SEED STERILIZATION AND ITS EFFECT UPON INOCULATION. 5 



So innocuous is the hydrogen peroxid tliat it iiia_y be used without 

 appreciable injur}' on germinated seeds. For transplanting seed- 

 lings from one culture medium to another without transferring asso- 

 ciated bacteria it has a special vahie as a sterilizing agent. It should 

 be added that while hydrogen peroxid is an excellent bactericide it 

 seems to have no more eti'ect upon certain fungi, notably the Penicil- 

 liums, than it has upon the higher plants. 



The leguminous seeds having hard coats withstood the toxic effect 

 of formaldehyde and mercuric chlorid much better than the nonlegu- 

 minous seeds, most of which were killed by the solution used. The 

 failure to sterilize completely some seeds, e. g., pea and bean, by the 

 same treatment with which other seeds were rendered sterile indi- 

 cates either chance infections more resistant than those on the other 

 seeds or the possibility that sometimes the infection exists under the 

 seed coats. It has been authentically established in the case of bean 

 anthracnose and loose smut of wheat that the seed itself is entered by 

 the fungus. In many of our attempts to sterilize seed the conclusion 

 seems inevitable that molds and fungi occur under the seed coats 

 rather frequently in some lots of seed. "N^liether this is due to infec- 

 tion in the field or to improper storage has not been determined. 



RESIDUAL EFFECTS OF SEED STERILIZATION. 



To test the residual toxic action of disinfectants used on seed, some 

 experiments were made by inoculating the wash water from treated 

 seed. A preliminary experiment with pea, wheat, and radish seed 

 treated with mercuric chlorid (0.5 per cent) for thirty minutes 

 showed that the fourth wash water from the treated seed was de- 

 cidedly toxic. After three times rinsing the seed in test tubes with 

 sterile water, the next rinsing water was poured into a sterile tube 

 and inoculated from a vigorous culture of Bacillus suhtijis, tubes of 

 sterile tap water being inoculated at the same time. Platings on beef 

 agar were made two days later, and the plates after incubating for 

 three days failed to show any growth from the wash water. The 

 checks gave abundant colonies. It was evident that the mercury 

 washed from the seed after three rinsings was still sufficient to be 

 fatal to B. suhtilis inoculated into the fourth wash water. 



Similar tests were made in which wheat seed was treated with mer- 

 curic chlorid, formaldehyde, and hydrogen peroxid. The same meth- 

 ods were used, except that the pcroxid-treated seeds were rinsed only 

 once, the formaldehyde series twice, and the mercuric-chlorid series 

 three times. Thus, the second, third, and fourth wash waters, re- 

 spectively, were used in these tests. The results, showing the number 

 of colonies of Bacillus suhtilis surviving in the wash water, appear 

 in Table III. 



[Cir.GI.1 



