vol. 8.1 Robertson. — Embryonic Fission in Grisia. 129 



has been started. He considers that the very thin wall of the 

 anterior end of the ovicell is not impenetrable to the sperma 



tozoa. If, indeed, the spermatozoa reach the ova at all, they 

 must penetrate the tissues of the colony at some point. Whitman 

 ('!»(}) has shown that a method of impregnation somewhat similar 

 to this is not uncommon in several groups of animals. In most 

 of the cases he mentions the spermatozoa are forcibly injected 

 through the cuticle, and wandering through the tissues, sonic 

 succeed in reaching the ova. Grisia is covered with a calcareous 

 layer which is pierced at intervals by pores that extend through 

 the chitinous ectocyst beneath it. The epithelial cells of the 

 body wall pass through these pores and spread out over the 

 surface, forming a very thin layer upon it. These pores afford 

 innumerable points where spermatozoa could effect an entrance. 

 Moreover, near the growing tip the outer covering becomes 

 thinner and the deposition of calcareous material does not keep 

 pace with the growth of the branches, so that the growing points 

 are covered with an extremely delicate chitinous layer only. 

 Since, as has been shown, the ovaries are situated at the growing 

 tips, it is practicable for fertilization to take place before, or at 

 the time that the ovum becomes associated with the hud. The 

 occasional occurrence of embryos in a zooecium, as for example 

 in the case shown iu PI. XIII, Fig. 1C (nub.), where cleavage 

 has occurred, indicates fertilization thus early, as does also the 

 early cleavage in an undoubted ovicell shown in PI. XIII, Figs. 

 19 and "20. One of the blastomeres resulting from the first 

 cleavage is shown in each of these figures (bl.). They are not 

 surrounded by the cells of the polypide bud (/»/. /»/.). and yet 

 the first division has taken place, so that cleavage occurs, appar- 

 ently, at or before the time that the ovum is surrounded by the 

 cells of the bud, and before the ovicell is formed. 



The view that Harmer advances in regard to the time of 

 fertilization is based upon his belief that the ovum is the product 

 of the polypide bud ('97). He considers that only certain buds 

 in each internode produce eggs, that these are ecpiivalent to fertile 

 polypides, and that they give rise to ovicells. The evidence from 

 my own observations, however, proves that eggs are produced in 

 every terminal internode, independently of either buds or poly- 



