MECHANORECEPTORS AND BEHAVIOR 347 



(1678) subsequently distinguished between the canal system and the gelati- 

 nous tubules bearing the ampullae now named after him, but the whole 

 system was regarded as mucus-secreting until in the late nineteenth century 

 Leydig and Schultze put forward the idea that it was really a sensory struc- 

 ture, perhaps serving a "sixth sense." 



The pattern of the canals— Although the distribution of the canals is 

 discernible in some species on superficial examination, for most the canals 

 have to be exposed in some way, either by injecting dyes or by serial section- 

 ing of embryonic material. 



The nomenclature of the various canals has caused much confusion and 

 was subject to considerable variation until Ewart (1892) appreciated that the 

 distribution of the canals is strongly influenced by their innervation. The 

 scheme which he then elaborated has, for the most part, been followed by 

 subsequent writers. Accordingly, the canals constitute the follow- 

 ing (Figure 7A); 



1) the lateral canal, innervated from the lateralis division of the vagus 

 nerve, and running from the tip of the tail to the head, where just behind the 

 orbit, it divides into: 



2) the supraorbital canal, innervated by superficial ophthalmic VII, which 

 passes over the eye and 



3) the infraorbital canal, innervated by buccal VII, which runs under the 

 eye, off which branches 



4) the hyomandibular canal, innervated by external mandibular VII. A 

 short canal, the supratemporal, crosses the head to unite the two lateral 

 canals, while discrete mandibular canals lie on each side of the lower jaw. 



In some way the configuration of these canals presumably reflects hydro- 

 dynamic flow over the head and body, but its biological significance remains 

 a mystery. 



Apart from Ewart's original description of Somniosus, full accounts of the 

 form of the canal system in a great variety of elasmobranchs are provided by 

 Garman (1888), Ewart and Mitchell (1892), Johnson (1917), Norris (1932), 

 and Von Bonde (1933a, b). 



Some canals of Chlamydoselachus (Hawkes 1906) and of Chimaera (Cole 

 1896) consist merely of a groove set into the surface of the skin. In other 

 forms these grooves are closed for at least part of their length, while in most 

 elasmobranchs these tubes then lie beneath the epidermis and connect to the 

 outside by means of tubules, which Ewart thought of as 'feelers'. In 

 Chlamydoselachus , however, some canals lack tubules and open directly. The 

 tubules are mostly quite short, but in some species they may be very long and 

 branching, as on the tail of the thresher, Alopias. In front of the eye the 

 tubules generally run anteriorly, whereas behind the eye the pores usually 

 face the tail. 



Although the canal system of most species conforms to the general 

 pattern just described, variations are found (Figure 7). For example the 



