348 FISHES OF WESTERN SOUTH AMERICA 
cases tuberculate. In O. luteus I find the early stages most readily identifiable by 
a somewhat adult-like body form. 
The genus is noteworthy for the great variation of scale-form, varying with 
the species, with body regions, and with age. The scales of the posterior and middle 
parts of the body tend toward the concentrically striate form. From about the 
eighth scale-row posterior to the opercular aperture forward to the head and up- 
ward to the nape the scales of most species are modified. In certain species they 
are thickened, enlarged, and polished to a luster rather than glossy; this type is 
often marked in part with pores. Another type is of a granular nature, nowhere 
smooth. The granules are small, tend to form definite rows around the posterior 
margins of the scales back to about eight rows beyond the opercle; each scale with 
scattered granules upon its surface elsewhere. There appears to be a definite 
correlation between the granular scalation and the heavier types of pharyngeal 
teeth, as though the mollusk-eating forms were subjected to more wear from rocky 
shores than other species. The special scale-forms do not become evident until 
the juvenile stages are outgrown, and increase with age. The scales of the head, 
the back, and those of the posterior part of the body are often deciduous, especially 
in the young, becoming firmer with age, and more regularly placed in rows. The 
specialized scales of the anterior regions, however, increase in size and decrease in 
number, losing their regularity. 
Branchiostegal rays appear to be always 5. Fins are usually rather uniformly 
placed, generally broad, rounded, with the origin of the dorsal at almost exactly the 
middle point between the extremity of the opercle and the caudal base; the caudal 
is nearly always broad, but varies from slightly rounded to slightly concave, some- 
times truncate; never deeply cleft. The anal is always about two rays behind the 
dorsal. The pectoral generally reaches about half way from its own base to the 
origin of the anal. Ventrals wanting in all species. 
In a few species the anal fin is often found sharply deflected in the larger 
specimens toward the right, enclosing a space along the right side of its base. There 
is a fold of skin, apparently somewhat deciduous, running along the base of the 
rays. The first ray is more or less surrounded by the border of the genital pore. 
The whole apparatus appears to be an elementary mechanism for spawning or egg- 
laying, and is associated with a backward-directed vent in forming an almost papil- 
lose, fleshy area. 
Two types of fin-structure appear to prevail: a type of fin having generally 
longer, more slender rays, more deeply imbedded in skin, smooth, rigid, more 
mucoid; the other type with fins longer, rays shorter, less glandular, more flexible, 
darker, harsher to the touch. 
The gonad appears always to be on the right side of the abdominal cavity, 
single, imperceptibly divided at the tip or not at all. In all the Valenciennes species 
except cuvieri, his figures convey an exaggerated idea of the size of the mouth and 
the degree of its elevation. 
The gill-membranes are more or less united, the rakers a rather uniform series, 
small, simple, widely spaced; the pharyngeal teeth not variable, with a shoulder. 
