An investigation of the anatomy and gas exchange characteristics of Mariscus congestus in three different habitats was undertaken in order to establish whether M. congestus from the three different habitats displayed any ecotypic responses when placed in a new similar environment. It was hoped that the results of this investigation would yield evidence that would support the ecotype concept similar to the investigations of Milner and Hiesey (1964), Green (1969) and Slayter and Ferrar (1977). On the basis of the site leaf anatomy, M. congestus investigated at the coast (site 1) differed in many respects from the inland plants (sites 2 and 3). These differences suggest that the coastal plants may have undergone a slight ecotypic divergence from the inland plants. The anatomical investigation also suggested that the leaves of M. congestus from all three sites may either be C₄ NADP-ME or PCK and that all had typical Chlorocyperiod anatomy. The habitat microclimates at sites 1-3 had different light and water regimes. There were no significant differences between the 12 month temperature environments of the three sites. There was however, a minor difference between the coastal (high temperature) and the inland (lower temperature) sites. M. congestus at the three sites had significantly different CO₂ assimilation rates, transpiration and stomatal conductance in response to the differing habitat microclimates. The water use efficiency of the sites were however, similar. Site 1 attained the highest CO₂ assimilation rates, transpiration, stomatal conductance, and water use efficiency and site 3 the lowest. Under similar conditions the gas exchange data for the potted plants indicated that M. congestus from the different sites was typically C₄. The optimal photosynthetic temperatures of all the sites was above 30°C and they did not show significant inhibition of CO₂ assimilation by different oxygen concentrations. The results of the laboratory investigation of the potted plants suggested that the only site-specific (ecotypic) response of M. eongestus was the light intensity at which the plants from the different sites were light saturated. The light and temperature response of field plants under field conditions was not comparable to the light and temperature response of potted plants under laboratory conditions. This may have been due to the field results being obtained under differing water and soil nutrient regimes. The potted plants may also have had a reduced root mass compared to their field counterparts and the potted plants may have also have become root bound. Under field conditions the plants had differing light saturation points and optimal photosynthetic temperatures compared to the potted plants. This investigation thus did not support the hypothesis stated in this thesis. The data in this investigation thus may indicate that plants with as diverse habitats as Mariscus congestus that are removed from their natural habitats display rapid changes in gas exchange characteristics in response to their new microclimates, with few ecotypic physiological characteristics of the old habitat being retained.

The principal objective of this investigation, was to collect the family Cyperaceae and to study their leaf, bract and culm anatomy. The second was to examine the collection for unique structures or forms, whilst a third was to classify the Cyperaceae according to their photosynthetic structures and types. Distribution of the Cyperaceae within the broad region defined as the Eastern Cape would be influenced by rainfall pattern. It was expected that C₃ species would predominate in more mesic environments and habitats, whilst the C₄ species would be found in drier less favourable habitats. Collection within the region (November 1993 to late January 1997), yielded 106 species, totalling some 600 specimens. Both sub-families of the Cyperaceae (Caricoideae and Cyperoideae), eight tribes (Abildgaardieae, Cariceae, Cypereae, Hypotvtreae, Rhynchosporeae, Shoeneae, Sirpeae and Slerieae) and twenty five genera were found to be present. Sixty percent of the species were C₃ and forty percent were C₄. Sixteen new species, which had not been collected within the boundaries of the region previously were also found. Over 43 percent of the species collected had unique anatomical characteristics that appeared to be influenced by habitat and or climate influenced. These are the characters influenced by hydromorphic, mesomorphic and xeromorphic environments. These anatomical characters: Thickness of leaves and bracts; thickness of the adaxial and abaxial epidermis of the leaves and bracts; flush, sunken and raised stomata; presence or absence of bulliform cells; presence and distribution of sclerenchymatous structures; presence or absence of a hypodermis; presence or absence of cavities in the leaves, bracts and culms; mesophyll or ground tissue structure, and the presence or absence of secretary structures. A few noteworthy anatomical characters that are influenced by climate are present in Cladium mariscus subsp. jamaicense (Schoeneae), the Cariceae, the Cypereae and the Sclerieae. Within Cladium mariscus subsp. jamaicense the pseudo-dorsiventral leaves and bracts, as well as the large lamina cavities, containing trans-lamina girders are unique. Papillate epidermal cells are limited to the tribe Cariceae. In the Cypereae many of the species lack bulliform cells and hypodermal layers. In Pycreus cooperi (Cypereae) the vascular bundles of the leaves and bracts, appear to be stacked in rows, that are inter-spaced with lamina cavities. In the Sclerieae the mesophyll structure is specific to the species level. Unique anatomical characteristics were also present in the leaves, bracts and culms of the genus Carpha. In this genus distinctive lateral vascular bundles were present abutting the large bundles and/or midrib bundle. The most distinctive anatomical characteristics that could be used to separate the members of the Cyperaceae were the structures and associated structures related to the photosynthetic pathway. The Eastern Cape Cyperaceae