372 Ralph S. Lillie 



boundary is sharply defiued; anteriorly it passes by a graduai tran- 

 sition into tlie move fully differeutiated region in front. The pygidial 

 regiou thus takes no part in the formation of somites, but is carried 

 passively backwards as the body elongates and preserves unchan- 

 ged its peculiar characteis tbrougliout the whole development. 



The histological structure of the growing zone may be seen by 

 a refcrence to the ligures above mentioned. Cross sections show 

 that the ectoderm of the extreme posterior region of the body of 

 the young larvae is greatly thickened, and of almost uniform dia- 

 nieter in ali portions of its circumference (Figs. 14, 21 and 24, 

 Piate 23). Figure 14 is a section passìng through the posterior region 

 of a larva of 5^ — 6 somites. The section is slightly oblique; dorsally 

 it cuts through the anterior portion of the paratrochal region, and ven- 

 trally through the region of commencing differentiation at the posterior 

 border of somite V. The ectoderm of the growiug zone is seen to 

 be composed of a single layer of large cubical cells containing 

 large clear nuclei, each of which usually exhibits a well-defined 

 nucleolus. On passing forward from this zone the dorsal and lateral 

 regions of the ectoderm are seen gradually to become thiuner, the 

 cell-limits become less sharply delined, the number of nuclei in 

 cross section becomes fewer, and the nuclei themselves become smaller 

 and more deeply stained, gradually assumiug the usuai charac- 

 teristics of the tissue-nuclei. The ventral ectoderm also becomes 

 thinner anteriorly, although to a less exteut than the dorsal and 

 lateral regions; and the ventral nerve-cord appears as a fibrillar 

 differentiation of the inner ends of its cells. 



The region of large cells with the large clear nuclei thus evi- 

 dently corresponds to the undifterentiated embryonic region of the 

 growing zone, from which cells are divided off anteriorly; in other 

 words, to the teloblastic region. Definite individualized teloblasts 

 however are not present in Arenicola at this stage. The number 

 of cells in cross section is not Constant and the cells show no dis- 

 position to become arranged in rows, but simply form an undiffe- 

 rentiated zone which passes by a graduai transition into the more 

 diflerentiated ectoderm in advance. Postcriorly the ectodermal cells 

 of the growing zone are, as already mentioned, sharply delimited 

 from the small, radially arranged, pigmented cells of the adjoining 

 pygidial region (Piate 24, Fig. 32, 33). 



The thinning-out which the ectoderm shows on passing forward 

 from the growing zone is in ali probability due in large part to a 



