148 
BUIvI^ETlN OE THE BUREAU OE EISHERIES. 
Suborder SalmopErc^. Trout perch. 
PERCOPSID^. Sand rollers. 
The trout perch, Percopsis guttatus Agassiz, from East Okoboji Lake, Iowa (Evermann), has broad 
sub triangulate scales, with the apical margin forming two sides of the triangle, each side with about 
17 sharp teeth. Nucleus very close to apex; circuli widely spaced; a slight tendency to basal plication, 
but no distinct radii. In the form of the apical region, with the nucleus very far apicad and the single 
row of sharp teeth, this reminds one of Gobius and its allies. In the Gobiidae, however, the basal radii 
are very well formed and numerous. 
Suborder BerycoidEa. Berycoid fishes. 
BERYClDdJ. Alfonsinos. 
Jordan & Fowler say of this family: “Covered with ctenoid, or cycloid, foliate, or granular scales.” 
I have from the United States National Museum scales of Beryx spiendens Lowe (fish 21 inches long) 
from Japan. They are about 8 mm. long and 9.5-10.5 mm. broad, with the exposed part colorless and 
the covered pale yellowish red. The apical margin is thin and rather irregular, wholly without teeth; 
lateral margins concave; laterobasal comers very prominent; lower margin convex, irregularly sub- 
crenulate; nucleus a short distance apicad of middle; basal circuli dense; weak basal folds in place of 
radii; broad apical region with growth lines but no circuli, and with scattered small round perfora- 
tions. These perforations in the apical region sometimes have elevated margins, and in the region of 
the nucleus their place is taken by small sharp spines. Probably they arise as spines in every case, 
and the spines breaking off leave perforations. Jordan & Fowler state that the scales of Beryx spiendens 
‘‘are fmmished with fine prickles, giving a somewhat rough touch.” In Beryx lineatus, as figured by 
Sauvage, the prickles or spines of the apical region are large and dense. In either case, the arrange- 
ment parallels that found in species of Macrurus, and it is interesting in this connection to note that 
Regan some years ago suggested that the Gadoids might have come from the same stock which also 
gave rise to the Berycidae. In Caulolepis subulidens Garman, as figured by Garman (Memoirs Museum 
Oamparative Zoology, xxiv, 1899), the scales are extremely modified, but still have some of the spines 
or teeth. 
TRACHICHYIDdl. Deep-sea berycoids. 
Garman, in the work just cited, figures the scale of Trachichihys mento Garman. It is transversely 
oval, without the laterobasal comers of Beryx, but it has a number of tme berycoid spines in the sub- 
apical region. Boulenger refers the genus to Berycidae. 
HOLOCENTRIDAJ. Squirrel-fishes. (PI. xxxvi, fig. 26.) 
The scales of the Holocentridae resemble in shape those of the Berycidae, being broad, with prominent 
lateroba^l angles. In every case, so far as known to me, the apical margin is armed with strong, straight, 
comb-like teeth. In some (especially Myripristis murdjan) the subapical region has spiniferous pits, of 
the same character as the spine-bearing holes of Beryx. The basal margin is straight or nearly so, except 
that in the middle it is thrown into one or more folds or lobes, indicative of the mdiments of a basal 
radial system. The scales are all broader than long, but those of Flammeo are not so conspicuously so as 
those of Holocentnis, and especially Myripristis murdjan; the last are larger than any of the others, about 
8 J^ mm. long and 14X broad, strongly reddish. The basal circuli in all are excessively fine, more so 
than in Beryx spiendens. The species examined are as follows: 
Myripristis murdjan (Forskfil). Strongs I., Carolines (M. C. Z.). 
Flammeo scythrops. Hawaii. 
Flammeo sammara (Forskfil). Hawaii. 
Holocentrus diadema Lac 4 pMe. Hawaii. 
Holocentrus laticeps Cuvier & Valenciennes. Kaui, Hawaii (M. C. Z.). 
