1893.] PROSTATE IH" THE OLIGOCH^TA. 481 



types is BotJirionem-on (IZ), where the " Cement-Driise " is attached 

 to the summii; of a short diverticulum of the atrium. 



It seems to me impossible to resist the conviction that we have 

 Tiere a series of stages in the production of a limited " Cement-Driise " 

 out of a continuous glandular investment of the atrium. In MoniJi- 

 gaster, whose atrium as regards the points under discussion is so 

 like that of Branchiura, the glandular cells which invest it ex- 

 ternally perforate the muscular layer to reach the inner epithelium ; 

 I imagine that this will be found to be the case with Branchiura 

 too. A reduction of this investing layer to a restricted region of 

 the atrium brings about the Cement-Driise of Tuhifex ; a greater 

 separation between the masses of glandular cells would produce 

 the state of affairs characteristic of Telmatodrilus. It is noticeable 

 that the absence of pei'itoneum upon the atrium of Branchiura is 

 seen in Tuhifex in the region of the Cement-Driise ; why this 

 should be I cannot suggest ; but that it is so is a further reason 

 for the justice of my comparisons. 



In the family Lumbriculidse there are indications of a similar 

 series of stages in the conversion of a complete glandular covering 

 of the atrium into a more partial one ; but the stages are fewer 

 than in the Tubificidse, in fact there are only, so far as our present 

 knowledge of the group goes, two such stages. The more usual 

 condition is seen in the majority ; in Hhgnchelmis, for example, 

 the atrium is lined by a layer of cells, outside which are masses of 

 pear-shaped cells which are aggregated to some extent into separate 

 masses ; the after all rather slight indications of discontinuity 

 between the masses of cells which clothe the atrium in Bhyn- 

 chelmis are still further emphasized in Sutroa. In this Annelid (8) 

 the atrium is invested by very distinctly separate masses of 

 glandular pear-shaped cells; the entire atrium is covered by a thin 

 muscular layer which might be regarded as the peritoneal in- 

 vestment ; between this supposed peritoneum and the glandular 

 coat of the atrium lie masses of developing sperm ; the structure 

 therefore, whatever be its morphological nature, is, in function, a 

 sperm-sac. 



It seems to me that the two families Tubificidse and Lumbriculidae 

 are nearly allied, in any case no one has disputed the morphological 

 identity of the organs termed atria in both ; it is therefore reason- 

 able, at least pending further information, to assume that the atria 

 correspond in detail, and in this case the Lumbriculidas, like the 

 Tubificid genus Branchiura, have an atrium which is frequently 

 devoid of peritoneum. The intermediate condition between the 

 atrium in the Lumbriculidae and that in the higher Oligochseta is 

 afforded by the Moniligastridse. Moniligaster itself has an atrium 

 which seems to be identical in all essentials with that of Branchiura 

 and the Lumbriculidae ; it has two layers of cells which are sepa- 

 rated by a well-developed muscular layer. I have shown (9) that 

 the outer layer of cells sends processes through the muscular layer, 

 which therefore are in a position to void their contents into the 

 lumen of the atrium. I myself have only seen this arrangement 



