1904. J FROM EAST AFRICA AND ZANZIBAR. 87 



The buccal mass is rather large, the radula fragile, with a wide 

 rhachis. There are 21 rows in all, some of the longest of which 

 contain 24 teeth on each side. The first tooth (fig. 3 a) is larger 

 than the rest, and projects into the rhachis ; it is slightly bent, but 

 hardly hamate. All the first teeth are similar and regular in shape. 

 The base is somewhat wavy and as if hollowed out. The other 

 teeth are awl-shaped, with an irregular and somewhat bifid base 

 (fig. 3 c). The liver is yellowish grey and not very large. In front 

 of it are two hermaphrodite glands, much as in T. cocci7iea, but 

 smaller. The verge is armed with numerous' short thorns of very 

 varying shape. 



I think this animal is probably Kelaart's T. ceylonica, for 

 which the genus was founded, and which appears not to have 

 been described since ; but it is difficult to be certain of the identi- 

 fication, as he gives no information respecting the radula. The 

 form and colour agree well, including the red lines on the branchiae 

 and round the foot. The chief diff"erence is that whereas his 

 specimen has 15-16 pure white branchiae "set round a large 

 disk," mine had 12 yellow branchiae set in a circle open behind. 

 But his specimen was nearly twice the size of mine, and probably 

 the larger individuals develop extra plumes which close up the 

 posterior gap. On the other hand, both specimens agreed in 

 having rather large feathery branchiae, an unusual character in 

 the genus. Kelaart says " they resemble a small tuft of marabout 

 feathers." 



Trevelyana crocea B. (Plate III. fig 4.) 

 [Bergh, in Semper's Reisen, xvi. 2, p. 850, figs.] 

 More than 100 specimens from the East and "West Coasts of 

 Zanzibar, where it is one of the commonest littoral molluscs at 

 certain seasons. 



Mr. Crossland, who collected them, informs me that this form 

 provided a most striking case of the migration of molluscs in 

 flocks to shallow water for the deposition of spawn. 



But few specimens were collected before a certain period of a 

 few days' duration, when the sand of Chuaka BaiJ^ just below low- 

 tide mark was occupied by astonishing numbers of these delicate 

 little nudibranchs. These were not washed up by accident, but 

 were all actively crawling on the sand among the weeds &c. 

 Many were in coitu, and when placed in basins of sea-water 

 most of the specimens were soon engaged in copulation or the 

 deposition of yellow egg-ribbons. By-and-by the swarm dis- 

 appeared to some unknown permanent habitat. If this were in 

 the deeper channels of the bay (1 to 2 fathoms deep at low tide) 

 they must have been found there by dredging. As this was not 

 the case, it seems most probable that the migrations of these tiny 

 animals extend to and from the deep sea three or more miles 

 away. An almost equally conspicuous swarm was formed by 

 individuals of Melihe Jimhriata^ and other species (e. g. Geratosoma 

 cornigerum, Chromodoris spp., and Pleurohranchus delicatus) 

 appeared occasionally for a few days in considerable though 



