106 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.43. 



Anterior dorsal low, longest ray a little longer than snout; second 

 dorsal close to first, lower in the middle, the longest rays being near 

 the caudal, tips of posterior rays reaching base of caudal; caudal 

 long, narrow, posterior margin rounded; anal similar to second dorsal, 

 its distance from tip of snout slightly less than one-third the distance 

 from base of caudal; ventrals 1.30 in head, small, far apart, tip of 

 longest ray reaching base of seventh anal ray; pectorals 1.43 in head, 

 as long as part of head behind anterior margin of pupil. 



Color in alcohol: Ground color brownish black with silvery reflec- 

 tions, hgh'test on top of head and sides of body posteriorly, darkest 

 on belly, the latter has a bluish cast and metalHc reflections ; vertical 

 fins dusky, margins and basal portions somewhat darker; paired 

 fins buff, their axils black; peritoneum and an area on posterior part 

 of gill chamber silvery white. 



Type.— Ca,t. No. 72923, U.S.N.M., 27.4 cm. in length, taken with a 

 beam trawl at station 5296 (lat. 13° 40' 09" N.; long. 120° 57'45"E.), 

 in Verde Island Passage, at a depth of 210 fathoms, on a bottom of mud 

 and sand. 



From our examples of P. japonicus Hilgendorf, which agree with 

 an example from Kagoshima, Japan, identified by Smith and Pope, 

 P. nigrescens appears to be distinct, separable on the greater number 

 of rays in the anal (66 to 72 in nigrescens, 58 to 65 in japonicus), much 

 lower fii'st dorsal, longer pectorals and ventrals, smaller barbel, 

 smaller eye, longer snout, broader interorbital, and darker colora- 

 tion. From Hilgendorf's meager description of the type of japonicus 

 it differs in having the head longer, the interorbital broader, the 

 first dorsal shorter, and the anal longer. P. edelmanni Brauer is 

 very similar in form but has a shorter head, no barbel, only 5 ventral 

 rays, 20 pectoral rays, and is of lighter coloration. 



Family MACROURID^. 



Subfamily B^THYOAIDHST^E. 



Our observations do not agree with those of Mr. Regan * as to the 

 position of the coracoid foramen in Gadomus and Melanohranchus. 

 In one of the types of G. longijilis (Goode and Bean) and in our exam- 

 ples of G. multijilis (Gunther) ^ from the Philippines the coracoid 

 foramen is not within the substance of the hypercoracoid, but between 

 the hypercoracoid and hypocoracoid as in the gadoids. From the 

 foramen a shallow but distinct fossa extends obhquely toward center 

 of each coracoid bone, its breadth and depth decreasing posteriorly. 

 In some individuals more of the foramen Ues within the hypercora- 



1 Regan, Ann. Mag" Nat. Hist., ser. 7, vol. 11, 190:% No. 65, p. 459, flg. 2.4. 



s We consider these species distinct, separable on the smaller number of pectoral rays in the former. B. 

 fuTvescens Alcock is also distinct, belonging to the genus Regania Jordan. 



