■50 BULLETIN 151, UNITED STATES NATIONAL MUSEUM 



Schweinfurth's travels led him through the Sudan to the northern 

 Belgian Congo, and though in 1920^^ I considered that A. caudospina 

 Meek was only a synonym of Werner's A. a. fiavicauda, it would per- 

 haps be more advisable to QxcXuAe fiavicauda as an East African race 

 until actual comparison of the types with East African caudospina 

 has been made. There is nothing in the brief description oi fiavicauda 

 to distinguish it from caudospina except, perhaps, the 7 to 8 praeanal 

 pores and the statement that there are no occipitals, which seems 

 rather doubtful. 



In 1910 Meek described five male agamas from Gilgil and Lake 

 Elementeita, Kenya Colony, as A. caudospina , characterized by 

 70-85 mid-body scale rows and an exceedingly depressed and spinose 

 tail. On the base of the tail and adjacent area of the back character- 

 istically there are a series of light transverse lines on a brownish 

 ground. The throat in alcoholic specimens shows the dark longi- 

 tudinal streaks which are present in males of most of the group. 



In 1921 Lonnberg described A. elgonis from Mount Elgon on the 

 basis of seven specimens of both sexes. He considered that it was 

 probably a race of lionotus distinguished by the higher number of 

 mid-body scale rows, 80-90, and the largest scales in the vicinity of 

 the ear hardly more than half the diameter of the tympanum. The 

 number of praeanal pores is 14. In life the head was coral red, the 

 body and legs ultramarine blue. I consider that this race should be 

 called A. a. elgonis. 



In 1923 I described A. lionotus dodomae on the basis of a large series 

 of lizards from the Dodoma Province of Tanganyika Territory. 

 Boulenger, who personally examined the type, together with the type 

 of A. lionotus, considered that the only character of importance — 

 apart from coloring — distinguishing it from lionotus was the higher 

 number of mid-body scale rows, 70-89. There are 10 to 13 praeanal 

 pores. The male has a pearshaped scarlet patch on the throat which 

 is surrounded by a rich navy-blue border. It should be called A. a. 

 dodomae. 



In 1923 A. lionotus mwanzae was described from Shanwa, Mwanza, 

 Tanganyika Territory, from a series collected in the Mwanza Prov- 

 ince. The males of this species are the handsomest of any of the 

 races, being metallic purplish pink from snout to mid-body, while the 

 posterior half of the body is rich metallic blue. Unfortunately in 

 scale characters it is identical with the Dodoma race, possessing from 

 70-90 mid-body scale rows and 10 to 13 praeanal pores. It should be 

 called A. a. mwanzae. 



In 1928 Barbour and Loveridge described A. colonorum usamharae 

 from the Usambara Mountains. It is the extreme eastern represent- 



" Loveridge, 1920, Proc. Zool. Soc, London, p. 141. 



