Herrick, Criteria of Homology. 207 



in primitive vertebrates there is in front of tlie trigeminus a gen- 

 eral cutaneous nerve belonging to a different segment, the profundus 

 nerve. The profundus nerve is rarely preserved in the adult, though 

 vestiges of it can be recognized in several selachians and ganoids, 

 Avliere there is evidence that the profundus nerve has fused witli 

 the ophthalmic rami of the trigeminus. These rami, therefore, are 

 to be regarded as trigeminus jjlus j)rofundus nerves in all cases 

 where it can be shown that the profundus elements are preserved. 

 Similarly, in fishes branches of the lateral line and gustatory roots 

 of the facialis often anastomose peripherally with trigeminal 

 branches. It is evident that the mixed ramus thus constituted has 

 no longer the same individuality as before. It cannot be classed 

 simply with the trigeminus or facialis ; it is both. It should be 

 given both names, or an entirely new name, or else some arbitrary 

 rule should be laid down regarding the selection of a single name 

 already current. The past usage in such cases has been most varied 

 and confusing, and the confusion has, in many cases, been worse 

 confounded by ignorance of the fact that there was any difference 

 in the composition of the anastomosing rami, or l)y indifference 

 to this fact even when recognized. The result is that to-day the 

 synonymy of the rami of the cranial nerves of lower vertebrates, 

 where such anastomoses are frequently and very diversely devel- 

 oped, is in worse confusion than that of the pre-Linnsean herbals. 



(8) A given ramus of a seginental nerve which contains more 

 than one component is not perfectly homologous with a ramus of 

 the same nerve in a different species which has a similar peripheral 

 distribution, but lacks one or more of the components or has an 

 additional component, even though the added component comes from 

 the same segmental nerve. Thus, the hyomandibular trunk of the 

 cod is not perfectly homologous with this neiwe in Menidia ; for 

 the cod lacks the visceral sensory component of this nerve which 

 is present in Menidia, though the other relations are practically 

 identical. 



(9) From the preceding considerations it follows that the com- 

 position of every nerve and rannis must be accuratxsly known before 

 its homologies can be understood. Dissimilar and unrelated func- 



