122 On Changes of Form in Fishes. 



Danish memoir), which was captured in the western part of 

 the Atlantic in the neighbourhood of Brazil ; it is 34-37 

 millims. long, discoid, nearly orbicular, colourless, with a 

 silvery band, &c. The most serious objection that could be 

 raised to this interpretation is the presence of a very different 

 young form (pi. v. fig. 3), not larger, and sometimes even 

 smaller, which, however, notwithstanding its small size, is 

 already in a comparatively more advanced stage, transitional 

 between Acronurus and AcanthuruSy and which must with 

 absolute certainty be referred to A. cceruhus. Whether this 

 apparent contradiction arises from the circumstance that we 

 have to do here with different though nearly allied species, or 

 is due to the fact that the metamorphosis may take place a 

 little earlier or a little later, is a question which I shall leave 

 undecided for the present. Another, younger specimen of the 

 same form, perhaps of the same species, but captured N.N.E. 

 of the Bermudas, and characterized especially by the compa- 

 ratively enormous development of the anterior (strictly the 

 second) spine of the dorsal and anal, which gives these little 

 nearly rhomboidal fishes a very peculiar aspect, makes known 

 to us the ^^Acromnms " phase at a period still less advanced, 

 and which cannot be very far distant from the time of exclu- 

 sion from the egg. 



As a contribution to the evolution of the Acanthuri I must 

 also cite the change which the dental apparatus undergoes in 

 A. strigosus [ctenodon). The adult fish presents this peculi- 

 arity — the teeth are pectinated only on one side ; the young- 

 individuals still in the '■''Acronurus " phase have them pecti- 

 nated on both sides. As these little fishes pass from the stage 

 of Acronurus to that of Acanthurus the teeth with unilateral 

 pectination make their appearance and predominate over those 

 with double pectination. 



18. FiSTULARiA viLLOSA ; Centriscus velitaris and 

 BREViSPiNis ; Centriscops and Orthichthys. 



Fistularia viUosa of Klunzinger is only a young form of 

 F. serratUy Cuvier. The small close-set spines which clothe 

 its skin occur also in young examples of F. tahacaria. It is 

 not easy to differentiate these two species (of the east and 

 west) in consequence of the modifications which their propor- 

 tions undergo during growth &c. ; but it is still more difficult 

 to distinguish the two forms of Aulostomus, the specific value 

 of which seems to me very doubtful. 



Centriscus gracilis, Lowe, of which our Museum possesses 

 several young examples from the Atlantic, south and north of 

 the equator, must almost be regarded as a pelagic species. 



