2o8 'Journal of Comparative Neurology and Psychology. 



tional systems must never be homologized either in a phylogenetic 

 or nijeristic series. 



(10) When the composition of a segmental nerve is fully known 

 each root and its ganglion (in case of the sensory roots) should 

 have a separate name and be treated as a functional and morpho- 

 logical unit. In determining the homologies of such a unit regard 

 must be had primarily for the function which it performs, as deter- 

 mined by its terminal relations, i. e., the type of peripheral end 

 organ and the location within the central nervous system of the 

 primary nuclei of origin or termination of its fibers. These con- 

 siderations take precedence over all others in doubtful cases. 



(11) The peripheral relations of nerves to other organs along 

 their courses are also important in determining their homologies ; 

 but resemblances in such relations must not outweigh differences 

 in functional composition, where these two factors are in conflict. 



(12) No inflexible rules can at present be laid do^vn for the 

 nomenclature of peripheral rami of mixed or variable composi- 

 tion. In the selection of names for peripheral nerves or rami pri- 

 ority should rule among competing terms, other things being equal. 

 But if the prior term implies a false morphology, or is unnecessarily 

 cumbersome or otherwise objectionable, or if it has been long obsolete, 

 it may be discarded in favor of a better one or one better known. 

 Peripheral rami of mixed composition can often be analyzed into 

 an ancient or primary branch of one nerve and cenogenetic addi- 

 tions by peripheral anastomosis from a different segmental nerve. 

 In such cases the ramus may be named as a branch of the nerve with 

 which its palingenetic connection is made, even though it is not exactly 

 homologous with the nerve so named in other species which lack the 

 peripheral anastomotic addition. Thus the r. mandibularis trigemini 

 is typically composed of general cutaneous and motor fibers. In 

 some vertebrates gustatory fibers enter it by peripheral anastomosis 

 from the geniculate ganglion of the facialis. The mixed nerve so 

 formed may for convenience still be termed the ramus mandibularis 

 trigemini, even though it is morphologically partly facialis, provided 

 the imperfection of the homology is explicitly recognized. In the 

 same way the lingual nerve of man may be assigned to the trigeminus. 



