878 



SCIENCE 



[N. S. Vol. XXXVII. No. 962 



also to their prolonged existence. Prudden's 

 experiments (1887) with water suspensions of 

 a staphylococcus, in tubes greased to prevent 

 crystallization at the temperatures employed 

 (15-25° F.), led him to believe that at the 

 same temperature the destruction of bacteria, 

 due to cold, was greater when the water did 

 not freeze, than when it did. Park, however, 

 made similar experiments (1900) with B. 

 typhosus and found that at the same tempera- 

 ture the reduction was 30 per cent, less in 

 water remaining liquid for three days than 

 where the water was frozen for the same 

 length of time. Park further cites an experi- 

 ment upon the freezing of typhoid bacilli in 

 which 50 per cent, to 70 per cent, are killed 

 " at the time," not more than 10 per cent, sur- 

 viving after one week and 1 per cent, after 

 four weeks, while Sedgwick and Winslow, after 

 a careful review of the literature and many 

 experiments, came to the conclusion that there 

 is " during the first half hour of freezing a 

 heavy reduction . . . amounting to perhaps 

 50 per cent. After this brief period of sud- 

 den but uncertain ' reduction ' the destruc- 

 tion of the germs proceeds pretty regularly as 

 a function of the time." Prescott and Wins- 

 low in their " Water Bacteriology " remark 

 (p. 17) that " Temperature has a direct rela- 

 tion to bacterial life, and the number of para- 

 sitic bacteria at least may be quickly lessened 

 by the action of cold." These conclusions are 

 supported by the fact that ice, and especially 

 old ice, even when formed from polluted 

 sources, is very low in bacterial life. 



On the other hand, it has gradually become 

 known that various frozen foods, such as ice 

 cream, frozen meat and frozen milk, often con- 

 tain very large numbers of living bacteria, and 

 this, too, even when kept for a long time, so 

 that a serious contradiction seems here to ex- 

 ist between theory and fact. To this contra- 

 diction my attention was first drawn some two 

 years ago during bacteriological studies of 

 frozen eggs, and especially by the fact that 

 such eggs, even after an exposure of many 

 months to a temperature of 0° P., still con- 

 tained millions of living bacteria. Obviously 



it was no longer possible to hold that either 

 mere cold or time is in and of itself neces- 

 sarily destructive of bacterial life; and in the 

 hope of bringing theory more clearly into har- 

 mony with experience I have within the last 

 year made numerous experiments calculated 

 to throw further light upon the general be- 

 havior of bacteria at temperatures about the 

 freezing point of water. 



Thus far I have worked almost exclusively 

 with a single species, B. coli, which, as is well 

 known, thrives at various moderate tempera- 

 tures and especially at the blood heat. I 

 have employed chiefly a 24-hour agar growth 

 suspended in water, in physiological salt so- 

 lution, in various dilutions of fat-free milk, 

 in various mixtures of pure glycerine and 

 water, and in solutions of cane sugar and of 

 commercial glucose. In some cases freezing 

 was done directly in test tubes; in other cases 

 in an ice cream freezer with the formation of 

 an ice " mush " or magma. By the courtesy 

 of the Quincy Market Cold Storage and Ware- 

 house Company of Boston I have been able to 

 hold the suspensions thus frozen at tempera- 

 tures as low as zero F. for periods of from 

 four to eight months. The experiments are 

 still in progress and some of them may be ex- 

 tended over a term of years. 



The following is a brief summary of re- 

 sults : 



I. When B. coli are frozen in Boston tap 

 water (in test tubes) as solid ice, and held at 

 — 20° C, only a fraction of one per cent, of 

 the original number remain alive at the end 

 of five days. Storage of a few weeks results 

 in complete destruction of the bacteria. These 

 results confirm those of Sedgwick and Wins- 

 low. 



II. When B. coli are frozen in Boston tap 

 water not solidly, but as a water ice or sherbet 

 is frozen, and held in this condition at — 20° 

 C, a large percentage remain alive for many 

 months. 



III. When B. coli are frozen in milk, pure 

 and diluted to various degrees with water, the 

 death rate of B. coli increases with the dilu- 

 tion, the largest numbers surviving in the un- 



