324 THE PSYCHROLUTIDiE. 



Several questions remain to be determined with respect to the repre- 

 sentatives of the fishes thus noticed: (1) What are the relations of the 

 respective genera? (2) What are their characteristics? (3) Are the 

 fishes of New Zealand and South America the same ? The possession 

 by the National Museum of the specimen described by Jordan and Gil- 

 bert as Psychrdlutes paradoxus and an authentic one of NeophrynichtJiys 

 latus permit a comparison of the two types, but, being unique to the 

 Museum, the rules do not permit their dissect ion. The questions can 

 therefore he in part, and only in part, elucidated. 



The characteristics that have been given as differentiating the genera 

 Psychrolutes and Neophrynichthys probably do not exist in nature. As 

 Professors Jordan and Gilbert have shown, a spinous dorsal is really 

 developed in their Psychrolutes; there is a spinous dorsal of short, 

 slender flexible spines imbedded in the skin and scarcely visible. There 

 are eight of these spines. On the other hand, the distinctness of the 

 spinous dorsal in Neophrynichthys has been exaggerated. In that genus 

 the spinous dorsal is obscure, externally, as in Psychrolutes, and it is 

 only when the skin is upraised that the spines can be seen and enumer- 

 ated. There is in this respect probably no difference between the typi- 

 cal Psychrolutes and Neophrynichthys. 



In fact, the two genera appear to be nearly related, but the relations 

 of neither are with the Gobiescoidte or any other form to which Dr. 

 Giinther has approximated them. Nevertheless, the genus Neophry- 

 nichthys manifests a decided general resemblance to a Batrachid both 

 in physiognomy and the loose skin with which the body and tins are 

 invested, and it is not at all surprising, ami, indeed, quite natural, that 

 a superficial observer who merely looked at the outside should be mis- 

 led by the resemblance and refer the family next to the Batrachids. 

 As has been already pointed out, however, it is with the C<>tti<l<v alone 

 that we have to compare them. If the skin is cut and lifted up from 

 the cheek of Neophrynichthys, a distinct suborbital stay is revealed ; 

 that stay is undoubtedly, as in the case of C<>t1i<l<c generally, the en- 

 larged third suborbital bone, and, likewise, as in the Cottidce,it ob- 

 liquely crosses the cheek and is attached to the inner angle or margin 

 of the pre-operculumin the manner that is characteristic of those fishes. 

 The genus Neophrynichthys and consequently also the genus Psychro- 

 lutes must be referred to the neighborhood of the Cottidce. Whether 

 the two genera really belong to that family can not be ascertained un- 

 til an examination is made of the skeleton. It is quite possible that 

 they may then prove to really represent a peculiar family. But in the 

 mean time, while no characters of anything like family value are appar- 

 ent, it is advisable to follow in the footsteps of Professors Jordan and 

 Gilbert and associate them with the Cottidce. In that family, how- 

 ever, we can isolate them in a distinct group or subfamily under the 



