(1) Department of Plant Production and Soil Science, University of Pretoria, Pretoria 0001, South Africa
(2) Department of Botany, University of Pretoria, Pretoria 0001, South Africa
(3) Author for Correspondence; e-mail: email@example.com
Published online: 13 July 2010
|Printer friendly Cite/link Email Feedback|
|Author:||Robbertse, Hannes P.J.; Grobbelaar, Nat; du Toit, Elsa|
|Publication:||The Botanical Review|
|Date:||Mar 1, 2011|
|Previous Article:||Androgenesis revisited.|
|Next Article:||Tropical vine growth and the effects on forest succession: a review of the ecology and management of tropical climbing plants.|