302 
BULLETIN OF THE UNITED STATES FISH COMMISSION. 
related to the two former, and seems itself closely related to E. ratius Muller. If the 
last-named species should he included in the new group it would have to carry with it 
E. carinatus Lamarck, E. lophoessus Gosse, etc. Moreover, Eattulus latus Jennings 
and E. bicuspes Pell seem related to E. multicrinis Ivellicott. On the whole, such a 
group could not he separated oft' without being open to all the objections which may 
be made to the classification on the basis of the toes. 
But the idea that a genus must represent a well-defined natural group, all the 
species of which are more closely related to each other than to any outside species, 
lias as a matter of fact been largely given up in practice. Generic divisions are 
more commonly made on artificial grounds, to break up an otherwise unwieldy 
group into convenient divisions. 
On this basis it seems to me that we may properly retain the old genera based 
upon the toes. In one group may be classed, as heretofore, those species which 
make the general impression of having a single long toe. This group must receive 
the name Eattulus. In the other group will be placed those that make clearly the 
impression of having two toes, and to this group the name Diurella belongs. 
Then arises the question as to how we shall define our two genera so as to decide 
in doubtful cases in which genus the species shall go. None of the definitions here- 
tofore given will suffice, for they have been made upon false grounds and without a 
knowledge of the real structure and amount of variation in the toes. 
We shall probably do best to frame our definition so that it shall retain in the 
genus Diurella those that have heretofore been looked upon generally as having two 
toes, while Eattulus shall include those that have generally been considered one-toed 
forms. This will be best accomplished if we define the two groups as follows: 
Diurella includes those species in which the smaller of the two toes is more than 
one-third the length of the larger; Eattulus, the species in which the smaller toe is 
but one-third or less of the length of the larger, or seems to be lacking. 
The history of our knowledge of the group shows that where, as in Diurella 
tenuior Gosse and I), insignis Herrick, the smaller toe is nearly (though not quite) 
one-half the length of the larger, the animal is naturally classified with the two-toed 
species. For this reason it is better to make the dividing point come at the propor- 
tion of one-third rather than at one-half. The division thus obtained is perhaps 
the most natural of any that could be made. The chief place where it fails is of 
course in the species that are near the dividing line, in separating such closely 
related species as Diurella tenuior Gosse and Eattulus gracilis Tessin. 
Another genus, to include species having the toes exactly equal (answering to 
Gosse’s genus Eattulus), might be recognized. But this seems to me hardly advis- 
able. The equality of the toes is only one point on a long scale of variation and 
seems, in the present case at least, not worthy of being so strongly marked. In 
our American EaMulidai this would separate from all others Diurella . tigris Gosse, 
D. intermedia Stenroos, D. sulcata Jennings, and D. cavia Gosse, Avhich certainly 
do not form a group well marked off from the other species of Diurella. If such a 
genus should be recognized it would have to receive the name Heterognathus 
Schmarda (1859), as this was the first genus founded for equal-toed forms, its type 
species, Heterognathus macrodactylus Schmarda, being without much doubt none 
other than Diurella tigris Muller. 
Specific distinctions. — As to the distinction of species, this seems not intrinsic- 
ally so difficult in the Eattulida as in some other groups of the Rotatoria, notably 
