THE PRIMARY SEGMENTATION. 133 



These segments have also been studied in living embryos of 

 several amphibia in stages with an open neural groove. They 

 have also been identified in the early stages of the newt, the 

 frog, and the torpedo. 



The neural segments have now been shown to occur in the 

 very early stages of a number of animals. The fact of their 

 presence in these early stages once established, they assume 

 new importance, They have a too definite history to admit of 

 being set aside as mere headings or undulations of no meta- 

 meric significance. The fact has to be confronted that these 

 neural segments are the early stages of the neuromeres, whose 

 characteristics have been determined by a number of observers. 

 So long as the neuromeres are supposed to be moulded over 

 the mesoblastic somites, they can have no particular importance 

 in the problem of head segmentation, but that view of the 

 neuromeres is an assumption made without a knowledge of 

 their early history. This early history, now made known, 

 places them in a new light, and, taken all together, the neural 

 segments furnish, I think, a more satisfactory basis for inter- 

 pretation of metamerism of the head than we have had before. 



There is one question regarding the validity of these seg- 

 ments that must be disposed of before we can proceed further. 

 All investigators know that appearances simulating regular 

 structures may be produced by the reagents used in prepar- 

 ing material for study ; and the question comes to us, Are not 

 these segments artifacts produced by the action of chemicals ? 

 Too great precaution cannot be taken in sifting this matter to 

 the bottom. The large number of specimens studied as a basis 

 for the facts already given were prepared by a variety of 

 methods. They were treated with reagents well known to 

 morphologists, such as picro-sulphuric acid, picro-nitric, Flem 

 ming's solution, Davidoff s corrosive-acetic, chromic acid with 

 a trace of osmic, corrosive sublimate removed with iodine ; and 

 in all cases the segments have been distinguishable, not in 

 patches but in such condition as to admit of being counted, 

 and there has been uniformly the same number of segments 

 in the head-region. It is not reasonable to assume that the 

 different reagents would all produce the same effect. 



