666 



(Hydrocyon, Chalcinus, Salminus), and present in small ones(Anodui 

 edentulus, Chilodus). 



2. MORMYRJ. 



Morruyrus longipennis, Rupp. 

 Mormyrus oxyrhynchus. 

 Mormyrus bane. 



Mormyrus cyprinoides. 

 Mormyrus, spec. 

 Mormyrops anguillaris. 



3. CLUPEINI. 



Megalops cyprinoides. 



Elops saurus. 



Coilia Grayi. 



Notopterus Pallasii (corpusc. very 



Butirinus macrocephalus. 

 Hyodon claudulus. 

 Osteoglossum Vandelii. 

 Osteoglossum bicirrosum. 



scanty). Heterotis niloticus. 



The plates of the abdominal carina in many Clupeini are formed 

 throughout of true bone, but do not belong to the present category. 



I am unable to find corpuscles in the scales of Lutodeira chanos, 

 Chatoessus punctatus and cepedianus, and Alosa vulgaris. In several 

 Cyprinoids (Labeo, Catastomus, Barlus), I have, in like manner, 

 failed to discover corpuscles in the scales proper ; on the other hand, 

 I have found very distinct dentinal tubes in the scales of Barbus, at 

 their hinder part. 



True osseous tissue will doubtless hereafter be found in the scales 

 of many other Physostomi which have it in their skeleton, but it is 

 not to be supposed that this will apply to all. 



In the Physostomi, as in the Ganoids, the bone-corpuscles lie in 

 the lower stratum of the scale ; still they are situated above the 

 fibrous layer, and immediately beneath the structureless layer, to 

 which in all scales I apply the name of " ganoin-layer," inasmuch 

 as it has in all cases the same signification. 



From the foregoing observations we are able to show still more 

 positively than could be done by J. Miiller, that the scales of Ganoids 

 have no peculiarity of structure to distinguish them from those of 

 the Teleostei. Nay, certain Ganoids, as Amia, have scales, which 

 in respect even of pliancy, rounded contour, and the surface-marking of 

 the ganoin-layer, agree with those of other fishes. 



In reference to those fishes which want bone-corpuscles in their 

 skeleton, I have still to remark, 1, that the corpuscles are also inva- 

 riably wanting in the semicanals upon the scales of the lateral line ; 

 for what Leydig designates as rudimentary bone-corpuscles in the 



