CALCIUM HYPOCHLORITE AS A SEED STERILIZER 423 
is exceedingly long or a noticeable amount is carried over with the 
seed. In making the transfers, however, an effort was made to 
leave behind as much of the germicide as possible. That the quantity 
of disinfecting material carried over with the seed into the bouillon 
was not sufficient to act as a germicide was proven by placing along 
with some of the treated seed in the bouillon an untreated seed. In 
every case within three days there was ample bacterial growth. 
With large seeds such as corn, beans, peas, and wheat, only those 
seeds were used which appeared normal and seemed capable of pro- 
ducing vigorous plants. With other seed, such as timothy, alfalfa, 
clover, etc., no such selection was made. The number of seeds placed 
in each series of six to ten tubes varied from one to three with corn 
and from four to one hundred with other seeds. The number, how- 
ever, varied with the size of the seed and the amount of available agar 
surface in the test tube. In several cases as many as 4,000 timothy 
seeds were treated at a time and transferred to a single container. 
The following table presents some of the data. 
Time Required for Time Required to Cause 
Sterilizing Seed. Injury to Seed. 
Seed 
Hours 
Hours 
Zea Mays L 
8 
18 
6 
18 
Triticum vulgar e Vill 
15 
20-22 
Phleum pratense L.^ 
8 
above 22 
4 
" II 
9 
" II 
Petroselinum sativum Hoffm.^ 
2 : 
15 
" II 
I 
" II 
5 
" 10 
Brassica oleracea L.^ , 
. 2 
" II 
Trifolium pratense L.^ 
24 
" 24 
Linum usitatissimum L.^ 
10 
" 14 
Fagopyrum esculentum Moench.^ 
No success at all 
" 40 
7 : 
30 
" 15 
Solanum tuberosum L.^ No success at all 
2 Seeds hulled before treatment. 
3 Tested only on peptone agar and at room temperature. 
^ Cuttings of tuber. 
The data show that the time required in most cases for sterilizing 
seed and the exposure necessary to produce injury are many hours 
apart. This allows considerable latitude with respect to the time 
during which the disinfectant may be allowed to act. It is evident 
