396 



its segmental importance. In reaching this point, first came the work 

 of Beraneck '84, and Orr '87, showing that the neuromeres of the 

 hind-brain are definitely connected with nerves ; following this, it was 

 demonstrated by Kupffer '86, and Mc Clure '89, that the neuromeres 

 extended throughout the neural tube, and that those of the trunk 

 region merge gradually into those of the head region. Lastly, the 

 neuromeres of the mid-brain have been especially studied by Waters 

 92: the condition of that brain region having been left undetermined 

 by previous observers. 



Among the problems that have arisen in studying the head 

 region are those that deal with the nature of the primitive segments 

 that have been modified to form the head, tracing the transformations 

 of these segments, and determining the number that enters into its 

 composition, and their arrangement with reference to the fore-, mid-, 

 and hind-brain respectively. These questions have attracted much 

 attention, and the fact that attempts to solve them have been made by 

 using as criteria myotomes, branchial sense organs, gill-arches and 

 neuromeres , without arriving at any final satisfactory conclusion, all 

 goes to show the great difficulty of the problems. It goes also to show, 

 that our knowledge of the segmentation of the head is still in a for- 

 mative state, and so long as that is the case, any actually observed 

 facts regarding that segmentation will be of value in helping clear 

 the ground for the ultimate solution of the problems of metamerism 

 of the head. It is this consideration that leads me to offer in 

 the form of a preliminary communication some new facts that I 

 have worked out on one of the Elasmobranchs — viz.: Squalus 

 acanthias (Synon. Acanthias vulgaris). I have been able to determine 

 in that form, that the primitive metamerism is exhibited in very 

 much earlier stages than has yet been recorded for other animals. 

 It first makes its appearance in the rudiments of the nervous system, 

 and this fact has an important bearing upon its interpretation: — 

 it is epiblastic, not mesoblastic. The division into segments is very 

 distinct long before the closure of the neural groove, and it not only 

 extends the whole length of the embryo, but, in Squalus acanthias, 

 extends also some distance into the embryonic rim. I have traced, 

 coherently, the history of these segments, in this Elasmobranch, from 

 their earliest appearance up to the time they form " neuromeres " and 

 begin to degenerate. I shall show that the relation of the parts of 

 the brain to the segments, is not the same in early conditions that 

 it is in later ones and, therefore, the interpretation of Waters that 



