Raymond Pearl 
261 
immer die kleinsteu Individueu welche conjugiren." Hertvvig* found the same 
thing in his cultures, namely that conjugants were below average sizef. 
While there would appear then to be general agreement that the conjugating 
individuals are usually smaller in size than the average for the general population 
of non-conjugants, it has not hitherto been noted, so far as I know, that con- 
jugants are also differentiated from the general population in variability and 
correlation. This is of course only natural since nothing but measurements of 
considerable numbers of individuals will give an appreciation of variation and 
correlation in size dimensions in so small an organism as Paramecium. The fact 
that conjugant individuals are markedly differentiated in variability from the 
general population if generally true is unquestionably of considerable importance 
in connection with the whole problem of repi'oduction and variation in the 
Infusoria, however the differentiation may be produced. The decrease in size 
preceding conjugation has usually been explained as due to a brief period at 
this time of unusually rapid division in which there is little time for growth 
between succeeding fissions. Calkins if has shown however that this period nf 
rapid division does not always precede conjugation. Consequently the reduced 
size must, in some cases at least, be due to some other factor. In his most recent 
paper on the subject, this investigator§ has brought out some very interesting 
facts regarding the protoplasmic condition of Paramecium in periods of depression, 
showing that with the reduced size usually observed at such times there are 
associated characteristic changes in cytoplasmic and nuclear structures. His re- 
sults, for the details of which the original paper must be consulted, certainly 
point to the conclusion that the reduction of size (and aL«o probably the changes 
in variabilities and correlations) observed in conjugants depend upon fundamental 
and deep-seated changes in the physiological condition of the organism connected 
with reproduction. It seems to me that this conclusion is much more probable 
than that we have to do with any direct selection in the ordinary sense. It 
is quite clear that there is no immediate elimination of those members of the 
culture which do not conform to the conjugant type. The possibilities in the way 
of change of variability in Paramecium without selection, but instead as a result 
of a direct general bodily rearrangement, or "Umwandlung," are known to be 
great||, and it seems to me altogether likely that in the differentiation of con- 
jugants from non-conjugants we are dealing with a case of this kind. 
There is a point which might be raised in objection to the view that the 
differentiation of the conjugant population is to be thrown back on deep-seated 
physiological causes. It might conceivably be maintained that the reduced varia- 
bility of conjugants which has been found is a result of the conjugation process 
* AhhnncU. der k. buyer. Akad. der ]nsse)iscli. ii. CI. Bd. xvii. pp. 153 — 233. 
t Note added Nov. 10. Mr Lister (loc. cit.) accuses me of being ignorant of the literature on 
this point. I leave it to the reader to judge what foundation exists for this accusation. K. P. 
X Loc. cit. p. 158. 
§ Calkins, G. M. : Journ. of Exper. Zool. Vol. i. pp. 423—461. 
II Cf. Pearl, E., and Dunbar, F. J. : Seventh Report, Michigan Academy of Science, pp. 77 — 86. 
Biometrika v 34 
