256 THE INVERTEBRATA 



obvious that the mechanism of copulation is far more complicated in 

 the Lumbricidae. Here the worms come into contact along their 

 ventral surfaces and each becomes enveloped in a mucous sheath. Close 

 adhesion is secured between the clitellum of one worm and the seg- 

 ments 9 and 10 of the other, partly by embracing movements of the 

 clitellum and partly by the chaetae of the same region being thrust 

 far into the body wall of the partner. The sperm passes out of the 

 male aperture and along the seminal groove to the clitellum ; how it 

 enters the spermathecae of the other worm has never been observed. 



The cocoons are formed some time after copulation. The worm 

 forms a mucous tube as in copulation. The cocoon is then secreted 

 round the clitellum and finally the albuminous fluid which nourishes 

 the embryo is formed between the cocoon and the body wall and the 

 worm frees itself from the cocoon by a series of jerks. All three 

 products, mucus, cocoon substance and albumen, are secreted by the 

 clitellum and each probably by a distinct type of cell. The eggs are 

 sometimes extruded and passed backwards into the cocoon while it 

 is still in position on the clitellum but the spermathecae eject the 

 spermatozoa when the cocoon passes over them. 



The embryo of Lumbricus is illustrated in Fig. 189. The prototroch 

 is absent but the gut and stomodaeum are developed early to absorb 

 the albumen in the cocoon. There are two mesoblast pole cells at the 

 hinder end which bud off the mesodermal strips: there are three 

 ectodermal pole cells on each side, the ventralmost a neuroblast 

 forming half the nerve cord and the two others nephroblasts giving rise 

 to longitudinal rows of cells which divide up to form the nephridia. 



The most primitive kind of nephridium (nephromixium) is that 

 described in LumbricuSy of which there is a pair for each segment, the 

 nephrostome projecting through the septum and opening into the 

 cavity of the segment in front. A great many modifications of this 

 arrangement exist especially in the Megascolecidae. Here, in addition 

 to the type already described which is distinguished as a mega- 

 nephridium, there are micronephridia of which enormous numbers may 

 exist in a single segment (2500 in Pheretima). These are small tubes 

 which may or may not open into the coelom by a nephrostome. They 

 may exist in the same segment as a pair of meganephridia. There is 

 good evidence for supposing that an originally single meganephridium 

 has been broken up into a multitude of micronephridia. In the 

 development of the earthworm Megascolides the segmental organs 

 first appear as cords of cells like meganephridia. These are thrown 

 into a series of loops and each loop is separated from the rest as a 

 micronephridium . 



Other modifications are those in which the nephridia open into the 

 alimentary canal instead of to the exterior. They maybe peptonephridiay 



