ASEXUAL REPRODUCTION IN SAGARTIA 189 



from an experimental cut. Mesenteries of this regeneration are 

 represented in table 5 by Roman type. The specimen divided 

 into two parts, a and be, in incomplete endocoels of the first 

 grade, but with the loss of one of the bounding mesenteries in a. 

 Eighteen days after this divisioijf, fission again occurred in be, 

 producing moieties b and c. On one side (the left in the table) 

 this new fission plane passed again adjacent to the same old 

 incomplete mesentery of first grade, which in this division 

 passed to c. Meanwhile the normal number of mesenteries had 

 been established, including the mate to this old bounding 

 mesentery, now represented on the far right of the formula of b. 

 The case of no. 18 is somewhat different. Here the division into 

 a and be was followed after but two days by the separation of 

 be into b and c. This occurred on one side along the boundary 

 between old tissue and that just beginning to regenerate. The 

 early state of this regeneration at the time of the second division 

 may have had something to do with the irregularities indicated 

 by the queries as to original completeness or incompleteness of 

 two of the mesenteries. Certainly, in the two days that elapsed 

 between the divisions, few mesenteries could have been estab- 

 lished in the earlier regenerating zone, and none would give any 

 indication of their final state of completeness if this had been 

 determined. It appears that, while a regenerated normally, pro- 

 ducing a mate to its incomplete bounding mesentery, the corre- 

 sponding regenerating zone of b shows no corresponding incom- 

 plete bounding mesentery (which would be represented on the 

 far right of the upper row of the formula for no. 18b). Rather, 

 the two regenerating zones have matched up together, with two 

 complete mesenteries from either side of the boundary forming 

 a new pair. It must be noted that the uncertainties of identi- 

 fication in this case are such that great dependence must not be 

 placed on the interpretation here suggested. Such a juxtapo- 

 sition of two regenerating areas may account for the large number 

 of mesenteries (including two pairs of directives) recorded in an 

 apparently single regenerating region in specimens nos. 88 and 89 

 of table 8. 



