BENEATH TROPIC SEAS 



arbitrary 100% of the largest fish. The elongated, 

 thread-like posterior ray of the dorsal fin is devel- 

 oped relatively late in life, the percentages of 

 this character in the same three individuals being 

 2.7%, 14.6% and 100%. In weight, the relations 

 between the three sizes of fish are tremendously dis- 

 proportionate. The actual weights are as follows : 

 the three-inch fish, five grams; the eight-inch fish, 

 sixty -five grams; and the forty -inch fish, thirty -six 

 pounds. This makes the percentages: 



Five-inch = .00031% or 32T6 of the large fish. 

 Eight-inch = .4% or -^io of the large fish. 



All my inquiries in regard to tarpon breeding in 

 Florida have been answered vaguely. There seems 

 to be a general agreement that the young are 

 hatched or live up the rivers or in ponds of fresh or 

 brackish water. Not only would it be of great 

 interest to have definite information concerning the 

 breeding, eggs and young of tarpon, but it would 

 clear up the matter of their near relationship to 

 those fishes which pass through a larval or 

 leptocephalus stage. ^ 



For the sake of helping in the identification of 

 this most interesting phase, I have translated an 

 article by Dr. van Kampen on the suspected larva 

 of the East Indian tarpon. Together with the 

 drawing of the larva itself, this will be found in 

 Appendix F of the present volume. 



' See note at end of Appendix F. 



74 



