546 Dr. W. G. Ride wood on Flying-fishes. 



is connected by ligament with the anterior ends of the 

 clavicles. The nearest allies to Gastropelecus are most 

 probably the species of Tetragonopterus } of the family 

 Characidae. In Tetragonopterus, however, there is nothing 

 unusual about the pectoral girdle ; the downwardly directed 

 plates of the coracoids extend in close contact with one 

 another for some little distance, as in many other Characidae 

 (e. g., Chalcinus trachypomus ; see Regan, Ann. & Mag. Nat. 

 Hist. (8) viii. p. 19, fig. 1, B), but the plates are not fused 

 into a single median lamina, and they exhibit no radiating 

 grooves and ridges. 



More remarkable than the great size of the bony keel of 

 the pectoral girdle of Gastropelecus is the high proportion 

 which the weight of the muscles attached to the keel bears 

 to the total weight of the body. Put roughly, the right and 

 left external pectoral muscles together constitute one-fourth 

 of the weight of the fish, whereas in the allied but unmodified 

 form Tetragonopterus the proportion is 1 to 140. The details 

 of the computation are as follows : — The weight of a speci- 

 men of Gastropelecus stellatus, taken from spirit and lightly 

 wiped with a cloth, was 6*50 grammes ; the weight of the 

 right and left external pectoral muscles dissected oft' was 

 1*59 grammes. The proportion of the latter to the former 

 is 1 to 4*088. In the case of a specimen of Tetragonopterus 

 anew, selected as nearly as possible of the same size as the 

 specimen of Gastropelecus, the total weight of the body was 

 7'01 grammes, the weight of the right and left external 

 pectoral muscles was *050 gramme • the proportion of the 

 latter to the former is thus 1 to 140*2. 



So remarkable is this difference in the relative size of the 

 depressor muscle of the pectoral fin in Gastropelecus and 

 Tetragonopterus, that it seemed advisable to compare by the 

 same method the depressor muscles of the common flying- 

 fish, Exoccetus, with those of its nearest relatives, e. g. the 

 skippers, of the genus Hemirhamphus , for which no one has 

 claimed the capacity for flight. The figures given by C. D. 

 Dumford (Ann. & Mag. Nat. Hist. (7) xviii., Nov. 1906, 

 p. 337) are computed from the depth to which a pin reaches 

 when inserted into the muscle at certain points, and not by 

 weighing the muscle dissected from the body. 



There is nothing in the appearance of the bones of the 

 pectoral girdle of Exocwtus and Hemirhamphus to suggest 

 that the muscles are vastly greater in the former than in the 

 latter ; the proportions of the parts are much the same in 

 both, and there is not in either case a great keel or median 

 lamina of bone for the attachment of the depressor muscles. 



