472 WILSON. [Vol. XI. 



dogmatic assumption that each and every localized form of 

 cell-activity must be referred to the agency of corresponding 

 pre-formed material germs. It has, moreover, the advantage 

 of enabling us to conceive the external aspect of the process 

 by which "permanent" cell-organs may have arisen in a his- 

 torical sense; and it enables us to bring under a common point 

 of view the origin of identical structures either by "free forma- 

 tion " or by the division of a preexisting parent-structure, as 

 would seem to be the case with the centriole,^ and is possibly 

 the case with the centrosome as a whole.^ The uninterrupted 

 continuance of a specific form of metabolic activity at a par- 

 ticular area may be accompanied by the persistence of the 

 cell-organ that is its morphological expression ; its cessation 

 may lead to the disappearance of the organ, as occurs in case 

 of the centrosome in echinoderm eggs after formation of the 

 polar bodies; the resumption of the action may create the 

 organ anew. Thus we are led by a study of cell-organization 

 to a point of view not far from that which Sachs has developed 

 on a quite different basis for the organization of the plant body,^ 

 and from which Loeb has so suggestively considered the devel- 

 opment of animals.* This view gives in itself, it is true, no 

 immediate insight into the mystery that still envelopes the 

 orderly determinations of differentiations; yet it seems to me 

 full of suggestions for further study. 



Biological Laboratory of Columbia College, 

 July, 1895. 



1 The fact may be recalled that in the Metazoa exactly equivalent structures 

 may be formed either independently or by the fission of a parent structure, as, for 

 instance, the gill-slits of ascidians. (See Willey, Amphioxus and the Ancestry of 

 the Vertebrates, Macmillan, 1S94.) 



2 The most careful search has thus far failed to reveal the existence of a cen- 

 trosome in some tissue-cells, which are nevertheless capable, under appropriate 

 stimulus, of undergoing typical karyokinetic division (as, for instance, the cells of 

 the pancreas and kidney). In the latter case the "free formation " of the centro- 

 some must be regarded as an open possibility, as has been admitted by a number 

 of competent writers. (See Heidenhain, Neue Untersuchungen Uber die Central- 

 korper, pp. 654, 655 ; Watase, Origin of the Centrosome, Biological Lectures, III, 

 p. 287.) 



^ Stoff und Form der Pflanzenorgane, Gesa??imelte Abkandhmgen, II, 1893. 

 * On Physiological Morphology, Woods Holl Biological Lectures, II, 1893. 



