no 



1 have succeeded in following step by step the segmentation of 

 laeramarina, Asellus coniinunis^ Porcellio scaber and Ai'madültdium sp.?, 

 and have also observed certain stages in the early development of 

 Oniscus murarius and Philoscia tittata. As a result I am able to state 

 with certainty that in none of these forms does a telolecithal segmen- 

 tation occur, but in all it is of the same type and is typically centroleci- 

 thal. In Porcellio and Armadillidiutn, to confine our attention to these, 

 immediately after fertilisation the nucleus occupies practically the 

 centre of the egg , where it lies imbedded in a mass of protoplasm 

 from which delicate processes radiate off into the yolk. Enclosing the 

 yolk is a delicate layer of protoplasm entirely destitute of nuclei and 

 probably , to judge by what occurs in the egg of laera , united with 

 the central nucleated mass by a reticulum of protoplasm, the yolk 

 being distributed in the meshes of this reticulum. 



The segmentation begins with a division of the simple central 

 nucleus and of the protoplasm which surrounds it, the peripheral 

 protoplasm and the yolk remaining undivided , as indeed they do 

 throughout the entire process. Thus two nuclei, each surrounded by 

 a stellate mass of protoplasm, [; are at this stage to be found near the 

 centre of the egg. This division is repeated, so that four , then eight, 

 and then sixteen nuclei, each surrounded with its mass of protoplasm, 

 are formed. But as the division proceeds the nuclei come to lie nearer 

 and nearer the surface of the ovum , until finally they reach the sur- 

 face and their protoplasmic envelopes fuse with the peripheral proto- 

 plasm , several of the nuclei at the same time approaching each other 

 somewhat , so as to form the anläge of the blastoderm , the remaining 

 ones being scattered at intervals over the surface of the Q%^. 



The details of the segmentation , and certain interesting pheno- 

 mena which accompany it, will be fully described with figures in my 

 forthcoming paper and I will confine myself at present to pointing out 

 that the segmentation of Porcellio and Armadillidium and of the 

 other terrestrial Isopods mentioned above, is identical in its character 

 with that of Asellus, the only notable difference being that in the latter 

 form a cleavage of the yolk into what Reichenbach has termed 

 primary yolk pyramids occurs at a certain stage , though wanting in 

 the earlier stages. 



These statements are at utter variance with the observations 

 recorded by Roule. It can hardly be possible that the European 

 Porcellio differs so markedly in its developmental processes from its 

 American representative as the two accounts would imply. My obser- 

 vations may be readily verified by the use of proper methods ^ and to 



2 Such as fixing in alcoholic picro-sulphuric acid, staining in Klein e nb erg' s 



