EAST AFRICAN REPTILES AND AMPHIBIANS 51 



ative of A. agama agama differing from the typical form in several 

 respects recently discussed. The throat of the male is crimson-lake 

 colored and faintly shows gray longitudinal lines. It is a species 

 occupying recently deforested areas and now living upon rocks, 

 though one was seen upon the trunk of a solitary tree. 



The status of A. doriae, described from Abyssinia (Ethiopia) by 

 Boulenger in 1885, is doubtful as far as East African records go. 

 Nieden simplified matters by saying that all Tanganyika Territory 

 records should be referred to A. colonorum (that is A. agama). Prob- 

 ably Lonnberg's records from Ngari na Nyuke and Nairobi Falls near 

 Juja Farm are referable to another race. Boulenger in 1896 referred 

 specimens from Lakes Hudolph and Stephanie to doriae but later 

 united doriae with liartmanni of the Sudan. It is said to differ from 

 A. agaraa typica in having the nostril pierced below instead of on the 

 canthus rostralis. After examining the long series of males attributed 

 to ^.cr^amaa^ama in the following pages I can not find one answering to 

 this description, though there is a wide range of variation both in the 

 size of the opening and its direction upward, outward, or backward. 

 For the present it seems advisable to omit it from the fauna of British 

 East Africa. 



To sum up, it may be said that where open plains and low scattered 

 rocky hills occur, both to the north and to the south of the equator, 

 marked difi'erentiation is taldng place from the typical arboreal 

 A. a. agama, mucros tend to disappear, keels on the scales become 

 more obtuse, and the kopje-dwelling agamas have an appearance so 

 like planiceps that many authors have referred East African captures 

 to that race. One might compare the speciation that has been going 

 on without the forest belt to that occurring in the squirrels of the 

 genus Heliosciurus recently discussed by Major Ingoldby.^^ 



Series taken from any one locality show a strong family likeness and 

 the type or types of the species which I now propose to designate 

 races, are, in most instances, of very distinct appearance. This 

 applies chiefly to the adult males; younger males are more homog- 

 enous, while females in most cases are indistinguishable, in some races 

 retaining their strongly keeled scales though those of the males are 

 much smoother. 



If the labeling is correct. Fort Hall appears to be the meeting place 

 of three of the forms. It should be borne in mind that the topo- 

 graphical features in the immediate vicinity of Fort Hall are of the 

 most diverse nature with wooded heights, thorny scrub, and rock- 

 strewn valleys in close proximity. 



>' Ingoldby, 1927, Proc. Zool. Soc. London, p. 471. 



