672 MR FRANK J. COLE ON THE 



though possibly not in the sense intended by Shore ; and I have already taken 

 advantage of this distinction when discussing the homology of the chorda tympani. 



L. The Anterior Spinal Nerves (fig. 2). 



Like Ltemargus, Hexanchus, and most Elasmobranchs, the anterior spinal nerves of 

 Chimsera, which as usual have no dorsal roots, pass through the cranium. How many, 

 however, is not so easily distinguishable as in other Elasmobranchs ; but as numerous 

 rootlets combine to form two main nerves (emerging, however, from the skull by three 

 foramina), at least two may be described. These two spinal nerves unite to form the 

 brachial nerve, and I propose to call them the cranial spinal nerves. 



First cranial spinal (I). — Arises by seven rootlets, which unite to form two roots; the 

 anterior root emerges from the medulla under cover of the posterior roots of the vagus, 

 and has four rootlets. The roots of the intestinal division of the vagus seem to 

 anastomose with this root, but really pass underneath it. The posterior root (three 

 rootlets) passes through a separate canal and unites with the anterior root just outside 

 the cranium. At the point of union a single dorsal branch (see Sp. d) is sent up to the 

 skin. The two roots having united to form the first cranial spinal, the latter dips down 

 suddenly at right angles towards the ventral surface. 



The posterior root of this cranial spinal has a more ventral origin than the anterior, 

 and it may have more than three rootlets. 



The first cranial spinal, therefore, has no dorsal root and one dorsal branch. 



Second cranial spinal (II). — Arises by six rootlets, which go to form two roots. 

 These two roots remain to some extent distinct, but there is an exchange of fibres 

 immediately outside the cranium. At this point two dorsal branches are given off to 

 the skin, and one of these soon divides into two (see Sp. d). This fact does not neces- 

 sarily indicate that this cranial spinal is a compound of three, since a single spinal nerve 

 of the Skate may have as many as four dorsal branches (see Ewart and Cole, 66). 



The anterior root of this spinal has at least two rootlets (and may have three), 

 whilst the posterior root has at least four. These two roots, after mingling outside the 

 cranium, continue as two nerves, which dip down sharply at right angles towards the 

 ventral surface. The anterior nerve anastomoses with the main trunk of the first 

 cranial spinal, forming a conspicuous nerve, whilst the posterior joins on to the product 

 of the above fusion some little distance further down, to form the large brachial nerve 

 (Sp. III). 



The second cranial spinal, therefore, has no dorsal root and three dorsal branches. 



The brachial nerve is distributed to the pectoral fin (one branch running in a canal 

 (Sp>. F.) bored in the pectoral girdle), and I have seen it sending a branch to the muscles 

 of the last or fifth branchial arch. This is the only nerve I have traced to this arch. 



My dissections of the remaining spinal nerves are not sufficiently complete to be 

 described here. 



