316 
15ULLKTIN OF THE UNITED STATES FISH COMMISSION. 
jMenidia is a favorable type for such study, not only on account of its central 
position and small size, but the tissues seem to react histologically better than those 
of some other Ashes. Moreover the nervous system of the family Atherinidai has not 
before been examined even by gross dissection, and several points have been brought 
out in the topography which are either new for teleosts or shed light on vexed morpho- 
logical questions, such as the discovery of a pre-trematic branch of the facial nerve, a 
ramus ophthalmicus profundus of the trigeminus, and a true spinal accessory nerve 
emerging with the vagus to supply the trapezius muscle. 
The research was conducted by means of reconstructions from serial sections cut 
through the entire head of the adult iisli and stained by a modification of the Weigert 
method,* the aim being to trace each nerve component continuously from its nucleus of 
origin or termination in the brain through the root and ganglionic complex to its 
peripheral end. Thus, ultimately, the exact composition of each peripheral nerve and 
ganglionic complex would be given. This attempt has been crowned with a fair 
measure of success, and plots have been prepared to exhibit graphically the courses 
of the several kinds of fibers. 
The doctrine of nerve components dates properly from the systematic separation 
of sensory and motor roots and the formulation of Bell’s law. (laskell’s suggestive 
“four-root theory” of the nerves has been a stimulus to further advance, though prob- 
ably that theory will not stand in exactly its original form. Our precise knowledge 
of the sensory components in the cranial nerves of lower vertebrates begins with 
Strong’s pai»er on “The Cranial Nerves of the Amphibia” in 18h5, and the present 
investigation was carried out upon the basis of that work. The cranial nerves of the 
fish which 1 have studied have reduced themselves to a plan which is so simple and 
so similar to what Strong found in the Amphibia that we are justified, 1 thiidi, in 
regarding this arrangement in its main outlines as a type to which the nerves of all 
the higher fishes may provisionally be referred (see diagram). A careful study of 
nearly all of the existing literature supports this belief. At any rate this scheme 
will serve, it is lioped, as a basis for future work in the attempt to unify and correlate 
our knowledge of the nerves of the lower vertebrates. 
The spinal nerves . — Following Gaskell. four components are now generally recog- 
nized in the spinal nerves of vertebrates: (1) somatic motor from the ventral horn 
cells, supplying the striated body musculature; (2) somatic sensory (general cutaneous), 
terminating in the dorsal horn and supplying the skin of the body; (3) visceral motor; 
(4) visceral sensory. The last two components are supposed to be related to the 
“intermediate zone,” or lateral horn region of the spinal cord, the sensory fibers 
coming in by the dorsal roots and the motor fibers going out by both roots. 
Ill Menidia, the r. medius (r. lateralis of authors) of the several spinal nerves 
usually anastomoses with a twig of the r. lateralis vagi. The two nerves can, 
however, be distinguished after the anastomosis by the difference in the size of the 
fibers, and it is seen that the spinal nerve never participates in the innervation of the 
lateral line organs, but always goes to the skin adjacent. The anastomosis has no 
* The technique employed hero and in numerous other experiments with the Weigert processes 
carried on in connectiou with this work has been fully reported in a previous paper — “ Report upon 
a Series of Experiments with the Weigert Methods,” State Hospitals llulletin, vol. ii, Nos. 3 and <1, 
Utica, N. Y., 1897; also a full abstract of the above in .Jour. Comiiarative Neurology, vol. viii. Nos. I 
and 2, .Inly, 1898. 
