48 SJÖSTBDTS KILIMANDJARO-MERU EXPEDITION. 17: 2. 



is of the reniotest, but that there should be any species common to regions so widely 

 separated is of much interest. The absence from a given fauna of certain genera 

 and species offen affords as mvich food for speculation and as many points of interest 

 as the presence of others, and this fauna offers no exception to such a statement. 

 The genus Anaplecta is moderately represented in West Africa but except for one 

 species from the Soudan is absent in the East. Epilampra is well represented in 

 West Africa but is entirely absent from the Eastern and Southern regions, its place 

 being filled to a certain extent by the genus Gyna, though this too is poorly repre- 

 sented both in numbers of species and individuals in the Kilimandjaro and Abyssinian 

 regions. Polyphaga is a characteristically dry-country genus, its absence from Dr. 

 SjöSTEDt's collection is therefore not surprising; in Abyssinia, Somaliland and South 

 Africa it is well represented. South Africa may be regarded as the head-quarters 

 of the family Perisphaenidae and such genera as Aptera, Pronaonota, Pilema, Melano- 

 hJatta, Hostilia, Pcvciloblatta and others are peculiar to this region and include some 

 of the most abundant species; in the Kilimandjaro region this sub-family is repres- 

 ented by four genera and six species only, two of the sjjecies being represented in 

 Dr. Sjöstedt's collection by a good number of individuals, nearly all of the female 

 sex, still it cannot be said that this sub-family is dominant in this region. The most 

 dominant species of the Kilimandjaro fauna, as evidenced by the number of indi- 

 viduals captured, are the species of the genera PseiidoderopeUis and Deropeltis, and 

 notably P. petrophila and D. autraniana ; these were taken in great abundance at 

 all seasons of the year and at varying altitudes and in a lettei' Dr. Sjöstedt informs 

 me that many were found under the boulders bestrewing the Masai steppe-country. 

 The discovery of a cockroach symbiotic with termites is of much interest and 

 that this species should be strictly congeneric with a species found in the nest of a 

 social wasp in S. America is a fact that can scarcely be explained on any other 

 supposition than that further collecting will reveal the wide spread distribution of 

 the genus thoughout the tropics. 



