made up essentially of cartilage and skin, I should like to 
emphasize that the rostrum, the opercular flap, and the barbels 
each. show declining rates of growth relative to the rest of the 
body. 
Discussion.—Assuming an earlier stage in the development 
of Folyodon than has yet been discovered, a stage in which recog- 
nizable rudiments of the rostrum, opercular flaps, and barbels have 
not been differentiated, and keeping the relative growth curves of 
these organs in mind, an examination of the 17 millimeter specimen 
suggests that the barbels are differentiated first and go through 
an initial period of rapid development followed by declining 
growth rates relative to the rest of the body. Subsequently, the 
rostrum and opercular flaps are differentiated and grow at high 
initial rates which decline to the rate of the rest of the body 
or lower. Since barbels, opercula, and the rostrum each originate 
from cells, or groups of cells, with the capacity for synthesizing 
cartilage, it seems likely that the growth histories of these 
organs may be expressed as an equilibrium reaction between the 
total amount of growing cartilage in these organs and the concen- 
tration of cartilage-forming substances in the blood stream which 
feeds this cartilage. 
Folyodon feeds on the same food throughout its life. 
Hence, it may be supposed that the digestive and assimilative pro- 
cesses remain the same throughout its life and, under comparable - 
environmental conditions, that the amount of cartilage-forming 
substances put into the blood stream is closely proportional to 
the size of the body. Thus, we may expect that the actual concen- 
tration of these hypothetical cartilage-forming substances in the 
blood stream is greater in early stages of development, when there 
is little or no cartilage, than in later stages when there is 
relatively much cartilage to use them up by its mere subsistence. 
The rate of growth of these cartilaginous parts may then be expect- 
ed to decline to the rate of the rest of the body, 
. While the assumption of an unutilized excess of cartil- 
age-forming substances may account satisfactorily for the high 
initial growth rates of rostrum, barbels, and opercular flaps, and 
for the subsequent decline in the relative growth rate of these 
parts, it must be modified to account for the fact that the barbel 
curve has a slope of 0.3 in the same’ fishes which show rostrum and 
operculum slopes of approximately 1. In order to account for such 
differences in relative growth rates we may suppose that there are 
intrinsic differences in their component cells as regards their 
rate of multiplication, and since the barbels are attached to the 
rostrum, we may suppose that these intrinsic differences arise by 
some process akin to starvation. The same sort of explanation may 
be used to account for the decline in the rate of growth of the 
rostrum below that of the rest of the body. corheges shows an 
increasing proportion of rostrum until the fis as a total length 
of about 250 millimeters after which the proportion of rostrum 
declines. It has already been mentioned that the heterogonic 
