JUNE 14, 1912] 
space relations between elements of the 
germ plasm are quite unnecessary in the 
heredity of such qualities. 
With the disappearance of Weismann’s 
conception of definite and complex space 
relations between the elements of the germ 
plasm, we perhaps see the last of the old 
doctrine of formal preformation. And yet 
it is a curious fact that one of the most im- 
portant influences which Mendelian stud- 
ies have exerted on our conceptions of the 
germ plasm seems to be in the stimulus 
which they have apparently given to all 
corpuscular theories of heredity. The doc- 
trine of unit characters is the real gist of 
Mendelism, and it seems obvious to associ- 
ate these unit characters of the plant, as a 
whole, with the theoretically postulated 
determinants, granules, ete., of the cor- 
puscular theories of heredity. 
The relative popularity of De Vries’s 
conception of the pangens is due to the em- 
phasis he lays on the conception of units 
representing diffuse characters of the plant 
as a whole rather than organisms or tis- 
sues. De Vries’s pangens are assumed in 
many cases to represent just such qualities 
and characteristics of color, size, etc., in 
the plant as are found to show the Men- 
delian behavior most perfectly. It is cer- 
tainly a striking fact that Mendel and De 
Vries independently reached the concep- 
tion of the importance of such characters 
in an analysis of heredity. That De Vries 
should regard them as units is due to the 
influence of his corpuscular theory of the 
germ plasm. They are more properly de- 
scribed as diffuse characteristics of the 
plant or its organs, as wholes, as De Vries 
has so strongly emphasized in relating them 
to the origin of his mutants which differ 
from their parents in general features af- 
fecting the whole organism. 
It must be remembered also, that while 
De Vries makes his pangens stand for 
SCIENCE 
917 
characteristic features of the plant or its 
organs as wholes rather than for specific 
organs or tissues, and rejects Weismann’s 
and Naegeli’s conception of a mosaic combi- 
nation of the units in a germ plasm, yet he 
is not free from the feeling that a definite 
spatial relation of the pangens in the germ 
plasm is necessary. The pangens must be 
in smaller and larger groups and these 
groups so arranged that the members of a 
group may become active at the same time 
at least. The arrangement must also be 
such as to provide for their proper distri- 
bution at each cell division. This sounds 
as if a relatively simple serial arrangement 
were all that is necessary, but as noted, the 
coupling of pangens representing all the 
secondary sex and sex-limited characters 
affecting widely distributed organs of the 
body, which in many respects are other- 
wise determined, with a single sex chromo- 
some or group of pangens, is not a simple 
matter. The doctrine of intracellular 
pangenesis, with its storage of pangens in 
the nucleus and their migration into the 
cytoplasm, provides for the behavior of the 
Mendelian factors hardly better than the 
formal unfolding of the architecturally pre- 
arranged corpuscles of Weismann’s theory. 
That the harmony in underlying assump- 
tions between the Mendelians and the 
adherents of corpuscular theories of hered- 
ity is only apparent seems to me the inevi- 
table conclusion of any careful analysis. A 
list of the characters associated by De 
Vries with his pangens is itself suggestive 
of difficulties. The first character he men- 
tions in his intracellular pangenesis is the 
green color of plants. That this is a char- 
acteristic of plants which have a certain 
unity of behavior in heredity may be true, 
but to see how it can be represented by a 
pangen granule in the germ plasm so as to 
appear in just the proper tissues in just 
the proper degree is not so obvious. Other 
