Fish Fauna of Johnston Is. — Gosline 
Of the five individuals more than 6 inches 
long from Johnston two, 6.4 and 8 inches 
long, are mature or maturing females. The 
ovaries are elongate, paired organs without 
sharp edges and contain small elongate eggs. 
The three other specimens, 7 to 7.4 inches in 
total length, contain in the ovarian position 
and behind the liver, flat, sharp-edged organs. 
The extent to which these structures are de- 
veloped in the three individuals varies greatly. 
At one extreme they are small and leaf like. 
At the other they are somewhat larger than 
the largest ovary in the females mentioned 
above, rather thick, and overlap the intestines 
at the sides, above, and behind. These large 
structures must, I think, be identified as ripe 
testes. 
The brown Scarus dubius , like the other two 
species to be dealt with, seems to have a 
green counterpart. Before discussing this, it 
seems advisable to say something about green 
parrot fishes in general based on experience 
with the scarids in Hawaii and elsewhere. 
Young parrot fishes, up to about 2 inches in 
length, are frequently, perhaps always, a plain 
light green color. These gradually become 
brownish with growth and I have never seen 
a green scarid between 3 inches and about 
7 inches long. Green parrot fishes (more than 
7 inches long) invariably have adult char- 
acteristics. In all the three species dealt with 
here, the size of the green counterpart is 
comparable to or somewhat larger than the 
mature brown form. Thus the green counter- 
parts of small brown forms, e.g., S. dubius and 
S. sordidus , are always relatively small while 
those of large brown forms, e.g., S. ahula 
( = perspicillatus) , are always relatively large. 
I have never seen a female green scarid, 
though, as just noted, some brown specimens 
appear to be adult males. Finally, green par- 
rot fishes seem to be relatively rare as com- 
pared with brown individuals, though the 
distinctiveness in the color markings of the 
green as contrasted with brown forms has led 
to the description of numerous green species. 
The green counterpart of S. dubius is rep- 
457 
resented in the University of Hawaii collec- 
tions by four specimens 170 to 190 mm. in 
total length that died in the Honolulu Aqua- 
rium and by one taken at Kailua, Kona, 
Hawaii. The aquarium specimens are so badly 
damaged that an exact correspondence in 
many morphological characters between them 
and the brown S. dubius could not be checked. 
They do agree in having two scale rows on 
the cheek, a relatively long horizontal limb 
to the free preopercular border, 13 pectoral 
rays, pointed ventrals, and a rather small head. 
In only two of the four could gonads be 
found. In these there were relatively small, 
elongate, sharp edged organs without eggs 
that must be considered testes. 
The 12 inch Kailua specimen is in better 
condition and hence of greater interest. It is 
an unripe male; I can only find a testis on the 
left side. The description of the morpholog- 
ical characters given for the adult brown S. 
dubius fits this specimen completely. The 
mouth is closed and the upper lip projects 
forward as a flap over the whole upper tooth 
plate. The pectoral rays are 13 on each side 
and there are two complete rows of scales on 
the cheek. There is a single outwardly-pro- 
jecting canine on each side of the upper jaw. 
The caudal is lunate and the outermost soft 
pelvic ray extends well beyond the others. 
Jordan and Evermann (op. cit.) recognize six 
green species of Scarus from the Hawaiian 
islands: Callyodon ’ perspicillatus (p. 347), C. 
jenkinsi (p. 353), C. gilberti (p. 354), C.formosus 
(p. 355), C. lauia (p. 355), and C. bataviensis 
(p. 356). Of these Scarus perspicillatus has the 
lower row of cheek scales incomplete, and 
Brock and Yamaguchi (1954: 154) have al- 
ready demonstrated that it is the adult male 
of "S. ahula.” The color description and plate 
given by Jordan and Evermann for Scarus lauia 
is decidedly dissimilar to that of the specimen 
at hand and seems to represent an entirely 
different species. As for the remaining forms: 
S. jenkinsi , S. gilberti , S. formosus , and S. bata- 
viensis , there is nothing in Jordan and Ever- 
mann to indicate that any or all of them are 
