EIGENMANN: the CHEIRODONTINiE. < 



notomelas, and C. piaba, to the clupeoid openings in Leptohrycon jatuarance, Macro- 

 psohrycon urugumjanm, Megalamphodus megalopterus, and M. micropterus. But the 

 size of the mouth in itself is of no great importance, for there is great variation 

 within such genera as Aphyocharax and Cheirodon. There is very great difference 

 in the size and shape of the premaxillary and maxillary, as the outhnes in the 

 text-figures show. 



Teeth. — The greatest interest centers in the teeth. In Grundulus all of the 

 teeth are peg-like, conical, single-pointed. Such teeth frequently appear on the 

 sides of the lower jaw, when the teeth are otherwise very different. They are also 

 frequently found on the distal part of the maxillary, when the number of teeth on 

 this bone are considerable, and they are also found in the premaxillary when the 

 teeth are feeble or numerous. The next degree of complexity is found in the three- 

 pointed teeth in the genera Spintherobolus, Probolodus, Aphyocharax, Macropso- 

 brycon, Mwroschemobrycon, and Oligobrycon. In Aphyocheirodon there are three- 

 to five-pointed teeth in the upper jaw and five-pointed teeth in the lower jaw. In 

 the remaining genera the teeth have typically five or more points. Frequently 

 the teeth in the sides of the lower jaw are not only smaller, but belong to a lower 

 order, i. e., they have fewer points than the others in the same mouth. The same 

 may be true, but to a less extent, of the teeth on the outer part of the premaxillary 

 and on the distal part of the maxillary. Usually the number of teeth in any bone 

 differs inversely as the number of points to each tooth, though this is not always the 

 case. In the species with many-pointed teeth the number of teeth is usually very 

 limited, none to three in the maxillary of Cheirodon, none to four in Odontostilbe, 

 but in one species of that genus ranging from four to seven. However, several of 

 the genera with tricuspid teeth have no teeth on the maxillary, others have as 

 many as twenty or more. 



To say that the teeth are unicuspid, tricuspid, or multicuspid, does not tell 

 the whole story. There have evidently been divergent radiations within each of 

 these groups both in the shape of the individual teeth and in their arrangement. 

 These teeth are so difficult to observe, even with the aid of the modern binocular 

 microscope and a spot-light, that in all cases where I had material the individual 

 bones were dissected out and mounted in balsam. Camera-lucida sketches were 

 then made. Under all the circumstances I think it will be best to entirely ignore 

 statements about the teeth in the older descriptions, whether made by myself or 

 by others. Statements that the entire edge of the maxillary is denticulate and 

 that there are no teeth in the maxillary are especially to be doubted. 



Reverting to the modifications of the three types of teeth, unicuspid, tricuspid, 



