EXTENSIONS 371 



and therefore whether Stentor studies are relevant to analysis of 

 the potentialities of cells in general. My opinion is already evident 

 from the fact that stentors have throughout this study been 

 referred to as cells. This follows if we define the cell as a packaged 

 nucleo-cytoplasmic duality capable in some degree of independent 

 life. We can allow that these ''packages" sometimes may have 

 ''holes" in them connecting to other packages (cell bridges), and 

 that the enclosed nuclear phase may consist of one or more nuclei 

 of one or two types. These units may be wholly free-living, like 

 Stentor. They may be autonomous but not free-living as in the 

 case of parasitic protozoa. Or as tissue cells they may be subject to 

 a system of intercellular reactions leading to the morphological 

 and functional wholeness of cellular organs and organisms. Even 

 in the latter case, the cell lives a double life, both dependent and 

 independent, as one of the authors of the Cell Theory, Schleiden, 

 remarked. A tissue cell is dependent on the organism for its 

 sustenance and participates in multicellular interaction and 

 organizing relations, yet its capacity for independent life is 

 abundantly demonstrated by culture outside the organism; just as 

 its self-dependence is shown by the fact that if the long process of 

 a nerve cell is separated from its nucleated cell body, neither 

 proximity to countless nucleated fellow cells nor being continually 

 bathed in blood plasma can save that nerve from disintegration 

 after its nucleo-cytoplasmic integrity has been violated. Indeed, 

 the study of somatic deletions in Drosophila has shown that the 

 absence of a single genetic locus, which may be tagged by its 

 correlation with yellow color, results in the independent death of 

 the cells which lack it, as if all the nucleus is needed all the time 

 just to maintain the hfe of the cell itself (Demerec, 1934). 



That larger organisms are comprised of multitudes of cells 

 would seem to imply the interaction between nucleus and cyto- 

 plasm is so intimate that no portion of the cytoplasm can be far 

 from an associated nucleus. The nuclear phase is not aggregated 

 into one "gland". Even in the neurone, in which the cytoplasm 

 may extend several feet from the major cell body with its nucleus, 

 specialized organelles — the neurofibrils — may have been 

 developed to allow nucleo-cytoplasmic interactions even at this 

 distance (Parker, 1929). 



All grades are found between complete independence of cells, 



