2 J. GRAHAM KERR. 



labour to making a very detailed investigation of the stages 

 in development here treated of, and I had originally intended 

 in my description to go into something like corresponding 

 detail. I have, however, altered my original intention in 

 this respect, for various reasons : amongst others because 

 in the interpretation of minute details of early development 

 one is necessarily much influenced by preconceived ideas; 

 and in the second place, because I find that these details 

 vary to an extraordinary extent in different eggs — some of 

 the variations being apparently due to variation in technical 

 methods of investigation, but many being certainly true in- 

 dividual variations. So potent are these disturbing factors 

 that I doubt very much whether a description going into very 

 minute detail must not necessarily be to a great extent mis- 

 leading, and so do harm. I therefore propose to limit myself 

 in regard to the early stages of development to the endeavour 

 to give an adequately complete general description of the 

 phenomena observed, with only so much detail as may seem 

 necessary to make the description clear. 



The investigation of a holoblastic egg 7 mm. in diameter 

 and packed with yolk involves great technical difficulties, as 

 the whole of each egg has to be converted into thin sections. 

 The full extent of these difficulties will only be appreciated 

 by embryologists who have essayed a similar task. In order 

 to help future workers I devote a few paragraphs to a general 

 account of methods. Then follows an account of the 

 phenomena observed, in which, as in my first paper, I 

 reserve reinarks of a general nature embodying views rather 

 than facts for a concluding section, so that any reader may 

 obtain the facts, which are naturally of greater importance, 

 with a minimum of trouble. 



299. As in Unit paper 1 naturally did nol make precise statements regarding 

 the interpretation to be put upon surface features without having assured myself 

 first by the examination of sections that they were correct, it is unnecessary 

 for Prof. Semon to feel the doubts about the behaviour of the blastopore in 

 Lepidosiren vrhich he expresses in his latest contribution on the de- 

 velopment of Ceratodus (Semon, ' Zoologische Forschungsreisen,' Band i, 

 S. 327). 



