484 



of them being crowded into a comparatively small space. The eggs 

 are thus very much distorted and never assume a spherical form. 

 Assuming as they do a great variety of shapes, it is difficult to get a 

 view of them which shows correctly what is taking place. This is par- 

 ticularly true of the earliest stages. To add to this difficulty the spadix 

 is continually undergoing movements, which constantly changes the 

 view. To avoid the sources of error thus incurred, I removed the egg 

 completely from the medusa and examined it by itself. Further I have 

 hardened the eggs in osmic acid and studied them by sections. Roth 

 methods show conclusively that there is no approach to an epibolic 

 gastrula. The segmentation is not perfectly regular and very frequently 

 presents appearances which resemble an epibolic gastrula. Further 

 study however shows that this resemblance is only superficial, being 

 due to slight irregularities and to difficiilties of observation. The seg- 

 mentation proceeds in a perfectly normal way and a typical morula is 

 reached. The morula is never spherical but is somewhat disk-shaped 

 owing to the pressure of the medusa wall. As the segmentation goes 

 on the cells become less and less distinct, until finally a stage is 

 reached in which all traces of segmentation have externally dissapeared. 

 The embryo lies now quite freely in the medusa cavity, entirely free 

 from the spadix. It is this which Al Iman has considered the Tubula- 

 rian egg. The stages of segmentation escaped his notice entirely. It 

 is really as we see the egg in quite an advanced stage of development. 

 The embryo is now transformed into an actinula, essentially as Cla- 

 mi eia n describes. 



Sections at various stages confirm the above results. No traces of 

 a differentiation into two layers is seen until the segmentation is quite 

 far advanced. The ectoderm makes its appearance quite slowly and 

 sections show it to be a true determinate ectoderm. The endoderm 

 appears somewhat later. After the ectoderm has become perfectly di- 

 stinct, a layer of cells, very indistinct at first but becoming more and 

 more definite, is seen just within the ectoderm. This endoderm ab- 

 sorbes the food mass within and soon forms a layer, one row of cells 

 thick. At the bases of the tentacles however it is several rows thick. 

 Just before the actinula becomes free two layers are perfectly definite 

 and enclose a distinct central cavity. 



The segmentation and formation of the germinal layers as above 

 given, coincides completely with Coelenterates in general. Tubtilaria, 

 which has been considered somewhat of an anomaly in hydroid deve- 

 lopment, presents thus no noteworthy differences from the rest of the 

 Hydroids. 



Druck von Ereitkopf und Härte! in Leipzig. 



