304. A. FRANKLIN SHULL 
is determined at, or prior to, an early egg stage; or if this determi- 
nation is a gradual process, it has proceeded so far before the early 
egg stages that manure solution is unable to reverse it. 
Experiment 6. In this experiment, a line was reared in spring 
water, and as in the preceding experiment two sister females were 
reserved from each generation for the purpose of breeding. One 
female was kept in spring water all her life, and all her offspring 
were reared to maturity in spring water. The other female was 
kept in spring water until she had laid 13 to 18 eggs, and was 
then transferred to filtered, undiluted manure solution. She was 
transferred to fresh manure solution every twelve hours thereafter, 
to prevent accumulation of bacteria, so that all of her eggs after 
the thirteenth to eighteenth were laid in manure solution. All 
the eggs laid in spring water were hatched, and the young reared 
to maturity, in spring water. All the eggs laid in manure solu- 
tion were hatched, and the young reared to maturity, in manure 
solution. 
That the manure solution used in this experiment was strong 
enough to exclude male-producers was shown by rearing in it 
five successive generations of all female-producers, as described 
in the preceding experiment. Further proof of its effectiveness 
is found in the additional control presently to be described. 
The details of this part of the experiment are given in table 8, 
where the rotifers of the early part of the family, that were 
reared in spring water, are given to the left of the vertical line, 
those reared in manure solution to the right of that line. The 
offspring are recorded in the order in which they hatched, this 
order being determined fairly accurately, I believe, by the rela- 
tive sizes of the young rotifers when they were isolated. Only 
one male-producer hatched from an egg laid in manure solution, 
and that came from the very first egg laid after the transfer of the 
female to the manure solution. 
The control of this experiment, that is, those families bred 
entirely in spring water, is given in table 9. Each family was 
reared from a sister of the parent of the corresponding family 
in table 8. These control families are divided by the vertical 
