History of a transient nervous apparatus in certain Ichthyopsida. 385 



There is, however, another fact, in addition to those previously men- 

 tioned, which tends to indicate the relations of some of these transient 

 nerves to the yolk-sac. It is that in such embryos as Nos. 410, 141, 

 419 etc. these nerves are biggest and longest in that region where 

 the epiblast and somatopleure are continued upon the yolk-sac. This 

 fact surely has as part of its significance the passage of such nerves 

 upon the latter. 



And, to put the same thing into rather different words, the nerves 

 cease, as described on page 348, where the embryo leaves the yolk-sac. 



e) Build and comparison of the sub-epiblastic ner- 

 ves of the transient system. The nerves appear to be quite 

 destitute of any medullary sheath or so-called "white substance of 

 Schwann". For a long time they are all merely naked axis-cylinders, 

 and are comparatively seldom composed of more than one fibre. In 

 later stages "nuclei" occur at intervals on them. Most of these nerves 

 are undoubtedly merely processes of single cells — most but not all. 

 In studying the immense series of Eaja - sections and in drawing 

 the figures endeavours have been made to depict only things about 

 which there could be no manner of doubt. Every section figured was 

 studied many times, often with the best lens available, and whenever 

 there appeared to be the slightest dubiety about any critical point, 

 no call was made on the imagination : the pictures were drawn just 

 as they appeared to the unbiassed eye. 



Of later stages there has been some hesitation about publishing 

 combination drawings of sections; for, owing to the increased com- 

 plexity of the nerves, it was difficult, or impossible, to be absolutely 

 certain that what one appeared to see had an actual existence in fact 

 in the sections ^). But in earlier embryos everything is quite clear, 

 and, as far as the nerves can be followed, their composition can be 

 determined with considerable certainty. 



There is one drawing (Fig. 24, plate 23) which did cause some 

 anxiety, but not on account of any doubt as to the correctness of 

 what was depicted. The drawing was unwittingly made twice, and on 

 the second occasion, when the best lens (Leitz 2 mm Pantachromatic) 

 was used as a check against mistake, the only point of difference 

 from the previous sketch lay in the nature of the ganglionic break in 

 the nerve on the left of the figure. On the first occasion only one 



[ 



1) One such inaccurate combination drawing is fig. 8 of the 

 preliminary paper (No. 3). 



