544 
Journal of Agricultural Research 
Vol. V, No. 13 
The sweet potatoes used in the first series of experiments were dug 
on September 30, thoroughly washed, and were kept covered in the 
laboratory until the following day. In the further manipulation each 
potato was split lengthwise into two parts as nearly equal as possible. 
So far as could be determined the potatoes were cut longitudinally in a 
dorsiventral plane. One half, marked “a,” was ground immediately, 
and samples were taken from the mash for the determination of mois¬ 
ture, sugar, and starch. The other half, marked “b,” was stored. Six 
halves were stored at eaph temperature, the corresponding halves hav¬ 
ing been grated and sampled as described. The operation of preparing 
and sampling the halves for a set of experiments at the three tempera¬ 
tures required two days. Simultaneously with the halves a number of 
whole sweet potatoes were put into the constant-temperature chambers 
in which the experiments were conducted. At the end of 12 days the 
stored halves were taken out, grated, and sampled. After the completion 
of this operation, which required two days, the stored whole potatoes (which 
during this time had been subjected to the same conditions as the stored 
halves) were split lengthwise, like the first set, and one half was prepared 
for analysis. The other half was stored for another period of 12 days, 
after which it also was grated and analyzed. It will thus be noted that 
the difference in composition of the two halves of the first set of roots 
showed the change during the first period of 12 days immediately after 
the potatoes had been dug, while the difference in composition of the two 
halves of the second lot showed the change for a second period of 12 
days immediately following the first period. 
Although the time during which the sweet potatoes were exposed to 
the experimental conditions was essentially the same for the comparable 
lots, some unimportant differences necessarily crept in. Thus, for 
instance, it was impossible to prepare a complete set in a single day; 
therefore, one half of the potatoes used in the experiment were prepared 
one day and the other half the following day. Consequently, the one 
lot remained in the laboratory about a day longer than the other. Also, 
although the different groups were taken out of their respective chambers 
in the same order in which they were put in, no attention was paid to the 
order in which the individual potatoes were removed, since it was neces¬ 
sary to work as rapidly as possible. On this account it is likely that 
some halves remained in the chambers a few hours longer and others a 
few hours less than the assigned period, but it is obvious, considering the 
slowness of the changes that take place, that these discrepancies can have 
no effect on the general result. 
The whole sweet potatoes stored simultaneously with the first set of 
halves also remained in the chambers two days longer than the halves, 
on account of the time required to grind and sample the stored halves; 
but this also is of no consequence, since the object of the experiments was 
