78 THE VOYAGE OF H.M.S. CHALLENGEE. 



demand too much from the imagination if we suppose that the genus originated there, 

 and has spread thence to other parts of the worhl. 



So far as some of the species are concerned, this seems ahnost certain. The two species 

 jjroaviis and hayanus are, while distinct enough, extremely closely allied and very similar 

 in general appearance. The first occurs near Gilolo, the second in the Red Sea. Now Pro- 

 fessor Semper^ has pointed out the close alliance between the moUusca, Crustacea, fishes, &c., 

 of the Red Sea and Indian Ocean and those of the Philippines and Western Pacific, and has 

 suggested that this alliance is due to the currents. He also states that a strong superficial 

 current enters the Red Sea, and, while taking many species in, prevents the return of those 

 which dwell on or near the surface. From this it seems not improbable that hayanus 

 has been developed from emigrant individuals of proavus. 



So far as general appearance and coloration go, two other a^edesi—Jlaviventris and 

 fraucnfeldanus — also resemble 2:>roavus,hut structurally they present far greater difierences, 

 and indeed iifrauenfeldanus, which I have not seen, be correctly described,^ it would seem 

 almost to deserve generic separation. Frauenfeldarms occurs at the Nicobar Islands,^or?'- 

 ventris near Ceylon, Madagascar, and St. Helena, and it does not seem impossible that 

 they have in proavus their common ancestor, or at least have had with that species and 

 hayanus a progenitor less remote than the common ancestor of the genus. 



Wiillerstorffi, micans, and princeps are three other species somewhat closely allied. 

 The headquarters of wiillerstorffi certainly appear to be the Atlantic, but notwithstand- 

 ino- this it probably came originally from the Western Pacific, for by the arrangement of 

 ocean currents it seems scarcely possible that it should have spread in the opposite direction. 



In like manner sericcus, if it occurs elsewhere than the Pacific, has probably been 

 carried thence by the currents. 



Of the distribution of the remaining species we know too little to make it worth 

 discussion. To conclude, it seems probable that the genus originated in the region of 

 the West Pacific, and that the species, or their ancestral forms, have been distributed 

 by the ocean currents. 



HALOBATODES. 



Halohatodes lituratus is reported as occurring in the Chinese Sea, between Manilla 

 and Hong Kong. The locality of the closely allied histrio is " Japan," but whether 

 it is marine or fresh-water is not stated. Compar is an Indian species, and stall comes 

 from Ceylon. 



This distribution is interesting, when it is remembered that the metropolis of the 

 species of Halohates is the region between the Indian Ocean and Western Pacific, and 

 that Halohatodes has probably been derived from Halohates. 



• The Natural Conditions of Existence, p. 279, 1S81. 

 ' See however the description, ante'a, p. 57. 



