34 PROCEEDINGS OF THE NATIO^STAL MUSEUM vol.84 



of head at posterior end of frontoparietal ridges, directed backward 

 and inward; preopercle armed with 4 spines, the upper one about 

 1.1 in orbit, directed upward and backward, slightly curved; on 

 the left side of the type it has a simple tip and two strong recurved 

 barbs along its upper margin ; on the right side both the tip and the 

 single recurved barb are bifid; the three lower preopercular spines 

 simple, sharp, somewhat longer than nasal spines, the upper two 

 directed downward and backward, the lower one downward and 

 forward. Pores of head very large and conspicuous; the suborbital 

 chain of bones bordered dorsally and ventrally by large pores; the 

 upper series continued around the orbit ; those pores that are dorsal 

 and posterior to the eye opening through the skin of the eyeball itself 

 instead of in interorbital space ; the lower series continued backward 

 on cheek to near base of upper preopercular spine, from which point 



FiGDBH 7. — Phasmatocottus ctenopterygius, new species. The true formula of the dorsal 

 fins of the type is VI, 12, not VII, 11, as shown in the figure, which may have been 

 made from a second specimen, since lost. 



it swings dorsally; pores of marginal preopercular series continued 

 forward as large simple mandibular pores, the anterior pair opening 

 in a common pit on the median line of symphysis. 



Origin of first dorsal very slightly behind upper end of gill open- 

 ing; base of fin 2.5 in head; fin of 6 detached spines unconnected by 

 membrane, the first two with approximate bases; first spine 1.3 in 

 third spine, which is longest, being 1.8 in head. Second dorsal sepa- 

 rated from first by an interspace about 2.1 in orbit; base of fin 1.5 

 in head; fin of 12 rays, first ray 1.5 in seventh ray, which is longest, 

 being 2.3 in head. Origin of anal about under second dorsal ray, 

 its posterior end under second ray from end of second dorsal; base 

 of fin 1.7 in head; fin of 1 detached spine (?)' and 8 rays. First 



^The small structure at the anterior end of the anal fin in this fish certainly appears 

 to be a spine, although it may possibly be the basal portion of a broken ray that has 

 attained a sharp point and smooth appearance. It is only the absence of anal spines in 

 all other cottids except Ochotskia armata Schmidt, with which this species shows no close 

 relationship, that causes me to question its character. It may be possible that this 

 species has retained one anal spine as a primitive feature, and only examination of a 

 second specimen can clear up this point. 



