746 NATURAL SCIENCE. ^^^^^ 



could not fail to be instructive, as exemplifying the variations of a 

 domestic species, since they present astonishing differences. The 

 pariah dogs of this coast much resemble a smooth-coated Dingo 

 ■which once lived in the Zoological Society's Gardens, but they are 

 smaller. The normal colour appears to be yellow, but others occur ; I 

 have seen partially drooped ears and a curled tail, but, of course, 

 mixture with European dogs now and then occurs. 



A common pet with the Swahilis is a small turtle-dove (Tiiytny 

 damayensis) which presents an interesting instance of commencing 

 divergence. All the Zanzibar specimens of this bird noticed had 

 a black line from the bill to the eye, usually wanting, or not readilj'' 

 perceptible, in Mombasa specimens. The Zanzibar specimens, 

 too, of the frog Xenopus hrvis are largely yellow on the under side 

 •of the thighs — a character which is not apparent in Cape specimens 

 in the Zoological Gardens. 



This curious frog is very common in Zanzibar, inhabiting the 

 ■small and shallow wells whence the natives procure water, and 

 presumably subsisting on insects which fall in. It is usually 

 ■seen in the water with its head out, but I have seen them just 

 at the edge, on the mud. On land they move along in jerks 

 rather than hop, and this possibly explains the presence of 

 •claws in so aquatic a species. I gave a specimen of this 

 frog to a mongoose {Cyossarchus fasciatus) and the animal mortally 

 wounded and left it. Next day, however, it killed outright and partly 

 ate another specimen, which appeared to be distasteful when bitten, 

 and was soon vomited. This latter incident also happened with a 

 lizard [Mahuia striata) which the mongoose had had the day before, 

 but the lizard was, nevertheless, entirely eaten. This mongoose, 

 curiously enough, appeared to be unwilling to attack birds, though it 

 did not seem to fmd them unpalatable. The lizard above mentioned 

 is extremely common both in Zanzibar and Mombasa, though it 

 appeared to me wilder and more difficult to capture in the latter 

 locality. In Zanzibar its great haunt is the wall of an old burial- 

 ground, just outside the town, in the crevices of which they retire to 

 rest, and can be gently drawn out by the tail, if caution be used 

 to prevent its breaking ; this calamity often happens, though the tail 

 appears much less brittle in Mabuia than in Lacerta vivipara, in which 

 it parts with the slightest struggle. A gecko [Hemidactylus mabuia) 

 was al'so very common in Zanzibar, and I was able, both in that town 

 and in Mombasa, to carry out some more experiments with regard 

 to the palatability of various animals, both with Mabuia and Hemidac- 

 tylus, specimens of which I had in captivity. In the back yard of the 

 hotel grew some castor-oil plants, infested by a large number of black- 

 and-yellow, sparsely haired caterpillars, which were greedily devoured 

 by one or two glossy cuckoos. These larvae ultimately developed 

 into a moth (probably an Euproctis) almost exactly resemhWng Porthesia 

 ■anyijlua. These moths were readily eaten by the geckos, though 



