408 Edward Phelps Allis jr., 



In the larva, as the reconstruction from serial sections shows, 

 the nasal section of the supraorbital canal begins at a pore that lies 

 between the anterior and posterior nasal apertures. From there it 

 runs forward, mesial to the anterior aperture, and beyond that aperture 

 turns mesially and then backward in a short curve. It then continues 

 directly backward to the level of the hind margin of the posterior 

 nasal aperture, where it opens on the outer surface in an elongated 

 depression common to it and the anterior pore of the frontal section 

 of the canal. In this nasal section of the canal there were, in the 

 specimen examined by sections, six sense organs on one side of the 

 head and seven on the other, and a primary tube, opening by a single 

 surface pore, arose from the canal between each two successive organs. 

 In the frontal section of the canal there were, on the one side only 

 that was examined, twelve sense organs; and certain of the primary 

 pores had here already undergone a first dichotomous division, as shown 

 also in the figure, this division being, in certain cases, so deep that the 

 primarily simple primary tubes were there each split into two separate 

 portions which arose separately and independently from each other, but 

 close together, from the main canal. Between the 10th. and 11th., and 

 the 11th. and 12th. organs of this frontal section of the line no primary 

 tubes are shown in the reconstruction, this indicating, if correct, that 

 the two tubes that should normally have here been formed have either 

 aborted, even at this eaily age, or have, for some reason, never been 

 formed. In the larva from which the sketches for fig. 1 were made 

 there was apparently also, in this same region, a considerable length 

 of canal without related pores. What the explanation of this may 

 be I can not suggest, but, as the larvae were both reared in an 

 aquarium, the absence of the two or more tubes here is perhaps an 

 accidental irregularity due to this artificial life, for the trunks of 

 what are apparently corresponding dendritic systems are here found 

 in the adult. 



Anteriorly the frontal portion of the supraorbital canal was, as 

 just above stated, incompletely continuous with the nasal section. 

 Posteriorly, it had not as yet anastomosed with the main infraorbital 

 canal; and from the position and direction of its hind end it is probable 



