188 Annals of the Carnegie Museum. 



the type. There is a well developed, simple, supraorbital tentacle 

 above the posterior part of the eye, one at the occiput, and two equally 

 spaced between these on each side of the head. There are two on the 

 preopercular margin, one on each nasal spine, and a very minute one 

 on the end of the maxillary. It is more completely invested with 

 spinules than was the type; the naked area along the base of the anal 

 is scarcely to be appreciated, the breast is covered, except a small 

 area in front of the ventral base, and just behind the branchiostegal 

 membrane, the head and postaxial area are naked. The lateral 

 plates are even less conspicuous than is shown in the drawing of the 

 type, and each anterior one bears a simple filament. The specimen 

 at hand agrees otherwise with the original description. 



26. Icelinus borealis Gilbert. 

 This was the commonest fish brought up in the dredge. The 

 largest specimens are four inches in length, considerably larger than 

 the typical ones. Of the three spines which are usually present in 

 the typical specimens below the forked preopercular spine, the upper 

 one is absent in these specimens, except in the very small ones. 



27. Astrolytes fenestralis (Jordan and Gilbert). 



This common species was taken with the seine in abundance. 



The soft rays of the dorsal in fourteen specimens number seventeen, 

 in three specimens sixteen, and in two specimens eighteen. In the 

 anal fourteen specimens have thirteen rays, four specimens have 

 twelve, and one has fourteen. The dorsal band of scales has from 

 thirty-two to thirty-five in the series, and the pores of the lateral line 

 number from thirty-five to thirty-seven. In two or three specimens 

 the lowermost prong of the preopercular spine has failed to develop, 

 or is very small, thus making the spine trifid. 



28. Axyrias harringtoni Starks. 



This species was found to be rather abundant in shallow water 

 where several specimens were taken with the seine in company with 

 Astrolytes and Artediiis. The largest was four inches in length. 



The following additions may be made to the original description. 

 The dorsal has seventeen soft rays and the anal thirteen in the type 

 (not sixteen and twelve), in each case the last ray being slenderer 

 than the others. This is the usual number of dorsal rays in the 



