484 ME. F. B. BEDDAED OS THE ATEItJM AND [May 16, 



genera PontodHlus, Microscole.i; &c. as the lowest forms of the 

 Crjptodrilidse except iu the sense that they are degenerate forms ; 

 it seems to me that it is the Eudrilidae which present, as regards 

 the atria, the most obvious likeness to the Geoscolicidae. In them, 

 as in the Geoscolicidse, the atrium opens into a terminal " bulbus " 

 which ma_y even, as in EudrUus, possess a pad developed out of its 

 lining epithelium apparently not uulike the " scudo ovale" described 

 by Eosa ; the glandular appendix which in the Eudrilidse opens into 

 this is sometimes (as in Heliodrilus) hardly differentiated, as it is iu 

 many others, into a muscular and a glandular portion. In Eudrilids 

 the position of the pore of the sperm-duct into the " appendix " is 

 so variable that there is no difficulty in comparing the two families 

 in respect of the fact that in both the sperm-duct opens into the 

 glandular appendix. Even among the other subfamilies of the 

 Megascolicidse (I am here following Eosa's classification) there 

 are not wanting indications of a close approximation between the 

 atria and those of these Geoscolicids : in some species of the genus 

 PericJirfta, for example, the atria open into a sue variable in size, 

 which itself opens on to the exterior ; it appears to me that this 

 sac is the equivalent of the terminal sac in the Geoscolicidse. As 

 to the presence of a retractor, which Eosa thinks distinguishes the 

 atrium of the Geoscolicidse, many Megascolicidse have bands of 

 muscles which appear to me to be perfectly comparable ; for 

 instance, in my genus Octochcvtus (9) among the Acanthodrilids 

 there are bands of muscles which, though perhaps not exactly 

 inserted on to the atria, are attached to the body-wall iu their 

 immediate neighbourhood. A better example still is furnished by 

 two species of Eudriloides lately examined by me ; in these there 

 are a series of muscular strands actually inserted on to the 

 muscular termination of the two atria. To assert that these are 

 in their nature different from those of the Geoscolicids seems to be 

 too strong. 



It seems therefore that the facts allow of no other view save 

 that the various structures termed atrium by myself are homo- 

 logous. The extremes are united by too complete a series of 

 intermediade forms to permit of any doubt upon the point. There 

 are, however, as has been pointed nut, differences in detail between 

 the atria of different groups ; these amount to so much in the 

 entire series that the complex glaud of EudrUvs, or Perichaia, 

 would perhaps unhesitatingly be regarded as different from the 

 simple atrium of Stylodrilus, were there no intermediate stages. 

 Though there would appear to be no great difficulty in deriving 

 the atrium of one family from that of another, it is not so apparent 

 which are the ancestral and which are the derived forms. To 

 determine or attempt to determine this, opens up the whole 

 question of the classification of these Annelids ; and in my opinion 

 a consideration of the facts relating to the at ri i confirms for the 

 most part the scheme of classification adopted bj' myself. In 

 order to determine which form of atrium is most primitive, it is 

 necessarj' to enquire into the origin of the atrium. 



