176 MR. CYBIL CROSSLAND ON THE [Mar. 7, 



aftei" a stay of a little over a week, in order to find means of 

 i-eaching the Island of Bonavista (text-fig. 24), in the east of 

 the group, to examine some " coral-reefs " marked on the charts. 

 This occurred three days later when the Government steamer 

 ' Mindello ' left St. Vincent for its monthly circuit of the 

 Islands. I thus saw the northern point of St. Antonio, with its 

 deep valleys, carpeted with vivid green, and the huge precipices 

 of its shores, the lower but rocky shores of St. Nicholas, and the 

 white sand-spit which forms the southern part of Sal — the Island 

 of Salt (text-fig. 21, p. 171). This was the first time I had seen 

 pure white sand in these islands, so suggestive of the vicinity of 

 coral. However, neither here nor at Bonavista, where the same 

 sand forms a large part of the western shore, is there any sand, 

 or indeed any other rock, of coral origin, and the " coral-reefs " of 

 the chart, like others in the vicinity, are simply limestone shoals, 

 not resembling coral-reefs even in form and with either very little 

 or no coral anywhere about them. 



The same absence of coral-reefs has been characteristic of the 

 past, for although the greater part of at least the western side of 

 Bonavista is of recent limestone, containing in places numerous 

 fossil shells, I found no particle of coital in it, either on the coast 

 or inland, or in the small shallow limestone beds near the town 

 of St. Vincent. 



Bonavista is not an appropriate name. The apj)earance of the 

 island is not at all picturesque, and its discovery has not been 

 much blessing to the human race. The island is a desert only a 

 little less complete than the greater part of St. Vincent. As a 

 little grass grows after rain, a population is established on the 

 island subsisting by cattle-breeding. Every few years the rain 

 fails to appear, and as much as a third of the population perishes, 

 since relief works are practically unknown to the Poi^tuguese 

 Government. At Port Sal Rei half the houses are in ruins, 

 and some have even been abandoned diu-ing course of erection. 

 Residence among such signs of misery is not pleasant, and I was 

 glad to leave at the end of a fortnight, having satisfied myself 

 that Coral or Zostera habitats do not occur in these Islands. 



My return to St. Vincent involved three days in a small Italian 

 " felua," but, in spite of the motion of so small a boat and the 

 primitiveness of the accommodation provided (on deck), much of 

 the time was rendered delightful by the number and variety of 

 fish, birds, and dolphins seen close at hand during a calm. 



3. Comparison of the Fauna with that op East Africa. 



Although several East- African species reach the Mediterranean, 

 and certainly others extend from this sea to the Cape Verde 

 Islands, it is at once and certainly evident that the faunas of 

 these two localities, taken as wholes, are distinct. Of species 

 common to the two localities, there are at least three species 

 of Crab, several Prosobranch molhiscs, the Polychsete Eunice 



