History of a transient nervous apparatus in certain Ichthyopsida. 409 



nations of embryological fact, and this forms a very serious obstacle 

 in the way of progress. 



Sooner or later a conviction may dawn upon investigators, that 

 the laws governing animal development lie deep, and that to discover 

 them we must also take wider and more comprehensive views of our 

 problems. 



Moreover, it should not be forgotten that little things, like a few 

 stray ganglion-cells, may hide an all-important story, and that here, 

 as elsewhere, in the apparent exception itself the key to the mystery 

 often lies concealed. 



Abstract. 



In this paper some of the results of investigations into the 

 transient nervous apparatus of certain Ichthyopsidian embryos (on 

 which the author has been engaged for several years) are described. 

 The researches are a continuation and extension of previous work on 

 similar structures in Lepidosteus osseus , of which a preliminary ac- 

 count^) was presented to the Royal Society of London in 1889. The 

 first part of the paper is devoted to the description of the develop- 

 ment, growth, and degeneration of a transient apparatus of ganglion- 

 cells and simple nerves, as demonstrated by the aid of a progressive 

 series of upwards of 120 skate embryos, whose sizes range from 5 mm 

 to upwards of 19 cm. 



Following on this, a general summary of the development of the 

 apparatus, of its probable functions and of its fate is given. At the close 

 of the paper the question of antithetic alternation of generations or 

 "development by substitution of organisms" is briefly discussed, and 

 an attempt is made to prove that such an alternation can be traced 

 in Vertebrate development, more especially in that of lower Verte- 

 brates like Raja. 



Development of the ganglion -cells of the apparatus. 

 — When sections of early embryos (5 mm) of Raja hatis are examined, 

 it is seen, more especially in the region of the pronephros and for 

 some distance caudad, that, at a period when the cells of the future 

 spinal ganglia are wandering out from the lips of the cord, there 

 are among them and on the cord itself certain cells in the act of 

 assuming ganglionic characters. This formation of ganglion - cells 

 begins immediately in front of the pronephric region, and, gradually 



1) Beaed, J., On the early development of Lepidosteus osseus 

 (preliminary notice), in: P. R. Soc. London, V. 46, p. 115 — 118. 



