JUNE 14, 1912] 
them. Still there can be no question that 
the doctrine of segregation and alternative 
inheritance no longer has the clear and 
simple form in which Mendel proposed it. 
It is equally certain that the changes which 
have been made were necessary in order to 
bring the theory into harmony with the 
vast mass of new facts which breeding on 
a large scale has brought to light. 
These modifications have largely been 
in the direction of providing for a greater 
degree of variability in the F, and succeed- 
ing generations than is provided for by the 
doctrine of segregation. Pre-Mendelian 
views held to a general breaking-up and 
tendency to vary in hybrids after the first 
generation. It was Mendel’s great contri- 
bution to apparently discover in this mul- 
tiplicity of forms a real simplicity in fun- 
damental features. Recent workers have, 
however, found it increasingly difficult to 
interpret their results within the limits of 
the Mendelian formule. 
One of the first steps in the modification 
of Mendel’s views was to admit greater 
variation in the first generation. It is gen- 
erally agreed now that there is no law of 
dominance, that the first generation may 
be intermediate, or a mosaic of the parental 
characters, and the possibility is present 
that there may even be di- or poly-mor- 
phism in the first generation. Practically 
all the pre-Mendelian conceptions as to the 
character of the first generation have now 
been confirmed and generally admitted. 
An important step toward the modifica- 
tion of the Mendelian account of segrega- 
tion in the second and following genera- 
tions was in the introduction of the pres- 
ence and absence hypothesis as a substitute 
for the Mendelian doctrine of pairs of posi- 
tive visible characters. The presence and 
absence hypothesis when applied to simple 
Mendelian pairs of contrasting characters 
seems perhaps to have a sort of explana- 
SCIENCE 
919 
tory value. To be sure, the case is some- 
what as if a geologist should explain a 
given hill as due to the absence of a moun- 
tain, failing to suggest anything as to the 
forces and conditions concerned in the 
production of both hills and mountains. 
There can be no question that Mendel’s 
doctrine of segregation assumed the 
probable occurrence of alternative inherit- 
ance for all homologous differential quali- 
ties in the parents of hybrids. It was the 
task of the breeder to discover such pairs 
of opposite characters and operate with 
them. This has been found to be impos- 
sible in many cases, and the presence and 
absence hypothesis is advanced as an ex- 
pression more nearly in accord with the 
facts as found in practical breeding work. 
The facts that have necessitated this change 
show that not all apparently homologous 
contrasting characteristics form  allelo- 
morphic pairs. When two contrasting 
flower colors are found not to form an 
allelomorphic pair, the pairmg may some- 
times still be discovered by operating with 
eroups of characters in the formation of 
allelomorphs, such as pigmentation against 
white or albinism, and here again albinism 
is by no means always found to be the same 
thing in heredity. 
In other cases the presence and absence 
hypothesis is introduced as a concession to 
the fact that the second and following hy- 
brid generations are much more variable 
than the Mendelian doctrine allowed. On 
the Mendelian hypothesis of allelomorphic 
pairs and their segregation, two individ- 
uals differing in a single character, such as 
flower color, could produce but two kinds 
of gametes and four groups of offspring. 
On the presence and absence hypothesis we 
can assume at once four elements, a factor 
for the presence and absence of each of the 
two colors. To take the stock illustration: 
fowls with pea and rose combs when crossed 
