on the fifth line of scales dorsal to the lateral line, counting dorso- 
anteriorly in the same rows of scales as heterofore, that is, in the 
rows that begin dorso-anteriorly and from there run ventro-posteriorly. 
At right angles, approximately, to these well marked rows, the scales 
are also arranged in less well defined rows that begin ventro-anteriorly 
and run dorso-posteriorly. The last four sensory patches of the dorsal 
body line lie in four consecutive scales of one of these last mentioned 
rows, and their appearance strongly suggests the possibility of their 
lying in the middle line of the embryonic tail. The sensory patches 
of this body line are not found on every row of scales, certain rows 
being skipped. 51 sensory patches, in all, were found on the left side 
of the specimen. 
The second, and irregular body line of sensory patches extends 
from near the head to the level of the most posterior dorsal finlet. 
The first patch of this line, on the left side of the 44-cm specimen, 
lay on the sixth scale of the sixth row posterior to the posttemporal 
bone, and was directly in the line produced of the first two sensory 
patches of the lateral line. The other patches of the line lay at this 
same level, or on scales occupying a level one scale ventral to it, there 
being, in all, sixteen patches in the full length of the line. Whether 
this sensory line, or the main lateral line, is the strict homologue of 
the main lateral line of Amia, can only be determined when the inner- 
vation and development are fully known. ‘The fact that the middle 
and irregular line lies very closely in the direct continuation backward of 
the first few sensory scales of the lateral system, while the main lateral 
line drops, irregularly, to a considerably lower level certainly needs 
some explanation. It’s possible that we have here the homologues 
of the main lateral, and accessory lateral lines of Amia. In that case 
the constantly varying relations of the main line to the scales of the 
fish, in the anterior portion of its length, would seem to throw some 
doubt on the value of the organs of the lateral sensory system in 
determining the exact homologues of the bones that are developed in 
relation to them. 
There is no canal formed in any part of either of these three body 
lines of Polypterus, and no canal whatever could be found in the fish 
posterior to pore 14 infraorbital, or posterior to the supratemporal 
commissure dorsal to the line of the main infraorbital canal. The 
complete canal said by Pornnarp (No. 12, p. 545) to correspond, in 
Polypterus, “to the rudimentary canal represented in Amia by the 
dorsal line of pit organs“, did not, accordingly, exist in my specimens. 
The Innervation of the sense organs of the canals of Polyp- 
