SENSE ORGANS OP AMIURUS. 267 



in the pike, but perhaps more closely resemble the structures which, 

 in Lota, communicate with the outside by scattered pores. The 

 structures to which I refer open by slit-like apertures very different 

 in character from the ordinary pores. It is only in the fresh skin that 

 they can be readily detected, and then it is owing to the deficiency of 

 pigment in the wall of the slit similar to that which occurs in the 

 mouth of the pore, that they stand off from the rest of the skin. In 

 size they vary considerably. Some are larger, others much shorter 

 than the pores, but all of them are very much narrower. The 

 most easily recognized are those which form a sort of accessory lateral 

 line stretching obliquely downwards and backwards from the upper 

 angle of the gill-slit. They are accompanied and probably supplied 

 by a distinct branch of the Ramus lateralis vagi, which runs along the 

 line of junction of the lateral and ventral musculature, but another 

 very distinct row is to be found almost parallel to the praeopercular 

 mucous canal, running down over the M. adductor mandibular. Both 

 of these are indicated by the dotted lines on Fig. 6. Again, in front 

 of the dorsal fin similar slits occur, several very distinct behind the 

 occipital pores, others less so, disposed transversely to the long axis 

 of the body. 



I have no preparations of the adult skin which pass through these 

 structures, but in a series through a young fish of two inches in 

 length, made for a different purpose, I find certain detached flask- 

 like sacs traceable through three or four sections, which communi- 

 cate freely with the outside by apertures which are, no doubt, the 

 above-mentioned slits. These sacs appear to be irregularly scattered, 

 at any rate, as Leydig observes in relation to the pike it would be a 

 work of some labour to map them out, but although often far re- 

 moved in the trunk from the lateral canals, they appear to be always 

 grouped near these in the head. They are especially numerous in 

 the neighbourhood of the nerve-hillocks, and are thus found especi- 

 ally on the snout, below the eyes, on the cheeks and in the occipital 

 region. I recognize the same structures also in the much younger 

 forms whence Fig. 7 is taken, and as well in the one series as in the 

 other, the difference between these sacs and the end-buds is very 

 striking. Although the central-cells of the end-bud may be retracted, 

 as noted above, so as to form a little recess in the mouth of the 

 ' beaker,' the whole organ does not extend down to the corium but 

 is lodged on a papilla extending half-way up through the epidermis, 

 20 



