SPEARFISHES OF THE CENTRAL PACIFIC 



511 



60 



< 



O 40 - 



60% 



MAKAIRA 

 MAKAIRA 



AMPLA 

 AUDAX 



■- a ISTIOMPAX MARLINA 



Figure 14. 



40 50 60 



SNOUT TO ORBIT (CM.) 

 -Relation of length of mandible (tip to angle of jaw) to snout from tip to orbit, 

 and Conrad, 1939, and Conrad and LaMonte, 1937.) 



100 



(Measurements from Gregory 



divergence among species is apparant; audax and 

 marlina show sliglitly positive allometric growtli 

 of the mandible in relation to the snout, whereas 

 ampla shows a slightly negative allometric growth. 

 Unfortunately, the POFI measurements of the 

 mandible (to the joint) are not comparable to the 

 measurement used h\ these authors, so compari- 

 sons between areas are not possible. 



Obviously, here is a character that is useless for 

 separating the species among the intermediate 

 sizes, but the divergence among the very large 

 specimens suggests that, in them, it may be useful 

 for distinguishing ampla from marlina. The 

 length of the mandible to the angle of the jaw, as 

 measured by Gregory and Conrad and by Conrad 

 and LaMonte, is preferred to the measurement 

 used by POFI; also, it may be measured with 

 considerable precision from photographs. The 

 plots of the published data suggest that specimens 

 of more than about 600 pounds in which the 

 mandible is more than 48 percent of the snout 

 (that is, goes into the snout less than 2.1 times) 

 will be marlina, whereas those in which the length 

 of the mandible is less than 48 percent of the snout 

 should probably be considered to be ampla. If 

 we apply this criterion to the type photograph of 

 marlina (Jordan and Evormann, 1920: pi. 17; 

 which weighed only 509 pounds), and to all of 



435062 O— 58 3 



Farrington's (1953) photographs of black marlin 

 of more than 600 pounds in which the characters 

 can be measured, we find that the length of the 

 mandible is contained in the snout 1.5 to 1.9 times, 

 with an average of 1.76. On the other hand, in 

 the photographs of ampla of more than 400 pounds, 

 shown by Farrington (1937), the length of the 

 mandible is contained in the snout from 1.9 to 2.4 

 times, with an average of 2.09. Here is a char- 

 acter that may well be useful in distinguishing 

 ampla from marlina, when the unequivocal 

 character of the pectoral fin has not been recorded ; 

 but additional measurements of large specimens 

 are needed to establish the difference. 



Clearly, too, this difference in the lower jaw 

 is the reason for the apparent differences that have 

 been observed in the snout. When the lower jaw 

 is very short, as in large ampla, the snout seems 

 extremely long and slender, whereas the snout 

 seems shorter when the lower jaw is long, as 

 in marlina. 



Length of pelvic fin 



In our previous discussion of allometric growth, 

 we pointed out tliat there was almost no change 

 in the length of the pelvic fin witii size of the 

 fish in any of the three species examined by 

 POFI. Consequently, we may compare these on 



