20. CHEYSOPHRYS. 493 



twelve for the Bengalese. It is evident from my examinatiuii that 

 the number varies in the latter. It must be considered a rule in 

 those fishes with a truly single dorsal fin, composed of a spinom and 

 soft portion, that often one or two soft rays, nearest to the spines, are 

 transformed into true spines, the number of the latter thus appearing 

 to be increased. That is the case in this species, where we find 

 specimens Mdth eleven spines and eleven rays, or others with twelve 

 spines and then with ten rays only. This view is very nicely 

 illustrated by a specimen of this (and of other) species in the British 

 Museum Collection, where the second ray of the anal fin is trans- 

 formed into a tnie spine. A variation of the spines, corresponding 

 to that of the rays, is much less observed in those Acanthopterygii 

 which have the dorsal divided by a more or less deep notch. 



3. Valenciennes says that his Ohr. longispinis from Japan and 

 Bengal has a somewhat longer second anal spine than the other 

 {Chr. berda, Val., from Pondicherry), and Bleeker states nearly the 

 contrary, viz. that the Bengal fish has that spine longer than the 

 Japanese one. There is variation also in this respect: immatiire 

 and half-grown specimens have that spine generally relatively longer ; 

 but we find specimens of the same size, from Japan and Bengal, 

 where this spine is equal. 



4. I could not observe a conspicuous difference in the height of 

 the dorsal fin, as stated by Bleeker. 



5. The examination of any of the Sparoids in diff'erent ages shows 

 that the molar teeth are subjected not only to a regular reproduction, 

 like the fi'ont teeth, but also to a difierent arrangement. Therefore 

 the number of the series of molars and theh- shape can be used as a 

 specific character only when founded on the examination of several 

 specimens. If there are several series of molars, the inner series 

 generally contains the smallest ones, and extends more or less back- 

 wards, so that sometimes a specimen appears to have one seiies more 

 than another. This is the case in the two fishes on which Sir J. 

 Richardson has founded Chr. auripes and .vanthopoda, the latter of 

 which had lost moreover one of the canines. Both the specimens 

 are stuffed skins, one-half of which only is preserved. 



15. Chrysophrys calamara. 



Calamara, Rmsell, i. pi. 92. 



Chrysophrys calamara, Cuv. fy Val. vi. p. 117; Bleeker, Verhand. 



Batar. Genoofsch. xxiii. Spar. p. 10. 

 berda, Bleeh. Topogr. Batav. and Ichth. Madura. 



D. |i. A. J. L. rat. 35. 



The height of the body is 2f-2| in the total length, the length 

 of the head about four times ; the (hameter of the eye is one-foiu-th 

 of the latter and equal to the length of the snout. The molars in 

 four or five series, the lai'gest ones in the outer series. The prjp- 

 orbital is much lower than the orbit. There are five series of scales 

 between the prajorbital and the angle of the proeopercuhun. Dorsal 



