274 A Biometrical Stiidtj of Conjugation in Paramecium 
of assorting in conjugation is the correct one, then the relative position of the 
mouth is one directly selected character. But it by no means follows that it is the 
only one. There may be other characters selected to just as close a degree in the 
conjugating. For the present it is sufficient to have shown that in a species living 
under natural conditions, a high degree of homogamy is not only possible, but 
actually exists. 
Finally it should be pointed out that the fact that we find such a high degree 
of homogamy in a protozoan form like Paramecium strongly suggests the possi- 
bility that in higher organisms there may be assortative mating of the gametes in 
the process of fertilization. Should such a homogamy of the gametes occur 
it would probably be of far greater importance than any assortative mating 
of somas. 
VIII. SiiiiDiiari/ of Results. 
The chief results of this study of variation and correlation in Paraineciuin 
caiidutioa in connection with the process of conjugation may be summarily stated 
as follows : 
1. Biometrical analysis of the variation in a considerable number of individuals 
shows that Paramecium follows the same general laws which have been found 
to hold for continuous variation in higher forms. The coefficients for variation 
in length of Paramecium cluster very closely about a value of 8 per cent. 
Certain of the frequency distributions approach reasonably well to the normal 
or Gaussian curve, but this is by no means universally the case. 
2. In following a single culture throughout the history of a conjugation 
epidemic, it was found that definite changes are produced in the population as 
a result of the action of the environment. These environmental effects, however, 
were more pronounced in the case of non-conjugants than in conjugants. 
3. There is no evidence that conjugation tends to produce increased vari- 
ability in ex-conjugants. All the evidence indicates, on the contrary, that 
conjugation serves to restrict the variability induced by environmental influences 
or, in other words, to preserve relative stability of type. 
4. Conjugant Paramecia are distinctly and markedly differentiated from 
the non-conjugant population living in the culture at the same time, in type, 
variability and organic correlation in respect to all characters studied. This 
differentiation includes not only the absolute dimensions of the body, but also 
the shape, as measured by the length-breadth index. This differentiation of 
conjugants from non-conjugants is in no sense insignificant in amount, nor is 
it confined merely to the means of the different characters. On the contrary, it 
is large in amount, and just as significant for the variabilities and correlations 
as for the means. 
5. There is a tendency for like to pair with like (homogamy) in respect to 
length of body in the conjugation of Paramecium of a closer degree than has 
hitherto been found for assortative mating in any organism. The homogamic 
