NO. 3 SCALES OF THE MORMYRID FISHES— COCKERELL 3 



(rarely five) lobes. The circuli are extremely fine, and are mainly 

 longitudinal, breaking into a roughened area apically, the whole 

 structure of the circuli being extraordinarily like that of Amia calva. 

 This is a matter of interest in connection with the known great an- 

 tiquity of the Albulidce. 



Another ancient family, placed next to the Albulidae in the system, 

 is the Elopidse. Doctor Evermann kindly sends me scales of Blops 

 saurus from Jamaica. They are also subquadrate, about 6 mm. 

 diameter, with fine circuli which are complete basally, and here 

 transverse, as in ordinary fishes, whereas Albula shows in this region 

 longitudinal strands consisting of minute segments.^ The basal 

 structure of Blops is evidently more "advanced" than that of Al- 

 bula — much more like that of Teleosts in general; that of Albula 

 (except for the radii and the scalloping) differs little from Amia 

 calva. Blops saurus has a variable number (about 12 to 22) of 

 strong basal radii, which really do radiate from the central nuclear 

 area of the scale, though failing centrally. The apical region of the 

 Blops scale is very thin, and shows numerous parallel radii. 



Supplementary Note on the Scales of Hiodon. 



Since writing the above paper I have received through the kind- 

 ness of Dr. S. Graenicher scales of Hiodon tergisus, from the St. 

 Croix River, Polk County, Wisconsin. In Boulenger's arrangement 

 the Hiodontidse, which are exclusively North American, follow the 

 Mormyridse. The scales are large (about 12 mm. long and broad), 

 with the nuclear area between 4 and 5 mm. from the apex. The 

 circuli are excessively fine and numerous, the same all around, and 

 of course transverse in the basal region. Rather numerous apical 

 radii are indicated by obscure broken lines, easily overlooked. The 

 basal radii are very distinct, irregularly placed, a group of seven or 

 eight closely adjacent ones going to each sublateral concavity of the 

 basal margin. The basal margin is not crenulate, and the basolateral 

 angles are distinct. 



This is wholly diverse from Albula, very distinct from the Mor- 

 myrids, but not far from Blops. The silvery skin and minute pig- 

 ment spots are quite as in Blops. Blops differs by the modification 

 of the apical circuli, which in the subapical field become coarse, 

 wavy, and transverse, actually at right angles to the lateral ones. 



'Compare Tims on scales of Gadus; Quart. Journ. Micros. Science, Oct., 

 1905, pi. VI. See also the structure of apical radii in the Serranids Apsilns and 

 Aprion, as figured by Sauvage in his work on the fishes of Madagascar. 



