150 Natural History Bulletin. 



eral contour of the fishing-frogs, Lophiid.^. but the position of 

 the gill-openings would exclude it from that family as defined 

 bv Jordan & Gilbert. The head is enormous in proportion to 

 the rest of the body, and notably depressed. The mouth is- 

 wide, opening verticalh'. The first dorsal is represented by a 

 pecuiiarh' modified spine, which is short and fleshv. hinged 

 like the '-fishing-rod"' before described, and fits nicely, when 

 not erected, into a distinct oval pit between the eves The 

 whole surface of the body is covered with prickly spines. 

 There is a curious system of lines of pores, more extensive 

 than in anv other fish that I have examined, reminding one 

 somewhat of Poricthys. One line borders the upper lip and 

 curves rather abruptly backward at the corners of the mouth, 

 until it joins a transverse line to be described presently. A 

 pair of lines start from the middle of the upper lip, embrace 

 the pit containing the first dorsal, diverge back of the e3es, 

 until they attain the level of the eyes and then pass directly 

 backward, curving down over the gill-openings until they 

 reach the lower margin of the tail, along which the\" run as 

 far as the base of the tail-fin. A third series of pores starts 

 from the middle of the lower jaw, forming a horse-shoe with 

 the ends awav from the jaw. and connected by a short line of 

 pores. . At each end of the horse-shoe a line extends outward 

 and backward to nearly the centre of the body, or just a little 

 in front, but much above, the ventral fins. At this point a 

 transverse line passes over the back across the dorsal lines. 

 and meets its fellow in the centre of the back. The ventral 

 fine extends almost directly backward from the origin of the 

 tranverse line, and ends on the proximal joint of the pectoral 

 fin. The bodv of this strange fish is capable of great infla- 

 tion, being almost all mouth and stomach. The soft dorsal 

 begins above the pectorals and almost reaches the caudal. 

 The anal is much shorter: the ventrals are very small. There 

 are no fleshy tags on the surface of the body except on the 

 first dorsal spine. The specimen is three and three-fourths 

 inches Ion"; and one and five-eighths inches broad, and one 

 and three-eiffhths inches in hei";ht when moderatelv distended. 



