No. 465.] STUDIES OF MYXINOIDS. 631 
1 o. of 
Individuals Gills Percent 
4 10-11 4l 
236 II 24.63 
95 11-12 9.92 
574 12 59.92 
33 12-13 3-44 
16 13 1.67 

A total of 958 individuals certainly gives a reasonably safe 
number on which to base conclusions. In this table as in the 
others, the fact stands out prominently that in the California 
hagfish twelve gills on each side is the number most commonly 
found. But when the number having it only amounts to 59.92% 
in a total of 958 eels, it cannot well be called the normal, or even 
the usual number, and consequently I agree with Dr. Ayers that 
the usual number of gill slits is eleven or twelve. 
That 13.7% in a total of 958 eels is found where the number 
of gills is greater on one side than on the other, is to me one of 
the most significant features in the proper specific classification 
of the Bdellostomids. With six and seven gills the prevailing 
number in Bdellostoma forsteri, eight in that found in Japan, 
ten the number in the Chilean form, and eleven and twelve 
in California, varying on the one hand, though rarely, to ten, and 
on the other hand more frequently to thirteen, it is surely no 
longer possible to divide these animals into different genera and 
species on the basis of the number of gills alone; the count 
of teeth (Ayers, '94), is equally unsatisfactory as a ground for 
division into species, and no other ground for such division has 
ever been advanced. 
As a matter of fact, the differing number of gills in the 
California hagfish is even more significant when the animal is 
examined more closely, externally and internally, than when the 
external apertures alone are counted. On dissecting the gill 
region of those animals with an unequal number of gills on the 
two sides, we find not merely that the number is variable, but 
also where the variation is most apt to occur. Referring back 
to my table, and noting the fish with an unequal number of gills 
