EXPERIMENTS WITH PRESERVATIVES. 23 



irregular, though the benzoate checked development in all. Equal 

 amounts of benzoate were used in tomato bouillon, with practically 

 the same results as in the ketchup. The development was checked 

 in all, and in some plates one-sixteenth of 1 per cent seemed to be 

 fully as efficacious as one- tenth of 1 per cent. When the mold was 

 examined under the microscope, the filaments were found to be much 

 swollen and distorted in shape, and filled with a coarsely granular 

 protoplasm, containing much fat, as indicated by the blackening with 

 osmic acid. The culture containing the mold which gave the least 

 development seemed to show the least distortion and swelling of the 

 filaments. 



The results indicated that in using sodium benzoate. as a preserva- 

 tive there is uncertainty as to results, even when using the maximum 

 amount allowed one-tenth of 1 per cent. They also indicated that 

 this preservative had an injurious effect on the living matter of the 

 mold. (See PL II; compare with normal growth, PL I.) 



SALT. 



The effect of salt in checking development was tested by using 

 tomato bouillon as a medium and adding 5, 10, 15, 20, 25, and 30 

 grams of salt, respectively, to 100 cc. These were inoculated with the 

 mold. The 5-gram solution seemed to have no effect on development. 

 When 10 grams were used growth appeared as soon as in the bouillon 

 without salt, but was not so extensive. In the 15-gram solution 

 growth was retarded four days, and most of that which did develop 

 remained submerged, the mold growing normally on the surface. 

 With 20 grams the growth was five days slower than the normal in 

 starting, and after that there was only a slight development. In the 

 25-gram solution, the growth started at the same time as when 20 

 grams were employed, but remained stationary, while with the 30- 

 gram solution, no development occurred. 



The yeast was checked slightly by 5 grams, and very materially by 

 the 10-gram solution, as it required two days for a thin, delicate film 

 to form, whereas in ordinary solutions a rather thick film is formed 

 within twenty-four hours or even in less time. There was no develop- 

 ment in the 15-gram solution. 



SUGAR. 



The effect of sugar was tested on both the mold and the yeast by 

 adding it to tomato bouillon. It was supposed that a low percentage 

 of sugar like the salt would plasmolyze the cells, and in this way check 

 growth, but it seemed to have no effect until the amount was increased 

 to 25 grams per 100 cc of bouillon. In this solution growth appeared 

 as soon as with the weaker solutions, but there was a smaller amount. 

 In the 25 to 40 gram solutions there was less development as the 



