OBSERVATIONS ON FISH SCALES. 
129 
Agosia Girard, the western dace. Proc. Biol. Soc. Wash., xxn, p. 157. Girard, Ichthyology of the 
Mexican Boundary Survey, pi. xxvn, fig. 6-8 ( notabilis=oscula ); pi. xxvn, fig. 10-12 (oscula); 
pi. xxviii, fig. 6-8 ( chrysogaster ); pi. xxvm, fig. 2-4 ( metallica-chrysogaster ). Girard’s figures show 
that the scales of the subgenera Apocope and Agosia are essentially of the same type, more or less 
elongate, with basal as well as apical radii. According to the figures, the scales of A. chrysogaster 
are distinctly different from those of A. metallica, the former rounded, the latter quadrate. Is it 
not probable that there are two species or races? 
Algansea Girard. (PI. xxxn, fig. 3.) Proc. Biol. Soc. Wash., xxiv, p. 213 {A. sallcei)', Zoologischer 
Anzeiger, xxxvi, p. 476 ( tincella ); Biological Bulletin, xx, p. 373 ( tincella , ajffinis, stigmatura). 
Girard figures the scales in Ichthyology of the Mexican Boundary Survey, pi. xxvn, fig. 2-4. They 
are of the same general type as those of Agosia , longer than broad, with basal as well as apical radii. 
According to the scales, A Igansea does not seem near to Hyhognathus (the scales of which are large and 
broad, without basal radii), but should be an herbivorous representative of the Agosia-Rhinichthys 
group, or related in some way to Temeculina. Among the herbivorous groups, it is by its scales 
nearer to the Chondrostominae and Chrosominae than to the Pimephalinae. May we not regard it as 
a modified chondrostomine, pointing somewhat in the direction of the Agosia group? In any case 
the scale characters must be considered relatively primitive and indicative of affinity with Old 
World types. 
Campostoma Agassiz, the stone-roller. Proc. Biol. Soc. Wash., xxn, p. 122. Girard, Ichth. Mex. 
Boundary Survey, pi. xxv, figures scales of C. ornatum, anomalum (as nasutum) and formosulum; 
they are all broad, with nucleus far basad of the middle, and apical radii only; the general type of 
Notropis and many other American genera. This genus is the type of a distinct subfamily, pecu¬ 
liar to America. 
Chrosomus Rafinesque, red-bellied dace. Biol. Bulletin, xx, p. 370, pi. 1, fig. 5; Proc. Biol. Soc. 
Wash., xxn, p. 121. Scale broader than long, with radii all around; the type that of the European 
Phoxinus. An isolated genus in America. 
Cliola Girard. Proc. Biol. Soc. Wash., xxn, p. 189 (C. smithii). Girard, Ichth. Mex. Boundary 
Survey, pi. xxxi, fig. 22-24 (velox=vigilax). The scales are broad, with about 19 apical radii in 
C. smithii, about 8 or 9 (according to Girard’s figures) in vigilax. The genus seems to stand between 
Pimephales and Notropis. 
Cochlognathus Baird & Girard. Girard, Icth. Mex. Boundary Survey, pi. xxxv, fig. 15-17, shows 
very broad scales with subbasal nucleus, the radii only apical, about 10 or 12, not nearly reaching 
the nucleus. According to the figures the basal margin is strongly convex, not straight as shown 
in Cliola vigilax. Thus the base of the scale is practically as in Lavinia, while that of C. vigilax 
may be compared with Ericymba or Phenacobius. 
Couesius Jordan, the chub-minnow. Proc. Biol. Soc. Wash., xxn, p. 159. Near to Semotilus; apical 
circuli angulate in the middle. 
Ericymba Cope. Proc. Biol. Soc. Wash., xxn, p. 162, pi. m, fig. 1; Biological Bulletin, xx, p. 374, 
pi. v, fig. 28 (normal scale) and fig. 29 (latinucleate scale). 
Exoglossum Rafinesque, the cut-lip. Proc. Biol. Soc. Wash., xxiv, p. 213. Scale resembling that of 
Mylocheilus. 
Gila Baird & Girard, the bony-tail. Biological Bulletin, xx, p. 374, pl.v, fig. 30 (scale of young G. 
robusta ); Girard, Ichth. Mex. Boundary Surv., pi. xxiv, fig. 10-12 ( grahami=robusta ); pi. xxxm, 
fig. 8-10 (emorii=elegans). The mature scale is more or less elongate, with a distinct basal lobe, 
the base being in general like that of the shorter scale of Lavinia. A Gila about 9^ inches long, 
collected by Messrs. E. R. Warren and J. W. Frey in the Bear River at Lily, Routt County, Colo., 
June 30, 1909, appears to be G. seminuda Cope & Yarrow. The caudal peduncle is almost 13 mm. 
in depth, the base of the caudal fin 20 mm. The fish is a typical Gila, and G. seminuda appears to 
have been based on an immature specimen. The ventral surface is without scales. Peritoneum 
rather dilute black, not spotted. Posterior division of air bladder large and broad, its breadth 16 
mm. Scales in lateral line about 96; the scales are elongate, with parallel, even concave sides; 
89970 0 —13-2 
