38 THE VOYAGE OF H.M.S. CHALLENGER. 



entirely absent, the lower rays being whitish. Ventrals nearly uniformly white, or with 

 the outer ray slightly tinged with grey. No black on the dorsal fin. 



This is the first species of Exocoetus in which I have found the coloration of the 

 pectoral fin varying. There can be no doubt as regards the specific identity of the 

 sjiecimens before me. 



Ten specimens, from 10^ to 13 inches long, were obtained at Honolulu. The 

 Sandwich Islands seem to be the only locality at which this species has been found 

 hitherto. 



Astronesthes niger, Eich. 



This fish was included in the Report on Deep-Sea Fishes, as specimens were 

 recorded from a depth of 2500 fathoms. However, as mentioned there (p. 203), it is 

 one of the most common of pelagic forms in the Atlantic as w^ell as in the Indian Ocean, 

 and, therefore, is caught on almost every voyage during which the tow-net is used. 

 Also, the Challenger obtained a surface specimen near the west coast of Africa on April 

 28, 1876. The habits of this fish are nocturnal. 



Halajihya, n. gen. 



I propose this generic name for three specimens, 26 mm. long., and 1"5 mm. deep, 

 which were obtained at the surface, in the open sea, on the passage from Sydney to 

 Wellington. They are evidently the young, and probably a very early stage of growth, 

 of a fish which shows some affinities to Microstoma. Costa (Faun. Regn. Napol. Pesc.) 

 figures on pi. xl. fig. 4, a small fish which must have been very similar to Halaphya ; 

 but neither description nor name seem to have been published by him in explanation of 

 the figure. 



There is, however, another fish from the Southern Indo-Pacific, which has to be 

 considered in determining the origin of these young specimens, viz., Gonorhynchus. 

 The fishes of this genus are rather scarce, but extend from the Cape of Good Hope to 

 New Zealand, and Japan. The form of the body, the position and shape of the fins, 

 and even the number of fin-rays of Halaphya are remarkably like those of Gonorhynchus, 

 but the form of the snout is entirely diff'erent. We should be obliged to assume that 

 with advancing grow^th the upper jaw is produced into a long proboscis, at the end of 

 which a barbel is developed. A change like this is not without parallel in the develop- 

 ment of fishes, but without knowing any of the intermediate stages we should not be 

 justified in assuming it in this case. Of the development of Gonorhynchus nothing 

 whatever is known. 



With the materials at present available the genus may be characterised as follows : — 



