64 



praBoperculum ; prseoperculum subrectangular, posterior border entire, 

 membranous, angle obtusely rounded ; operculum scaly. A row of elon- 

 gated scales on each side of the vertical fins, forming sheaths at their 

 bases; two elongated scales between the ventrals. Lateral line inter- 

 rupted ; tubules simple. The soft portion of the dorsal fin higher than 

 the spinous; spines subequal; second anal spine the longest ; the pos- 

 terior rays of the soft dorsal and anal prolonged. Pectorals rounded, 

 seven times in the total length. Ventrals pointed; their length is con- 

 tained 8f times in the total length. Caudal rounded. Color (in alco- 

 hol) olive-green, diluted inferiorly, with seven blackish, longitudinal 

 bands, which increase slightly in breadth posteriorly; the four upper 

 bands commence above the level of the inferior border of the orbit; the 

 two following begin in the axillary space; the lowest indistinct. Five 

 of the bands are continued on the caudal peduncle; the upper and lower 

 cease opposite the termination of the dorsal and anal fins. A narrow 

 white line on the middle of the snout, extending from between the eyes 

 to the extremity of the snout. Short linear spots on the dorsal fin ; the 

 rays and spines green. No ocellus on the upper part of the base of the 

 caudal, or anywhere on the caudal peduncle. 



Total length, 4.20 inches. 



Locality: Honolulu, Oahu. 



This fish is so characteristic that we do not hesitate to pronounce it 

 hexatcenia, although there are a great many discrepancies between the 

 above description and those given by Bleeker and others. We find 

 three series of scales on the cheeks, instead of two, and six canine teeth 

 in the upper jaw, instead of eight, as is given by Bleeker. Kner and 

 Steindachner found the hinder border of the prseoperculum finely ser- 

 rated, while in our specimen it is entire and membranous. According 

 to Bleeker the diameter of the eye is one-third of the length of the head, 

 and the length of the superior maxilla one-fourth of the same; the pec- 

 torals 5§ in the total length, and the ventrals six times in the same. 

 We find a discrepancy between the text of this author and the meas- 

 urements of his figure. The latter accord more nearly with our own 

 measurements. In the present specimen there is no evidence of the 

 ocellus at the base of the caudal, which all other writers have observed. 



