298 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.58. 



broad as long; slightly constricted at base; barely separated by the 

 hyaline conus. 



Head small, rounded, well elevated dorsally, passing in front into 

 a stocky, slightly tapering snout of only moderate length. Eyes 

 large, distinctly constricted between the large, spherical optic ganglia 

 and the almost spindle-shaped, strongly ventrally produced bulbi; 

 eyestalks quite stout, transparent, larger than and about twi':-e as 

 long as the snout. Ocular photophores not yet certainly identified. 

 Funnel moderately large, reaching just past the base of the eyestalks. 



Arms normally developed, but small, the third pair, which are 

 largest, but little longer than the snout; formula of relative length, 

 3, 4, 2, 1. Suckers small, round, the two rows well separated, but 

 the suckers rather crowded in the row. 



Tentacles long and stout, their clubs long, slender, tapering, and 

 not expanded except for a rather wide, delicate, membranous keel 

 on the distal portion. Suckers small, numerous, in four rows on 

 club, whence they pass on the carpus into two zigzag rows which 

 extend perhaps half way down the stalk. 



Total length, about 43 mm,; length of mantle (dorsal), 22 mm. 



Type.— Cat. No. 338697, U.S.N.M. [S.S.B.6271. 



Type locality.— 100-0 m.. Station 10208, North of Little Bahama 

 Bank; March 21, 1914. 



Remarlcs. — This strange cranchiid differs from S. melancliolicus 

 Chun in the strongly keeled tentacle clubs, more elongate and taper- 

 ing body, relatively larger fins, and the somewhat difl'erent shape of 

 the eyes. From the preceding species it is readily separable by 

 reason of its shorter eyestalks, strong ventral elongation of the bulbi 

 of the eyes, strong keels on the tentacle clubs, and larger fins. 



Although the original diagnosis of Sandalops has to be considerably 

 modified to permit the reference to it of the two species here de- 

 scribed, they appear to show better affinity with this than other 

 cranchiid genera. Indeed one can feel by no means confident that 

 & full ontogenetic series will not show all three of the named forms 

 to be merely developmental stages of a single species. To proceed, 

 however, on such a gratuitous assumption, unsupported by any 

 direct evidence from the specimens themselves, would certainh', in 

 view of the many discrepancies manifested, be very unwise. Sepa- 

 rate treatment is obviouslj' the safest plan for the present. 



Genus PYRGOPSIS de Rochebrune, 1884. 



PYRGOPSIS LEMUR, new species. 



Plate 16, fig. 5. 



Diagnosis. — ^A cranchiid of small size, with an elongate, spindle- 

 shaped body, most inflated near the middle, thence tapering to a 

 sharp point between and slightly past the large, nearly semicircular 

 fins. Texture of mantle membranous, the gladius evident as a hj^a- 



