ANATOMICAI> STRUCTURE OF KKKKIA. 3OI 



DiverticnIiDu (Hgs. i, 2, 18, 19 and 21). There is, as 

 in Ocnerodrilus, a paired oesophageal diverticulum in 

 somite ix, emanating in the anterior part of the somite, 

 and not in the posterior part as in Ocnerodrilus. The 

 junction with the oesophagus is situated in the upper part 

 of the (jesophagus and the diverticula are so bent down- 

 ward that they are not seen when the oesophagus is viewed 

 from above, being actually covered from above by the 

 upper projecting wall of the oesophagus. 



Both diverticula point directly downward and only 

 slightly forward. Thev are much rounded and blunt 

 as regards exterior form, while the inner structure re- 

 sembles that of the diverticulum in Ocnerodrilus. Com- 

 pared to the diverticula of Kerria halofhila, they are 

 much smaller than in that species. In fact the whole 

 alimentar}' canal differs considerably from Beddard's 

 form. In Kerria halophila there is a gizzard in somite 

 vii, the sacculation of the oesophagus is insignificant and 

 lastly the diverticula are very large filling the whole somite. 

 But they connect with the oesophagus in the anterior part 

 of the somite in the same way as in m}- new forms, and 

 this appears to be a characteristic of the genus, and one 

 to which some importance may be attached. The saccu- 

 lated intestine commences in Kerria halophila in somite 

 xiii, which is one somite further back than in our species. 

 This is probably a good species characteristic, as I have 

 found the same variation in species of Ocnerodrilus and 

 there it appeared constant. There is only one interior 

 cavity with large projecting ridges, one of which is much 

 larger than the other. That these ridges in Kerria as 

 well as in Ocnerodrilus constitute the first differentiation 

 towards a real interior division is evident. While describ- 

 ing a number of species of Ocnerodrilus I failed to find 

 a single specimen with a perfectly divided diverticulum. 



