478 MEDUSAE OF THE WORLD. 



component cells incloses the other. The inclosed cell then divides and develops into an 

 embryo, while the enveloping cell becomes very large, sends out pseudopodia-like processes, 

 and serves apparently only for the attachment, locomotion, and nutrition of the embryo. The 

 embryo grows and hangs freely in the gastric cavity of the Cunina proboscidea, while the envel- 

 oping giant-cell attaches it to the entodermal wall. The embryo then becomes free in the 

 gastric cavity of the Cunina and develops into a medusa which produces medusa-buds from 

 its aboral pole. These budded medusae are sexually mature when they emerge into the ocean 

 and have only 4 tentacles, no marginal lappets, 20 sensory-clubs, no otoporpae, no stomach- 

 pouches, and a ring-like gonad in the ectoderm of the subumbrella. They thus resemble a 

 Solmaris rather than a Cunina. 



MetschnikofF calls this extraordinary developmental process "sporogeny." 



Stschelkanowzeff, 1905, 1906, however, in a carefully studied research, has given an 

 explanation of this remarkable development of medusae within the stomach-cavity of C. pro- 

 boscidea which is probably more nearly correct, and which would dispense with MetschnikofF's 

 hypothesis of "sporogeny." 



According to Stschelkanowzeff both the male and female sexual products originate in the 

 ectoderm of the stomach-wall of the free-swimming medusa of C. proboscidea. The eggs, 

 however, migrate through the supporting lamella into the entoderm of the stomach-wall of the 

 medusa and are there fertilized by the small oval spermatozoa. There are 30 chromosomes. 

 The segmentation is total and practically equal and a solid morula develops. This morula 

 comes to lie in the gastrovascular cavity of the medusa, but retains an attachment with the 

 entodermal wall, which forms an open fold or follicle around it. The entoderm of the larva is 

 formed by delamination, the innermost cells of the morula becoming entodermal. Thus we 

 have a small, sac-shaped larva with its thick, outer ectodermal wall composed of several layers 

 of cells and a single layer of smaller cells constituting its degenerate entoderm. Then 2 ten- 

 tacles grow out from the aboral surface of the larva, and soon 2 others, 90 apart from the 

 original pair, appear, giving in all 4 tentacles. The rudimentary bell and velum then grow 

 out from the sides of the body of the larva, about 15 sensory-clubs appear at intervals around 

 the margin, and a rudimentary marginal nervous system develops. Practically no gelatinous 

 substance is found and the velum, tentacles, and bell collar are very small and rudimentary. 

 The mouth breaks through on the upper side of the larva and sexual products (male or female) 

 develop in the ectoderm of the subumbrella wall of the stomach of the medusa. The medusa 

 then lies with its mouth and velum uppermost in the gastrovascular cavity of its mother. 

 Soon, however, it degenerates into a mere sac filled with genital products and thus ends the 

 second generation. Stschelkanowzeff finds no nurse-cell in the early stages of development 

 of the larvae, such as MetschnikofF observed. 



Stschelkanowzeff has shown that it is probable that the eggs from this second generation 

 become free and develop into planula larvae which attach themselves to the stomach or lips of 

 the medusa Geryonia. Here they develop numerous medusa-buds, which are in turn set free 

 and develop into Cunina proboscidea. It appears, then, that there are three generations of 

 this medusa as follows: 



I. The large free-swimming Cunina proboscidea having 9 to 14 tentacles. 

 II. The small 4-tentacled medusae, which develop from the eggs of C. proboscidea and are found 



attached to the stomach-wall within the gastrovascular cavity of their mother. 

 111. The budding stolon-like larva;, which develop from the eggs of the small, 4-tentacled medusie 

 and are parasitic upon the medusa Geryonia. The medusae asexually set free from this 

 parasitic larva are Cunina proboscidea. 



As Stschelkanowzeff shows, MetschnikofF probably mistook the eggs of C. proboscidea 

 for "spores." Stschelkanowzeff could find no asexual budding in the medusae developing 

 within the stomach-cavity of C. proboscidea, although this has been observed by Metsch- 

 nikofF and also by H. B. Bigelow and by the author in other species. It is probable, how- 

 ever, that it is not of invariable occurrence. There is a marked discrepancy between the 

 observations of MetschnikofF and those of Stschelkanowzeff in respect to the existence of 

 the giant nurse-cells, but Bigelow finds them in Pegantha smaragdina, and indeed Stschel- 



