RECORDS OF W.A. MUSEUM 
[22 C 
Mr. C. Tate Regan has, at my request, very kindly compared 
Gunther's type specimen oi Melambnphes mgroris wth Steindachner’s 
figure of Girella zebra, and writes to say that he has no doubt they 
represent the same species. Castelnau {loc. cit. p. 68) has given 
reasons for supposing that Gunther’s determination of the Victorian 
fish as G/y/AiWoK Higf'om, Cuvier and Valenciennes, is incorrect. 
I fully agree with this conclusion, particularly as the latter species 
is said to generally resemble G. vakti and G. beiigaleiisis, which are 
very different in appearance to MehimbapJtes. 
1 have examined four specimens from Albany, 180-290 mm. 
long. Of these one has thirteen dorsal spines, two have fourteen, 
and one has fifteen. There are 64-70 pores along the lateral line, 
and go-ioo .scales immediately above it. A skin received from 
Count Castelnau as Neotephroeops zebra, is also in the Australian 
M useu m . 
Family CHAETODONTIDAE. 
HOLACANTHUS (CHAETODONTOPLUS) 
PERSONIFER, sp. nov. 
Plate XXXI. 
D. XIII. 19 ; A. III. 19 ; V.I.5 ; P.19 ; C.15. Body more elongate 
than is usual in the genus, the depth before the ventrals 2.3 in the 
length to the hypural. Head 4.7 in the sanie.^ Eye 3f in the head, 
as long as the snout without the upper lip. Interor’uital space 
convex, about half as wide again as the eye. Suborbital bone 
a little wider than the eye. Maxillary vertical, lower jaw 
the longer. Posterior preopercular margin finely dentate, lower 
entire ; the spine is very large, longer than the width of the 
interorbital space. Teeth cardiform, tricuspid, the median cusp 
much larger than the others ; they are arranged in about five rows 
in each jaw of which the outer ones are the longest. 
Entire head, body and fins, with the exception of the ventral 
rays, covered with minute, strongly ctenoid scales which are very 
irregularly arranged. Lateral line distinct anteriorly, scarcely 
1 The anterior portion of this specimen having been partially cleaned out and 
refilled with straw, the head is somewhat distorted, and these 
measurements may not be quite accurate. 
