376 DR ARTHUR T. MASTERMAN ON THE 



another was placed in a mixture of alcohol, acetic acid, and formalin, in the proportions 



of 



70 per cent, alcohol, . . .90 parts. 



Glacial acetic, . . . 3 „ 



Formalin, . . . 7 ,, 



This mixture, recommended to me originally by my friend Mr E. J. Bles of 

 Cambridge, I have found invaluable for a variety of histological purposes. For marine 

 collecting it is very handy, and specimens may be kept in it till required for sectioning. 

 (Another series was preserved with osmic acid, but these were not successful.) Both 

 series were excellently preserved, and gave equally good results. Sections were cut 

 in paraffin and celloidin, but as the paraffin alone appeared to be perfectly satisfactory, 

 I finally abandoned the use of celloidin as being less expeditious. In all pelagic 

 echinoderms celloidin appears to be essential, but these larvae certainly did not present 

 the least difficulty in manipulation. 



A large amount of labour and time were expended in searching for transition stages, 

 as it was felt that unless an absolutely complete series were forthcoming, the work 

 could hardly mark an advance upon previous efforts. 



I hope the sections actually figured will be sufficient to show the completely 

 serial nature of the work, though all the stages have been worked through in all 

 three planes of orientation, as also the transition stages. Naturally, the somewhat 

 revolutionary result in regard to the fate of the dextral ccelomic elements has led me 

 to hesitate in publishing my conclusions, especially as badly orientated specimens 

 deceived me at first into supposing that Cribrella agreed with the usually accepted 

 account of other asterids. Whether the conclusions drawn from the development of 

 Cribrella are accepted by authorities on the group or not, the facts as described here 

 may, I hope, be relied upon as accurate ; and I have, by using general names in the 

 descriptive part, tried to keep separate the two parts of the paper, i.e., the facts and 

 the conclusions which are drawn therefrom. 



The development from the fertilised egg onwards may be conveniently divided into 

 three parts : — 



( 1 ) The embryonic period up to the time of hatching, comprising the phenomena of 

 segmentation, gastrulation and formation of the ciliated demersal larva. 



(2) The period of larval life; the earlier part free, and the later more or less 

 attached by the sucker. 



(3) The post-larval period, during which the larval body becomes of secondary 

 import to that of the young starfish. This might be termed the period of meta- 

 morphosis, except that there is no real metamorphosis or reconstruction. It is one of 

 the most important features of an echinoderm development such as this, that there is 

 no real histolytic or metamorphic stage, and that larval organs are merely moulded 

 into those of the adult, if we except the reduction of the pre-oral lobe. 



Upon the actual duration of development it is difficult to speak with certainty 





