76 HAROLD 1 1 1. A I'll. 



. 



other specimens taken by the U. S. F. C. Str. "Albatross" in the 

 open ocean off the Washington coast and in Bering Sea, are 

 totally without them. These last named specimens, coming from 

 the same' depth (40 fm.) occurred on a pebbly bottom or one of 

 broken shell and it is possible that the Puget Sound individuals, 

 without the coelenterate, occurred in a similar habitat. Be that 

 as it may, it is a suggestive fact that in the fishes under con- 

 sideration the hydroid was "attached in large numbers about the 

 gill opening, on the throat and in the axilla," in other words over 

 the ventral surface that is already the most concealed portion of 

 the body. Referring to Hypsagonus quadricornis Prof. C. H. 

 Gilbert writes in Jordan and Evermann's "Fishes of North and 

 Middle America" (p. 204): "In the aquarium the fish appears to 

 walk, resting alternately on the upper and lower pectoral rays 

 and on the front rays of the anal." Under such circumstances 

 the eddies produced in the bottom ooze would naturally bring 

 the greatest amount of organic material to animals ventrally 

 situated. The appearance strongly suggests that the advantage 

 lies rather with the- hydroid just as it does with the several species 

 of barnacles attached to the skin of the whale. Whether the 

 association is any more intimate in the case Alcock cites it is 

 impossible to state conclusively, but the evidence is certainly not 

 entirely convincing. 1 



Prof. C. C. Nutting, to whom I have submitted specimens, has 

 kindly identified them as a species of Perigonimus, its nearest 

 relative being apparently P. vestitus Allman. As in other mem- 

 bers of the genus the hydrorhixa forms a highly branched, fre- 

 quently anastomosing, system over the surface of the fish, but 

 so far as noticed this contact is purely superficial, there being no 

 evidence of parasitism. And furthermore the presence of sin. ill 

 cntomostracans and nondescript organic remains in the gastric 



'Since this paper was sent to press I have examined upwards of two d< 

 specimens of this same species of rock perch (M. inermis) collected by my 

 colleague Prof. J. O. Snyclcr, at Onomichi, on the Inland Sea, in the Province of 

 Hingo. Japan. All of these are excellently preserved and in no instance has a 

 hydroid been found upon them. It thus becomes more certain that the associa- 

 tion described by Allcock is not an undoubted case of commensalUin. I'i>''< 

 Snyder has called my attention to the fact that according to K> -.MM \iin. and 

 Mag. N'at. Hist., 1905, Vol. XV., p. 20) Minous inermis should be Alhiotn m mo- 

 dactyhis (Bloch and Schneider). 



