Resumen
The presence of Ancylidae in South Africa was established as early as 1848 but the ?rst comprehensive study on this family was reported in 1923. This author came to the conclusion that extensive work still had to be done before it could be declared that knowledge pertaining to the South African representatives of this family could be considered as satisfactory. Another important contribution on South African Ancylidae was published in 1939. After a detailed morphological and histological study on Burnupia mooiensis three decades later the authors stated that the axonomy of the Ancylidae still remained unsatisfactory, a view that was supported by authors as recently as 2002. The National Freshwater Snail Collection (NFSC) was established in 1956 but he oldest sample of Ancylidae on record dates back to 1954 and at present 14 species of Burnupia are represented in the database of the NFSC. This contribution deals with the geographical distribution and speci?c characteristics of the habitats of B. capensis and B. stenochorias the two most widespread species of this genus in South Africa. Very little is known of the conservation status of the Ancylidae of this country but some results pertaining to this subject are brie?y discussed. Burnupia species are relatively inconspicuous organisms that are usually attached to solid objects on the substratum or aquatic vegetation and need to be speci? cally looked for during surveys. Personnel of government authorities and municipalities who made substantial contributions to the number of samples in the NFSC therefore received special training in sampling echniques during workshops. Identi?cation of the specimens was largely based on shell and radula characteristics and on he location of the site of collection. Only samples of which the sampling sites could be located on the 1:250 000 topo cadastral map series of South Africa were considered for further analysis. Respectively 702 and 435 samples of B. capensis and B. stenochorias met this requirement. The loci (1/16 th square degrees) in which the collection sites of these species were located, were distributed in pre-selected intervals of mean annual air temperature and rainfall, as well as ntervals of mean altitude, to illustrate the frequency of occurrence within speci?c intervals. A temperature index was calculated for all mollusc species in the database from their frequencies of occurrence within the selected intervals and the results used to rank them in order of their association with low to high climatic temperatures. Chi-square values were calculated to evaluate he signi?cance of the difference between frequencies of occurrence in, on, or at the various options or each parameter investigated and furthermore an effect size value was calculated to determine he contribution of each parameter towards establishing the geographical distribution of this species based on the data in the database. A multivariate analysis in the form of a decision tree was also constructed from the data which enabled the selection and ranking of those variables hat maximally discriminated between the frequency of occurrence of these two species in, on, or at the various options for each parameter as compared to all other mollusc species in the database. The majority of samples of both species were collected in rivers and streams and in habitats described as perennial with slow running and clear water. The presence of associated vegetation and a predominantly stony substratum was mentioned for the majority of sampling sites. The geographical distribution of these two species overlapped to a large extent and they were recorded together in 55 loci. However, the 257 loci from which B. capensis were collected covered a more extensive geographical area than the 108 loci from which B. stenochorias was recorded. Both species were poorly represented in the Limpopo, North West and Northern Cape province, a phenomenon that could probably be attributed to a paucity of perennial habitats and the fact that both species in question have poor abilities to overcome conditions of desiccation prevalent in seasonal habitats. Due to the fact that the majority of samples were collected in rivers and streams and on a stony substratum and also because of their largely sessile mode of existence, it is suggested that the feasibility to exploit these two species as indicators of heavy metal pollution should be investigated. It is a known fact that molluscs can accumulate heavy metals in their soft tissue and can also deposit organic substances in their shells. Large scale surveys for freshwater molluscs was terminated in the 1980?s and most of the collection sites recorded in the NFSC were not revisited since, therefore little is known of the conservation status of the freshwater molluscs of South Africa. A comparison of results of several surveys conducted in the Kruger National Park over a period of nearly four decades revealed a marked decline in the number of positive sites with regard to several mollusc species including representatives of the Ancylidae. The well documented geographical distribution of several species of Burnupia in the database of the NFSC could serve as a base of reference for an investigation of the conservation status of this genus of the Ancylidae in South Africa. The fact that these species are well represented in rivers and streams and on stony substratums could facilitate such an investigation in the sense that surveys could be focused on these types of water body. A comparison of the results of such surveys with the documented geographical distribution of all freshwater molluscs in the database of the NFSC could contribute substantially towards the knowledge currently available on the conservation status and species diversity of our indigenous molluscs.