580 PROCEEDINGS OF THE AMERICAN ACADEMY. 



Certain earlier authors, as Schneider ('79) and Rohon ('82), re- 

 garded the third pair of dorsal nerves as the first true sensory spinal 

 nerves, this interpretation being based partly on the fact that the first 

 pair of motor roots is associated with this pair of sensory roots. These 

 authors enumerated the "spinal nerves" separately, designating the 

 third sensory pair as the first spinal sensory roots. At present it 

 seems safest to enumerate this pair of dorsal nerves as the third, 

 without attempting to distinguish between so-called cranial and spinal 

 nerves. 



Heymans et van der Stricht ('98) have given a careful account of 

 the distribution of the branches of this pair of nerves in Branchio- 

 stoma lanceolatum, and Dogiel ( : 02) discusses nerve " III " in the 

 same species at considerable length. As has been noted, however, 

 Dogiel ascribes no definite territory of distribution to the third pair 

 of nerves, its number depending upon the number of roots present 

 near the place of exit of nerve III from the neural tube. Tliis un- 

 certainty makes comparison difficult and much involved. It may be 

 stated, however, that in his discussion of the innervation of the border 

 of the mouth Dogiel designates the commonly accepted third nerve as 

 III, as is shown in his Figur 1. 



I have studied this nerve in both species, chiefly upon specimens 

 impregnated w4th methylene blue, although gold chloride and picro- 

 carmine bring out excellently many details in the nerves of the oral 

 hood. The finer branches of the nerves of the buccal region are us- 

 ually impregnated in from 15 to 30 minutes in the methylene-blue 

 mixture previously described, while the main trunks of these nerves 

 require a longer immersion. A rather long subsequent exposure to 

 air (20 to 30 minutes) gave peculiarly brilliant results for the finer 

 branches. Fixation with ammonium molybdate and osmic acid, 

 after Dogiel's ( : 02) method, is well adapted to this region. The thin 

 tissues fix rapidly, and may be dehydrated quickly without loss of 

 impregnation. It is advisable before fixation to detach the anterior 

 portion of the animal, cutting a little behind the velum. By dissecting 

 along the mid-dorsal line, a flat, fairly thin preparation may be ob- 

 tained for mounting in balsam. Thick pieces of tissue fixed in am- 

 monium molybdate are usually unsatisfactory for sections or other 

 methods of study. 



In the two species studied surprisingly little difference was found in 

 the distribution of the branches of the tliird pair of nerves. Varia- 

 tions occur as frequently in Branchiostoma caribaeum as in B. lanceo- 

 latum, and are similar in character. In both species nerve III usually 



