JUNE 14, 1912] 
mal, mesophyl, vessels, etce., and the deter- 
mination of the qualities of the organisms 
as wholes which depend upon the interac- 
tion of these cells and which only indi- 
rectly represent and are represented by 
the organization of the cell and germ 
plasm. A leaf can not be represented di- 
rectly in a cell, but the color of the leaf 
may be represented in the color of the cell, 
and its size may be determined by the ca- 
pacity of the cell to divide and grow. 
Such Mendelian unit characters as color, 
length of life, ete., are properties of indi- 
vidual cells and agree with Detto’s con- 
ception of metidentical characters which 
become by multiplication of the cells more 
or less diffuse properties of plants as 
wholes or organs as wholes. 
It is easy to distinguish the heredity of 
the cell form as such from the heredity of 
the form of the many-celled colony. No 
one thinks now of asserting that the organ- 
ization of the cell is identical with that of 
the many-celled individual. That dogma 
of the old preformationists disappeared 
with the improvement of the microscope. 
If we can, however, fix clearly in mind 
that such representation of the adult or- 
ganism as is present in the egg in no way 
resembles in space configuration or in 
complexity the arrangement of orgams and 
tissues in the adult animal or plant, we can 
attack the problem of form development 
as it really exists, and free from many en- 
cumbering traditions of preformation and 
epigenesis. 
There can be no doubt that the cell and 
nucleus have a highly complex mechanical 
organization. It is a commonplace of his- 
tology that cell and tissue structures are 
relatively constant through genera and 
families—regardless of variation in the 
size and form of organs and of the plant 
as a whole. Cell size is also relatively con- 
stant through genera and even families. 
SCIENCE 
923 
Quantitative variations in the size of 
leaves, seeds, ete., are due to the number 
of cells they contain, and this, of course, 
depends on the number of times the cells 
have reproduced themselves by division. 
In some cases Mendelian characters can 
thus be identified with qualities of the 
cells. 
The attempt from the standpoint of 
Mendelian conceptions of dominance and 
segregation to analyze the behavior of the 
generalized qualities of plants and their 
parts in development and heredity has 
been stimulating to research in a high de- 
gree, but the attempt to express the re- 
sults of such analyses in terms of unit 
characters may be found to be only a relic 
of the preconceptions of the earlier cor- 
puscular and preformational theories of 
heredity. Many Mendelians are inclined 
to think of their germ plasm as merely 
chemical in its essential constitution and I 
have referred above to the uncertainty of 
the evidence as to any ultramicroscopic 
organization of the nuclear chromatin of 
the cell. With the clarification of our con- 
ceptions in the domain of colloid chemistry 
we may hope to gain new viewpoints which 
will be more serviceable in the interpreta- 
tion of biological facts than the conceptions 
of atoms and molecules which have so far 
dominated the corpuscular theories of 
protoplasmic structure. But we must also 
expect, perhaps, that the real distinctions 
between the organization of protoplasm 
with its long history of slow evolution and 
the in vitro aggregates of the chemist will 
be emphasized rather than obliterated. 
R. A. Harper 
CoLUMBIA UNIVERSITY 
SCIENTIFIC NOTES AND NEWS 
Dr. Franz Boas, professor of anthropology 
at Columbia University, has been given the 
doctorate of science by Oxford University. 
