Vol. 7, 1921 
GENETICS: W. H. TALIAFERRO 
167 
which differ inter se by very minute differences. In some cases, as in 
that of rat 413, we are almost forced to believe that a given "wild" in- 
fection is actually a "pure line." A possible explanation of these facts 
suggests itself when we consider the mechanism of infection with T. 
lewisi. As stated above the rat receives the infection from the flea. 
As the infected flea is feeding it deposits its feces, which contain the in- 
fective forms of the trypanosomes, on the fur of the rat. The rat then 
becomes infected by licking up the moist feces containing the infective 
forms. In nature it is probable that a given rat receives its infection 
from a single flea. This flea deposits a number of infective forms. The 
chances are that most of these will be killed by drying and that only a few 
will actually start the infection in the rat. This being so, the variability 
of the infections simply reflect the number of infective forms of the tryp- 
anosomes that actually start growth in the rat. In fact it is conceivable 
that a number of the "wild" infections are actually "pure lines." 
At the present time the author is of the opinion that if we consider 
all the infections in nature collectively we can say that there are a large 
number of "pure lines" which differ among themselves but which (as 
is shown by work in the laboratory) are per se constant in size. Due 
to the peculiar way in which rats receive their infections, however, we 
find that any given infection consists of only a very few of these "pure 
lines." 
Conclusions. — At the present phase of the work the following conclu- 
sions may be drawn: While growing the same "pure line" in different 
rats may cause significant differences in the mean size, these differences 
are small. The differences in the mean are never greater when the "pure 
line" is grown in different species of rats than when it is grown in different 
individuals of the same species. Passage of the "pure line" from rat to 
rat is not followed by any significant changes in the coefficient of vari- 
ation. In marked contrast to this it is found that passage of the "pure 
line" through the flea is invariably followed by a significant increase in 
the coefficient of variation which is interpreted as showing that the "pure 
line" breaks up into heritably diverse lines following such passage. Al- 
though passage through the flea has such a marked effect on the varia- 
bility it has no significant effect on the mean size. "Wild" infections 
occurring in nature exhibit differences in their mean sizes which are greater 
than can be explained by the fact that they occur in different rats. The 
coefficients of variation of "wild" infections are, on the average, higher 
than for the "pure line." The lowest of these coefficients, however, is 
no greater than the coefficient of variation for the "pure line" and 
even the highest can be explained on the assumption that the 
"wild" infection consists of only a very few "pure lines." It 
seems probable that there occur, in nature as a whole, a large number of 
