ASEXUAL REPRODUCTION IN SAGARTIA 203 



the so-called ventral pair of directives. Type II might be 

 considered as having been produced by reduction from either a 

 bilateral or a biradial condition, but Carlgren regards it as a 

 modification of Type I through the failure to develop of the 

 innermost mesenteries (d) there represented. According to this 

 view, then, d of Type I is lacking in Type II, and c^ of Type I 

 becomes d of Type 11. Type IV is a combination of Types I 

 and II, and Type X is the result of a doubling of Type I. Type 

 VII represents an extreme expression of the tendency to the 

 bilateral arrangement of mesenteries shown normally in Type I. 

 Type IX represents normal development, such as Type I, in a 

 basal piece having no part of the old column to interfere with 

 the formation of the ventral directives. In explanation of Type 

 III it is assumed that there are here two regenerating regions, 

 each similar to that of Type II. In one of these regions the 

 directive mesenteries are replaced by mesenteries of the old 

 part. Type VIII is similar to Type III, but lacks old parts 

 which might occupy the position of one pair of directives. Type 

 V is a mixture of Types II and III; Type VI, of I and III. 



It may be pointed out that the explanation for Type III is 

 not entirely in harmony with the order of development of that 

 type, but would demand the order c^ and c\ d, c- and c^. It 

 seems to me, however, that we cannot profitably consider at 

 present the more abstract questions of the relation between the 

 biradial and bilateral types of development. We cannot expect 

 to solve these problems of form determination from examination 

 of data collected for other purposes and assembled into tables 

 and diagrams. The most that can be expected of such material 

 from this standpoint is that it may present definite problems and 

 suggest favorable points of attack. 



One of these problems concerns in a concrete way the relations 

 between the biradial and bilateral plans of regeneration. Carl- 

 gren has shown that both types may occur in the same species 

 or even in the same regenerating pieces, and he has made a 

 beginning in ascertaining the conditions determining the plan of 

 development. He has shown that in Metridium larger pieces 

 containing part of the base, column and mesenteries develop, 



