268 



intestine to assist digestion, as described originally by d'Ude- 

 kem, being, in fact, unicellular glands ; but they are also 

 thrown off into the perivisceral cavity and form a great 

 number of its corpuscles. The endothelial cells of Limno- 

 drilus seen on its segment organs, and incompletely lining 

 the muscular body-walls, are large, clear, and globose, and 

 these too are throvvn off to contribute to the perivisceral cor- 

 puscles. In Enchytraiis the cells of the hepatic membrane 

 have a very definite oat shape, and so have those of the peri- 

 visceral fluid, and they are very numerous. (PI. XVII, fig. 6.) 

 Their form is so well marked that they are quite characteristic 

 of the genus. The perivisceral corpuscles differ from the 

 hepatic only in having lost their yellow granular contents. 



In Lumbriculus , besides elongate corpuscles from the 

 hepatic membrane, are also to be found somewhat round cor- 

 puscles with large amoeboid pseudopodia, which apparently 

 are thrown off — that is to say migrated — from the endothe- 

 lium. (PI. XVII, fig. 1.) In addition to these, in Tuhi- 

 fex and Lumhriculus there are very long, band-like cor- 

 puscles, which are nothing else but migrated muscular- 

 fibre cells. (Plate XVII, fig. 3.) In Tubifex the endothe- 

 lium of the perivisceral cavity is very incomplete, being 

 chiefly developed in a narrow band on each side of the cavity 

 between the dorsal and lateral rows of setae ; the muscular 

 tissue consequently presents a free surface to the perivisceral 

 fluid, and the cellular elements of the muscular tissue may 

 be seen in the act of separating, to pass a floating existence 

 in the perivisceral fluid. Besides these various contributions 

 to the perivisceral corpuscles, there is a more remarkable lot, 

 which are derived from the generative organs. 



The Oligochseta do not expend all the generative elements 

 which they develope. The large masses of sperm-cells and 

 of ova which commence development do not all attain maturity 

 and become extruded ; but a considerable number, varying in 

 individuals, undergo a degeneration, and become corpuscles of 

 the perivisceral fluid, where they are especially to be observed 

 in the autumn months. (PI. XVII, fig. 2.) The dimorphism of 

 the ovary in Tubifex observed last year by Dr. Fritz Ratzel, 

 is, I believe, to be explained by the fact that after a certain 

 number of ova have been deposited, or in accordance with 

 other conditions as to copulation, &c., the ova do not develop 

 to their full size, but form closely packed masses of cells of a 

 smaller size (his Enchytrajus form of ovary), which gradually 

 lose the character of ova altogether, and are at length sepa- 

 rated, and float in the perivisceral cavity as corpuscles. Such 

 a series of changes I have carefully traced in Tubifex. The 



