1894,] MAMMALS OP NTASALAND. 141 



almost straw-coloured fur ; the other two are not quite so far 

 advanced. Laying them beside the skins previously received it 

 appears, although this must be for the present a merely tentative 

 explanation, that the grey fur characteristic of October skins 

 gradually bleaches under the influence of the summer sun, until 

 its black rings become first brown and then rufous, this change 

 being quite independent of the shedding and replacement of the 

 fur itself. At the same time there is a change in the paler rings 

 between white and yellow, but in which direction and at what 

 particular season the series before me does not conclusively show, 

 chiefly because, altliougli marked with the month of captui-e, the 

 exact days have not been noted, so that there is often a little 

 uncertainty as to their exact succession. Coincidentally with this 

 bleaching of the dark rings the true change of fur occurs, the fur 

 first falling ofi^ on the head, then on the shoulders and tail, and 

 remaining on the rump until in January it is, as already noted, 

 nearly straw-coloured, with rufous subterminal and yellowish 

 terminal rings. The bleaching of the fur from black to rufous 

 during life may seem almost impossible, but that it really occurs is 

 shown by the darker rings of the tail-hairs, which in October are 

 all deep glossy black, but in November those near the bases of the 

 hairs, where they are not exposed to the sun, are still nearly or 

 quite black, while the terminal ones are brownish red. 



The young specimens, all apparently of about the same age, 

 introduce a further element of complexity into the question, for 

 while four of them (Dec. and Jan.) are in a rufous stage, the fiffh 

 (December) is grizzled grey, exactly like the grizzled grey parents 

 killed in October. I can make no suggestion for the elucidation 

 of the mystery, but I would suggest, to any one having the oppor- 

 tunity, the collection of a mother and her whole litter of young, the 

 skins to be marked with their exact date, and with the fact of 

 their belonging to one another. 



I may venture to hope that further collections will contain more 

 specimens of this very remarkable species, so that I may later have 

 the pleasure of giving a complete account of its changes all the 

 year round. 



The fourth collection, made from May to August 1893, contains, 

 unfortunately, no specimens of S. mutahilis. 



19. Mus DOLICHUBUS, Smuts. 



a. Ad. al. $ . Zomba. 12/92. 

 h. Imm. al. Zomba. 12/92. 



The following are the measurements of the well-preserved adult 

 specimen : — Head and body 97 mm. ; tail 155 ; hind foot, without 

 claws, 22 ; ear from notch 1 5"5. 



Mammae 1 —2 = 6. 



20. Mus MODESTUS, Wagn. 

 a. Ad. al. Zomba. 1/93. 



