962 MRS. O. A. MERRITT HAWKES ON THE [Dec. 11, 



course, for a distance of 5 cm. This portion of the canal, which 

 is closed, is probably the original hyomandibular canal, to which 

 the other parts have sul^sequently been added. In one specimen 

 there were here 15 neuromasts. Tubviles are present in the first 

 part of the canal only. Somewhat caudad to the mouth a branch 

 (G, H, I) is given off, which, curving round the mouth, runs 

 along the lower jaw to the symphysis (Plate LXVIII. fig. 1, 

 H, L, A), This branch is an open canal, and in one case had 

 19 neuromasts. On the left side of one specimen 1 cm. of 

 this canal was closed. It may be mentioned that in this closed 

 region, although there wei'e four neuromasts, there were no ex- 

 ternal openings corresponding to them. A second open canal 

 (J, L, K, H) leaves the main hyomandibular canal at J. This 

 branch (H, L, B) makes a wide curve downward and then forward 

 to meet the last-mentioned canal at H, shortly before its termi- 

 nation. One side of one specimen had 42 neuromasts in this 

 branch. In six out of the seven specimens examined externally, 

 the hyomandibular canals of the two sides appeared to join at I. 

 In the seventh specimen, however, the regions H, I, of the two 

 sides were parallel and unconnected. Doubtless this was the 

 original condition. The branch H, L, B, on one side of a single 

 specimen, had a break of 1 cm. At first sight this break might 

 seem to indicate that this was the point where the two 

 branches G, H, K and J, L, K united, but when we see the 

 canals in relation to their nerve supply, it is obvious that this 

 cannot be the case. Undoubtedly the two canals met and united 

 at H. There is a dorsal branch canal (L, M), narrow, shallow, 

 and superficial, which curves upward and forward towards the 

 lateral canal. In one specimen 10 neuromasts were present. 

 One of the specimens had, on both sides, another small hyoman- 

 dibular branch (N) and a cuiious partial twist in the region J , L, 

 which terminated in a fine branch (O). Branches N and O are 

 no longer functional. There is only a very slight depression in 

 the skin, bordered by the usual triangular scales which functionally 

 close the " open " canals. In the depression there is neither 

 sensory tube nor neuromast. JSTo nerve could be traced to these 

 canals. It maybe, however, that microscopical work would discover 

 what macroscopical methods have failed to disclose. If these, 

 probably degenerate canals occurred in a large number of speci- 

 mens they would indicate that the canal-system of Chlamydo- 

 selachus in this region had become simplified by the partial atrophy 

 of certain of its branches. As, however, these canals were found 

 in only one specimen out of seven, they must, for the present, 

 probably be regarded as an individual variation. 



The Gommissurcd or Supra-temjyoral canal (or canals), the aural 

 canal of Garman, is placed, as noticed by Garman, anterior to the 

 openings of the ductus endolymphaticus. It is never the usual 

 straight, transverse canal connecting the right and left lateral 

 canals. It varies considerably, as shown in text-fig. 140. There 

 are indications of two instead of one commissural canal, but it is 