could be divided into three distinct groups based on photosynthetic structure, namely one C₃, a C₄ and a potential C₃-C₄ intermediate group. The C₃ group was found to have non-radiate mesophyll and an outer parenchymatous sheath with small chloroplasts (Cariceae, Cyperaceae [in part], Hypolytreae (Chrysithrix capensis], Rhynchosporeae [R. brownii], Schoeneae, Scirpeae and Sclerieae). The C₄ group has radiate mesophyll and an inner parenchymatous sheath with enlarged chloroplasts (Kranz or PCR sheath). The C₄ species are present in tribes Abildgaardieae, Cypereae (in part) and Rhynchosporeae (R. barrosiana). In the Eastern Cape, a few species with C₃ anatomy have anatomical characteristics that are similar to the species with C₄ anatomy (Cyathocoma hexandra [bracts], Cyperus tennellus var. tennellus [leaves and bracts], Ficinia bulbosa [leaves], F. dura [leaves and bracts], F. lateralis coastal [leaves and bracts], F. oligantha [bracts], F. pingiour [bracts], F. stolonifera [leaves and bracts], F. tribracteata [leaves and bracts], F. zeyheri [leaves and bracts], Isolepis cernua [leaves and bracts], I. costata var. macra [bracts], Schoenus nigricans [leaves], Scirpus nodosus [bracts] and Tetraria cuspidata [leaves and bracts)). The vascular bundles within this intermediate group, fall within the Hattersleyand Watson (1975) minimal cell lateral count and maximal cell distal count criteria for C₄ grass species. However, no biochemical data exists to see whether they are C₃-C₄ intermediates or whether the Hattersley and Watson (1975) C₄ criteria for grasses applies to smaller, or scutiform Cyperaceae or not. Based on the results presented here, five distinct structural forms/types were found to be present in the C₃, C₄ and C₃-C₄ intermediate groups. The C₃ and the potential C₃-C₄ intermediate species may be divided into two types, based on the number of vascular sheaths present. In the first or A-type, vascular bundles are surrounded by two sheaths and in the more dominant B-type, by three. The A-type was found in the Cypereae (Cyperus denudatus and C. textilis) and most of the Scirpeae. B-type anatomy occurred in the Cariceae, Cypereae (c. difformis, C. pulcher, C. sphaerospermus, C. tennellus var. tennellus and P. mundii), Hypolytreae, Rhynchosporeae (R. brownii), Schoeneae, Scirpeae (Bolboschoenus maritimus, Ficinia cinnamomea, F. fascicularis, F. lateralis both, F. pingiour, the genus Fuirena, I. diabolica, I. fluitans, I. prolifera and Schoenoplectus paludicola) and Sclerieae tribes. Based on the vascular sheath structure, the C₄ species could be divided into three groups, namely bulbostyloid, chlorocyperoid and fimbristyloid, where the bulbostyloid structure occurred in Bulbostylis schoenoides. Cyperus (in part), Kyllinga, Mariscus and Pycreus (except P. mundii) had a chlorocyperoid structure. Genera with fimbristyloid structure were recorded in the genera Abildgaardia, Bulbostylis and Fimbristylis. The bulbostyloid type represents a potential a fifth C₄ anatomical type within the C₄ Cyperaceae. As a result of this observation, it is possible that the C₄ syndrome may have evolved five times in the Cyperaceae and not four as previously suggested by Bruhl and Perry (1995) and by Soros and Dengler (2001). The C₃ Cyperaceae species within the Eastern Cape are more dominant in higher elevation habitats the C₄ species, similar to the C₃ grasses. The only C₄ species that occur at high elevations are those with three sheaths. The C₃ and C₄ species within the region occur in similar low rainfall habitat ranges, where the C₄'s are more dominant in xeric habitats on drier soils than the C₃ species, similar to the grasses. Where more C₃ species occur in higher rainfall habitats than the C₄ species. With the exception of the Afromontane Bulbostylis schoenoides and R. barrosiana, the C₄ species similar to the grasses are dominant in high light and temperature habitats with low rainfall, unlike the C₄ Cyperaceae of Japan and America. Only five species occur in the desert like conditions of the Karoo-Namib biome (Cyperus laevigatus, C. rupestris var. rupestris, I. cernua, M. capensis and M. uitenhagensis), which have less than 250mm of rainfall per annum. Only three species are habitat-specific or may be endemic to a specific area within the Eastern Cape, namely A. capensis, Chrysithrix capensis and R. barrosiana. A. capensis in marshes on the Amatole mountains near Alice and Hogsback. C. capensis to the Tstsikamma mountains of the Wite Els Bosch forests. R. barrosiana to the marshlands of the Cape Morgan coastal Nature reserve at Kei Mouth. The anatomical types of the C₃ and more especially C₄ Cyperaceae are not specifically found in a particular rainfall regime or habitat type, which is contrary to the thesis hypothesis. However, the C₃ species are mostly correlated with hydrophytic to mesic habitats, with the exception of Ficinia and the two sheathed species. Ficinia is dominant in mesic grasslands and halophytic habitats. The two sheathed C₃ species are mostly present in halophytic habitats. The C₄ species are also more dominant in mesic to xerophytic grasslands, as expected in the hypothesis. Where only a few species occur in habitats correlated with increasing rainfall and temperature similar to the C₄ Cyperaceae of Japan and America. It may thus be that the development and evolution of the different C₄ anatomical forms (or phylogenetic forms) within the Cyperaceae may have enabled these species to establish themselves in habitats that were alien to their origins. It may be that the ability to regulate photoassimilate and water transport within the Cyperaceae enables their success in a dynamic and unpredictable climate, such as the Eastern Cape. Many of the anatomical characteristics reported in this thesis and its appendices are unique to the tribes, genera and/or species of the Eastern Cape Cyperaceae and thus may be valuable to future taxonomic classifications of the family. The research presented here should provide a good working platform for future, more detailed research on this often forgotten component of the vegetation.