No. 409.] CASES OF SALTATORY VARIATION. 35 
genus. It is possible that some of the genera of fishes without 
ventrals have arisen from such prepotent variants, a supposition 
that is reinforced by the following case of variation. 
3. The variation of greatest importance to be recorded here 
was found in nine specimens of Ameiurus melas. These were 
collected at random, out of a lot that must have numbered 
fifty or more, in a small cave at Glasgow, Ky. The fact that 
the specimens showed a remarkable series of variations was not 
detected until after they had been brought to the laboratory, 
so we can safely assume that they were collected at random 
and not with respect to the variations to be recorded. The 
B A 
Fic. 2. — A meiurus melas, showing supplementary narial barbels (see directive lines). 
specimens seen in the cave were all of approximately the size 
of the individuals here recorded. Unless they represent a 
dwarf race living in the small brook in this cave, they are prob- 
ably of the same age. One of us visited this cave two years 
ago, and he neither remembers to have seen any catfishes in 
the cave stream at that time, nor are any specimens in the col- 
lection made at that time. While the fact that he does not 
remember to have seen them and did not collect any two years 
ago is not conclusive evidence that there were no catfishes in 
the cave at that time, the size of the specimens, taken together 
with this, makes it probable that the present specimens are 
under two years old and got into the cave either as a batch of 
eggs or during their schooling stage. It is well known that 
the young of the catfishes remain together as a school for 
