308 



MR. S. A. NEAVE — EEPORT ON A JOURNEY 



within a few hundred yards of Molilo's big village, some six miles further 

 east. In spite o£ a prolonged search until after sunset on the same day, not 

 a single other specimen was even seen. In this locality there was no actual 

 running water, though the spot was an exceptionally damp and cool one, 

 there being a quantity of dense shade in the immediate neighbourhood. 



Fip', 4. — A f*mall pool near the Liiangwa, at the end of July ; 

 this would be dry from September to November. 



It would appear that G.fusca in the dry season, at least, frequents the 

 tributary streams of the Luangvva rather than the banks of the main river, 

 which have wide sand-banks and belts of reeds separating them from the 

 water^s edge, and perhaps do not provide sufficient shelter from the scorching- 

 sun of this locality. I am inclined to attribute the presence of my original 

 specimen of G. fvsca at Hargreaves, in the wet season, to the fact that a 

 fjxir-sized stream, the Mvuvia (of which the above mentioned Kanga is 



