230 
BULLETIN OF THE UNITED STATES FISH COMMISSION. 
CICHL1D/E. 
Individuals of the genus Heros are as numerous in the streams of Cuba as individuals of the Cen- 
trarchidn ■ are in the streams of equal size in the Ohio valley. They were found by us down to tide 
water, but not in it. Only a single species has been recorded from Cuba, and nothing has been said 
either concerning its distribution or its variation. No one, except possibly Poey, has before this com- 
pared numbers of specimens from different places or even from the same place. Such a comparison 
is therefore very desirable, and the material collected far surpasses all other collections made before. 
We have altogether 236 specimens from various localities. An examination of all of these proves 
either the presence of several instead of a single species on the island or a remarkable variation with 
localities. A definition of the variations has proved very elusive. The numbers of fin rays and scales 
are uniform, so that the differences exist in the proportions and the color. But the coloration also 
has a certain underlying uniformity. There is a spot near the middle of the side, another at the base 
of the caudal, and an obscure third above the gill-opening. There are numerous small spots on the 
fins and on scales of the sides, especially below and on the opercles, and sometimes on the cheeks. 
There is also a longitudinal streak from the eye through the lateral spot to the caudal spot, and a 
definite number of crossbars, both streak and bars most conspicuous in the young and in light- 
colored adult individuals. This uniformity of underlying structure makes defining of species or varie- 
ties a difficult proceeding. The polymorphism is further complicated by instances like the following: 
The specimens from San Antonio are readily referable to a certain form found at Calabazar, although 
they differ from Calabazar specimens in quite readily distinguishable features; but one of them differs 
notably from all other specimens collected at San Antonio, and would unhesitatingly be considered a 
species distinct, from the other specimens from the same locality. But at Palacios the same form 
branching from the Calabazar form approaches the characters of the single specimen from San Antonio. 
I venture to describe here certain of the most aberrant forms as new, without, however, feeling 
sure that they are really distinct varieties or species or that some of the other forms referred to II. tetra- 
canthus are not also new. 
Heros tetracanthus Cuvier & Valenciennes. * 
Heros tetracanthus torralbasi Eigenmann, new subspecies. 
(25 specimens, 60 to 181mm. long, from Calabazar.) 
These specimens come from the Almendares River, and as this flows near to Habana it is very 
probable that the type of Ileros tetracanthus came from the same river. Cuvier & V alenciennes say that 
Poey’s drawing, on which their tetracanthus was based, resembled Ambloplites in outline, and possessed 
spots in the angles of the scales. This very well describes some specimens I have (figs. 12 and 13). 
