MUSTELUS L.EV1S. 91 



fold of the epidermis, the so-called ' pocket/ which covers the 

 growing end of the line." Whether or not this '^ pocket " re- 

 pi'esents the first beginnings of the lateral canal, it is evident 

 that a canal that thus first appeared at the hind end of the 

 lateral sensory line must, appearing as it does before the line 

 reaches the hind end of the body, be continually being pushed 

 bodily backward, followiug and accompanying the growing 

 end of the line. There is thus here such an important depar- 

 ture from what is found in Amia, that I should hesitate to 

 accept the account as correct. 



My own observations, limited to the four series of sections 

 that I possess, tend to confirm Balfour's statement that the 

 lateral canals are formed by an involution of the deeper layer 

 only of the ectoderm. The covered gutter, rather than canal, 

 that is thus first formed then becomes, by a pressing together 

 of its walls, a sharp and apparently solid ridge projecting 

 inward from the inner surface of the ectoderm. This is the 

 condition found nearly everywhere in my 36 ram. embryo. 

 Where the cord is deepest it receives a branch from the 

 underlying and related lateral nerve, but there is as yet no 

 perceptible indication in my sections of a definite sensory 

 organ related to this nerve. Between each two consecutive 

 points where these nerves thus join the cord the cord becomes 

 less deep, and in the regions where it is the flattest there is a 

 small pit-like depression on the outer surface. This depres- 

 sion looks in certain sections like a shallow groove, while in 

 others it has vertical sides, and seems to cut clean through 

 the outer layer of ectoderm down to that deeper layer that is 

 alone directly related to the caiial itself. This depression, or 

 pit, quite certainly represents a future pore, the pore thus 

 apparently appearing while the cord that represents the canal 

 is still everywhere attached to the ectoderm. 



In my 55 mm. specimen the lateral canals were mostly 

 found as cords of tissue lying in the mesoderm, beneath the 

 ectoderm, and connected with the latter at intervals by 

 smaller cords, which represent the tubules of the adult. In 

 the large main cords there was a small central lumen, either 



