observations on fish scales. 
169 
Suborder Gobioidea. 
GOBIll).®. Gobies. 
Jordan remarks that the Gobiidse have no near relations among the spiny-rayed fishes. Gunther has 
figured the scale of Gohius, and I am indebted to Dr. Jordan for material of the Japanese Mapo fuscus 
(from Tanegashima), Ctenogobius virgattihis (from Shiogama) and C. hadropterus (from Tanegashima). 
These fishes have basal radii as in the acanthopterygians, but the nucleus is at the very apex of the scale , 
and the marginal teeth form a single apical row as in the Aphredoderidse; exactly, also, as in the ctenoid 
African Characinidse. In the ctenoid African characinids the basal radii are not developed, or rather 
are represented by mere folds; in A phredoderus they are 
very distinct, but not numerous ; in the Gobiidse they are 
very numerous and close together. In the gobiids the 
relatively short (though very strongly developed) apical 
teeth are more like those of the ctenoid characinids than 
those of Aphredoderus, and the characinid Distichodus , 
with its more or less pointed or angled apical margin, 
suggests the very strong angulation of that margin seen in 
the Gobiidae. Percopsisix^s the apical nucleus and angled margin of the gobiids, but the marginal teeth 
are less developed, and the basal radii are absent. 
The gobiid scales before me may be separated as follows: 
Lateral margin (from apicolateral to basolateral angles) somewhat larger than apical margin of one 
side; some radii lateral (entering margin above basolateral comer); radii about 22 . .Mapo fuscus 
Lateral margin shorter or not longer than apical margin of one side ; no radii lateral i 
I. Lateral circuli quite dense; radii about 30, some imperfect. 
Ctenogobius virgatulus Jordan & Snyder 
Lateral circuli not dense; radii about 14, some imperfect. 
Ctenogobius hadropterus Jordan & Snyder 
Latinucleate scales are common and are of course much modified, though they retain the character- 
istic margin. 
Suborder Discocephali. 
ECHENEID^. Remoras. 
I had examined the shark sucker, Leptecheneis naucrates (Linnaeus), collected at Woods Hole, Mass., 
and concluded that there were no scales. Dr. E. Linton, who was working on the parasites of this fish, 
suggested that scales might be found in the skin, and so it proved. The completely embedded scales 
have an elongate-lanceolate form, pointed or obtuse at the ends, length about 2^4 mm. The sculpture 
consists only of rather widely spaced concentric circuli. No other similar scale is known to me, 
though in the matter of sculpture (though not at all in form) the scales substantially agree with those 
of Lota. 
Fig. 41. — Leptecheneis naucrates CEcheneids) . Apical 
end to the right. Bureau of Fisheries. 
