286 C. M. CHILD. 



lively simple. It should be borne in mind, however, that cells 

 and tissues of a certain constitution, i. e., Cerianthus-protoplasm, 

 must be acted upon in order that a certain result characteristic of 

 the species may be obtained. In another species the result of a 

 stimulus similar in degree and kind may be more or less widely 

 different. At best the problem of form is complex but I believe 

 that in general we must look to function for an explanation of 

 form, whether direct or indirect, and not to form for an explanation 

 of function. This opinion does not involve the paradox of dif- 

 ferentiation of function in a structurally isotropic protoplasm, but 

 requires a distinction between chemical and physical structure of 

 protoplasm and morphological structure in general. Function in 

 a general sense is an expression and result of the first, while the 

 second is, at least in many cases, an expression and result of 



function. 



EXPERIMENTAL DUPLICATION OF PARTS. 



By partial longitudinal splitting of the body, the separated por- 

 tions remaining attached to the undivided part, it was possible 

 to produce specimens with two oral or aboral ends as the case 

 might be. In cases where one of the parts separated by the 

 longitudinal cut was considerably smaller than the other, and 

 especially in cases where the cut was somewhat oblique, the 

 smaller part or that attached to the undivided region by the nar- 

 rower portion very commonly separated from the remainder of 

 the body within a few days and regenerated- independently. In 

 many cases also closure was much delayed by the irregular in- 

 rolling of the edges and one or both of the split portions became 

 greatly reduced or constricted off. In a certain number of the 

 pieces, however, the split portions roll longitudinally and their 

 edges unite thus giving rise to the duplication. In case of dup- 

 lication of the oral end the two parts are in reality " halves," each 

 possessing approximately half the full number of tentacles. 

 Where closure is perfect or nearly so each disc of course pos- 

 sesses a mouth and oesophagus formed by union of the longitudi- 

 nal cut surfaces of the half oesophagus in reality half structure. 

 Whether in cases where one of these half structures contains the 

 old siphonoglyphe the other ever forms a new one I do not know. 

 I have never found a new one, but it may appear in time. 



