278 R. A. HARPER 
Darwin thought it legitimate to postulate particles, the "gemmules," 
which carry such characters. Spencer thought all functions of the or- 
ganism must be represented by physiological units" of the protoplasm. 
Haeckel explained assimilation as the "perigenesis of plastidules." Weis- 
mann brought this method of attack to its climax in his complex series of 
assumed biophores, determinants, ids, and idants; though perhaps Heiden- 
hain's graded series of life units, including super- as well as subcellular grades, 
has outdone that of Weismann. The cytologist has been unable to dis- 
cover in the viscous semi-liquid, semi-solid colloidal mass of the cell any 
adequate evidence of the existence of such a particulate structure. 
What seems to me the most important advance in our knowledge of 
cell architecture has been in the direction of the recognition of localized 
spatially differentiated regions of the cell body in which certain processes 
occur. To illustrate this point we may take the case of our increasing 
knowledge of the elaioplasts. Bodies associated with the appearance of 
oil in the cells have long been known in certain tissues of Funkia and Orni- 
thogalum. Wakker's work on the elaioplasts of Vanilla vastly increased 
the clearness of our conceptions as to these structures. In the cells of the 
vanilla plant the elaioplasts appear as quite definitely bounded, specifically 
organized regions in the cytoplasm. They show a reticulated stroma with 
more or less included oil which with specific treatments may be made to 
exude upon their surfaces. But they are not sharply bounded, and they 
are quite irregular in outline. In general they are not so liquid or so dif- 
ferentiated from their surroundings as to round up, and here again we have 
the evidence as in the case of the chloroplasts that the cell elements may be 
sufficiently viscous to maintain irregular outlines by virtue of their semi- 
solid condition. Beer has been able to recognize and has figured elaioplasts 
in many tissues in the Compositae. 
Other scattering observations indicate that they are of widespread 
occurrence in connection with oil production in many families of plants, 
and yet no one would deny that fatty materials may appear in cells, espe- 
cially in the liverworts and fungi and in animals, in the entire absence of 
any specially differentiated region of the cytoplasm for their formation. 
We get thus, as I conceive, a clear-cut notion of the modern cytologist's 
conception of protoplasmic structure. Oil may be formed anywhere in the 
cell. If the conditions are favorable for its production in large amounts, 
this production is likely to become localized and to bring with it marked 
regional transformations, so that we can even speak of such a region as an 
elaioplast or organ of the cell. If such conditions are permanent the organ 
is permanent, and by virtue of the extreme non-diffusibility of colloids 
and their tendency to form surface tension membranes, may be perpetuated 
even through long periods of inactivity. 
Similarly the plastidsare regions of the cell body in which carbohydrates 
are deposited in solid or semi-solid form as starch grains, to be later redis- 
