584 PROCEEDINGS OF THE AMERICAN ACADEMY. 



of the inner mouth plexus lose their slender, thread-like appearance 

 as they approach the plexus, and become flattened and band-like 

 (PI. 2, Fig. 7, ?'; PL 4, Fig. 18, i.). These branches lie interior to those 

 (o.) forming the outer mouth plexus, the difference in focus increasing 

 in their distal portions. The branches forming the inner plexus also 

 lose their fibrous appearance in methylene-blue preparations as they 

 approach the plexus, and appear granular; this change may be noted 

 in all branches of dorsal nerves innervating deep-lying structures. 

 At the place of main di\'ision of the ventral ramus of left nerve IV a 

 very large branch is also given off, which bends at once beneath the 

 ventral border of the trunk muscles, and passes to the right side of 

 the oral hood. Here it may emerge at varying places, usuallj^ how- 

 ever, near the ventral border of the trunk muscles of the right side, 

 and often anterior to the ventral branches of right nerve IV (Fig. 18). 

 This branch of left nerve IV subdivides in various ways in the right 

 wall of the oral hood, and the branches thus formed break up into 

 that portion of the inner mouth plexus of the right side which lies 

 posterior to the part formed by the branch of left nerve III, previ- 

 ously described. No other nerve branches from nerves on either side 

 of the body appear to be concerned in the formation of the inner plexus 

 of the mouth border on the right side, in either species. 



Another large, more posterior branch of left nerve IV leaves the 

 ventral ramus at its main place of division. This branch bends 

 beneath the trunk muscles toward the interior, and after reaching the 

 median region ventral to the notochord, runs posteriad as a thick, 

 straight trunk. In methylene-blue preparations in which epithelial 

 structures are not retained, this nerve branch appears to end abruptly 

 a short distance in front of the velum, and slightly to the right of the 

 median line (Fig. 8, IVa). This nerve may correspond to the 

 branch of nerve V described by Hatschek ('92, p. 144) as innervating 

 the sensory groove of Hatschek. Much variation certainly occurs in 

 the distribution of the branches of the dorsal nerves in the buccal 

 region. In this connection it may be noted that Heymans et van der 

 Stricht ('98, PI. VI, Fig. 22) figure a branch of left nerve IV passing 

 directly to the right side of the velum. I, however, have been unable 

 to find any connection between this nerve branch and the nerves of 

 the velum ; but, in some instances another large branch of the ventral 

 ramus of left nerve IV clearly did join in the innervation of the velum. 

 This branch is usually given off from the ventral ramus of the latter 

 nerve at or near the main place of di\'ision, and most frequently enters 

 the velum on the left side of the body. In one specimen (B. lanceo- 



