THE OSTEOLOGY AND RELATIONSHIP OF LEIOG- 

 NATHUS, A GENUS OF SCOMBROID FISHES. 



BY EDWIN CHAPIN STARRS. 



Doctor Boulenger, in his arrangement of fishes in the Cambridge 

 Natural History, has eliminated the family Leiognathidae from among 

 the Scombroid fishes, where it hitherto has been considered, and has in- 

 cluded it in the family Gerridae, a family of more or less evident Percoid 

 affinities. 



That Doctor Boulenger had some reason for this decision any one 

 will admit who makes a comparison of the mouth parts of the members 

 of these two families. Comparing then the genus Gerres with the genus 

 Lciognathus, which are typical representatives of their respective fami- 

 lies, we find these parts almost identical. Both have a long mandible, 

 with a concave lower outline, and a small mouth. The ascending limb 

 of the premaxillary is exceedingly long, extending over the top of the 

 head between the eyes and allowing the premaxillaries to be drawn out- 

 ward or forward to an enormous degree. The posterior tip of the maxil- 

 lary lies below the preorbital as an ovoid plate. When the premaxil- 

 laries are drawn forward the maxillaries are inclined somewhat for- 

 ward also. 



But aside from the mouth parts these genera have little in common. 

 Gerres has large scales over the head and body. Leiognathus has very 

 fine scales on the body, and the head and breast are naked and covered 

 with silvery skin suggestive of that of the Scombroid fishes. Gerres has 

 a very short anal base ; Leiognathus a very long one, and the soft rays of 

 both the dorsal and anal fins are arranged each in a step or notch, as in 

 so many of the Scombroid fishes, indicating a tendency towards a break- 

 ing up of the fins into finlets. 



The following investigation of the osteology of Lciognathus was un- 

 dertaken (1) to learn its skeletal characters; (2) to compare them with 

 those of Gerres; and (3) to ascertain what, if any, its Scombroid affin- 

 ites might be. 



(5) 



