much of its area of distribution; S. mossambicus and S. hornorum well
known for their hybridization and its effect on sex ratio (Chen 1969; Hick-
ling 1960); S.S. spilurus and S.S. niger, which were studied in Kenya by Van
Someren (1962) and Van Someren and Whitehead (1959a, 195913, 1960a,
1960b, 1961a, 1961b, 1961~); S. macrochir, which was studied in Zambia
and Shaba and transported to Ivory Coast where it was experimentally inter-
bred with S. niloticus (Jalabert et al. 1971; Lessent 1968); and S. andersonii
studied in Zambian ponds by Mortimer (1960) and in the Kafue floodplains
by Dudley (1979) and Kapetsky (1974).
The taxonomy of the species of Sarotherodon is dealt with fully in a new
work (Trewavas, in press).
Although the practical fish farmer cannot be expected to go into the
niceties of specific distinctions, I suggest that key suppliers of stock should
know what species they are using and whether these are pure. It is only
necessary to recall that the sex-ratio may be affected by mixing the species,
many of' which interbreed easily. There is also the possibility that one
species, and not necessarily the most desired, may oust the other as T. zillii
did T. rendalli at Yangarnbi (Gosse 1963).
Further I suggest that in Africa itself fisheries officers should look carefully
at the native species before importing stock from other regions. If on due
consideration they decide to import stock they should keep a careful record
of it.
Addendum: Alternative and Preferred ClassificationSince writing the above, I have had discussions with fish culturists in
Kenya. They are using a brood tank whose design is based on the breeding
habits of members of subgenera Oreochromis and Nyasalapia, namely the
'lek' system with migration of the brooding mother to a 'nursery' area. They
refer to this as the system characteristic of Sarotherodon. It is not, however,
the system practiced by species of Samthemdon sensu stricto, exemplified
by S. galilaeus, and I have advised them to refer to it as the Oreochromis
system of breeding. This and other considerations now lead me to believe
that we should place more emphasis on the gap between Sarotherodon and
the other subgenera by placing the latter in a separate genus, the earliest
available name for which is Oreochromis. Danakilia is also generically distinct.
The mouthbrooding tilapias used for fish culture accordingly would be
grouped in two genera, Sarotherodon and Oreochromis the latter comprising
four subgenera as shown in the accompanying table.