INVEBTEBRITA, CRYPTOGAMIA, MICROSCOPY, ETC. 651 



of the ovtiiy, is composed of the same cells (furnished with a vibratilc 

 cilium) as the eutoclerm of the radial canals. 



Towards the base of the ovary, where it becomes confounded with 

 the lower surface of the bell, the entodermic layer is as yet only 

 formed of a single stratum, as in the radial canal ; but in proportion 

 as we advance towards the interior of the ovary we see the entodermic 

 cells divide in a direction perpendicular to their length, and thus form 

 two superposed layers of entoderm ; the division of the cells continuing 

 in all directions, we thus hud the entoderm grow thicker and thicker. 



Between these two lamellaB of entoderm and ectoderm forming the 

 ovary, is a third more delicate, structureless lamella — the intermediate 

 lamella — sharply separating them and assisting us to define with 

 certainty which layer produces the ova. These ova are always found 

 under the intermediate lamella, and being thus separated from the 

 ectoderm by that lamella can only be developed from the entoderm. 

 This is further borne out by observing directly all the graduated 

 transitions between the ordinary entodermic cells of the young ova. 

 The changes in an entodermic cell destined to be developed into an 

 ovum consist in the increase of the vohime of this cell and the trans- 

 formation of the nucleus into a germinal spot. 



In the entodermic cells lining the radial canals the protoplasm is 

 perfectly transparent and devoid of granules ; the nucleus appears as a 

 clear round spot containing a central round and denser nucleolus. 

 The cells with their nuclei and nucleoli subsequently increase in size, 

 and the protoplasm becomes more and more granular. The nucleolus, 

 at first simple and furnished with a small vacuole, commences to 

 divide. It lengthens, becomes constricted in the middle, curves into 

 the form of a horse-shoe, and finally divides into two parts, each 

 possessing a central vacuole ; each half again divides into two parts, 

 but in a direction perpendicular to the first, and so on. 



These phenomena, constant and normal in the Medusfe of the 

 White Sea, are the exception in those from the Bay of Naples. In 

 the latter the division of the nucleus takes place in a different 

 manner. AV^hen the nucleolus, after elongation, presents a median 

 constriction, it does not divide into two parts, but simi)ly elongates in 

 the form of a band twisted upon itself; constrictions then foi-ming at 

 several parts, it becomes a long moniliform ribbon rolled up in 

 several turns. Each division of the chaplet is fusiform and round ; 

 it regularly contains in the middle a very small vacuole, and is 

 united to tlio neighbouring divisions by a thin and sometimes rather 

 long articulation. Sometimes tliis band, which reminds us of tlio 

 nucleus of some of the Infusoria (Sicnior, Sjnrosttimnm), splits into two. 

 Finally the articulations of the chaplet separate, and instead of a 

 nucleolus, there is formed at the centre of the nucleus a group of 

 several dozens of small round balls which collect into a si)hero placed 

 at simic distance from the walls of tlie nucleus. These balls continue 

 to divide until they reach several hundreds in number. During all 

 this time the ovum enlarges and attains its definitive diameter, which 

 surpasses nearly twenty times that of tho entodermic cells from wliicli 

 it originates. 



