No. 4-] AMPULLAE OF THE SELACHII. 175 



furnish no support for the suggestion made by Fritsch ('90) 

 that the centrum represents a sense organ which has lost its sen- 

 sory function because of the pressure of the overlying mucus. 



It therefore seems improbable that the ampullae serve as 

 organs for the reception of stimuli which result in sensations 

 of taste or audition, since the structures in vertebrates endowed 

 with such functions have a characteristic epithelium consisting 

 of supporting cells and sensory cells with sensory hairs. If the 

 ampullae prove to have sensory function their structure (free 

 nerve endings beneath an overlying epithelium) makes it more 

 probable that they are of the nature of tactile sense organs. 



Facts in support of the glandular theory are less easily 

 attainable. The mucus is present in large quantities, it is 

 true, filling ampullae and tubes; but I have never seen any 

 indication of glandular activity in the cells, nor have any 

 observations pointed to the fact that mucus is discharged from 

 the cells. The very large size of the nucleus leaves little room 

 in the cell for the processes of secrttion, and the position of 

 the nucleus at the peripheral end of the cell, where we should 

 expect the mucus if made to be aggregated and discharged, is 

 another fact opposed to the theory that we are dealing with a 

 typical glandular structure. 



An interesting correlation has been observed between the 

 habits and the number of ampullae found in three of the sela- 

 chians common at Woods Holl. As is well known, the shark 

 is very active in its movements, the skate is less active, while 

 the Torpedo is comparatively sluggish. The following table 

 gives the result of a careful counting of the ampullae in the 

 three forms above mentioned. 



Number of ampullae in dogfish (Galeus cam's} . 1595 



Number of ampullae in skate (Raja erinacea) . . 779 

 Number of ampullae in Torpedo (Torpedo occidentalism 220 



In the structure of the individual ampullae one is struck with 

 the complexity exhibited in Galeus, with the ten to twelve 

 ampullary pockets, as compared with the relative simplicity of 

 the six pockets of the Torpedo ampulla. In the skate the 

 number of pockets is usually seven or eight. 



