1888.] EaUAlORIAL AFRICA. 9 



These specimens vary very considerably in the colour of the 

 belly-hairs, which in some of them, notably in a, are pure white, 

 while in others they are of the usual dull grey and white found in 

 Western Coast specimens. 



19. SCIURUS ANNULATUS, Dcsm. 



a. 2 . Lado, 16/2/84. 



b. 6- Tobbo, 18/5/83. 



c. ?. Tobbo, 18/5/83. 



20. SCITJRUS PYRRHOPUS, F. CuV. 



a. S. Tingasi, 16/9/83. 

 h. 2 . Tingasi, 10/10/83. 

 " Iride fusca, ' Kejo.' "— E. 



Specimen a has the hairs of the belly, usually pure white, richly 

 washed with red. 



S. stangeri, S. rufobrachiatun, and the present species are all 

 strictly West-African forms, this being by far their most easterly 

 recorded locality. 



21. SciURTJS BOEHMi, Reichcn. 

 a, 6. c? & 2 • Tangasi, 7 and 9/83. 

 c, d. (J & 2 . Stat. Gadda, 2 and 3/84, 

 e. S . Nendja (Monbuttu), 9/7/83. 



/• 6. 



" ' Nangeri.' Common from 4° N. lat. southwards. As yet 

 only taken to the west of the Bahr el Djebel." — E. 



These beautiful little Squirrels quite agree with the description 

 given by Dr. Reichenow (Zool. Auzeiger, 1886, p. 315) of some 

 specimens obtained by Dr. R. Bohm in the Marungu country, at the 

 south-west corner of Lake Tanganyika. Whether the species is 

 really distinct from S. congicus, Kuhl, is a question which will have 

 to be decided later, when further specimens are available from other 

 localities. It appears to me to be by no means improbable that not 

 only will iS. hoehmi be found to grade into S. congicus, but that the 

 latter also will equally pass into S. poensis. Smith, of which I have 

 seen specimens with just a faint indication of a whitish back-stripe. 

 This stripe then becomes bright and prominent in S. congicus, with 

 a darker band below and external to it ; while further, in tS. hoehmi 

 this latter dark band is quite black, and is supplemented internally 

 by a second even more prominent black stripe. These gradations 

 from the normally wholly unstriped S. poensis to the brilliantly 

 banded 8. hoehmi show how little these dorsal bands can be trusted 

 for the division of the Squirrels into groups, or even for the discri- 

 mination of the species. No appreciable seasonal change is visible 

 between the different specimens of the series. 



The Squirrel from " Tamaja " referred to and figured by Leche ' 

 as S. lemniscatus, Le Coute, appears also to belong to this species, 

 which, while having the same number of stripes as S. lemniscatus, 



' T. c. p. 117, pi. iii. 



