482 



SCIENCE. 



[N. S. Vol. XXI. No. 535. 



and pounded ice, the time of exposure be- 

 ing 2 hours. The freezings were made in 

 5 c.e. portions of bouillon in test-tubes of 

 resistant glass. The thawings were made 

 in tap water at 16° to 18° C. The inocu- 

 lations for each set of plates were made in 

 the same way, i. e., usually with the thin- 

 nest meniscus it was possible to obtain 

 across a 1-mm. platinum oese. The petri 

 dishes were carefully selected, those taken 

 being approximately 9 cm. in diameter, with 

 flat bottoms. The regular method of work 

 was to make three poured plates (cheeks) 

 from the inoculated tubes after insuring 

 thorough ditSusion, which was obtained by 

 stirring Avith the platinum rod, shaking 

 and allowing to stand one half hour. The 

 tube was then immediately lowered into 

 the liquid air and frozen slowly from the 

 bottom up to avoid cracking. (This us- 

 ually required four minutes.) As soon as 

 the one half hour or other predetermined 

 time of exposure had elapsed, the tube was 

 removed, warmed for about 3 minutes in 

 the laboratory air and then thawed in 

 water (which usually required another 5 

 minutes). As soon as the thawing was 

 completed, three more poured plates were 

 made, and these together with the three 

 check plates were then incubated in the 

 dark at 30° C, until the colonies were in 

 good condition for counting— a period 

 varying, according to the species, from one 

 to several days. The plates were all put 

 on a leveling apparatus as soon as poured, 

 and in general the distribution of the 

 colonies in the nutrient agar was very uni- 

 form. When the plates were sown thin 

 enough, the entire surface was counted (60 

 sq. cm.) ; for the thicker sowings the aver- 

 age of 10 or 12 sq. cm. was used, or of one 

 half the plate. The following samples 

 from two of the thirty or more slides ex- 

 hibited will give a general idea of the 

 method and results: 



BACILLUS TYPHOSUS. (SIX POURED PLATES.) 



Colonies per Square Centimeter ; 



After Freezing 

 Before Freezing. (2 hrs. in Salt 



and Ice). 



Plate 1 46 2/60 



Plate II 39 



Plate III 42 2/60 



Average 42 1/45 



Per cent, killed, 99.5. 

 Frozen, Dec. 15. Incubated at 30° C. Counted, 

 Dee. 19. 



BACILLUS TYPHOSUS. (SIX POUBED PLATES.) 



Check on Salt and Ice, 1 e., 1/2 of Same Culture. 

 Colonies per Square Centimeter : 



Alter Freezing 

 Before Freezing. (2 1irs. in 



Liquid Air). 



Plate 1 50 1/60 



Plate II 51 3/60 



Plate III 43 2/60 



Average 48 2/60 



Per cent, killed, 99.3. 

 Frozen, Dec. 15. Incubated at 30° C. Counted, 

 Dec. 19. 



The following conclusions may be 

 drawn : (1) The effect of very low tempera- 

 tures has been greatly overestimated. As 

 destructive results were obtained with salt 

 and pounded ice (—17°. 8 C, or less) as 

 with liquid air. (2) The critical point ap- 

 pears to be somewhere around 0° C. If an 

 organism can pass this point in safety, it 

 is believed that even absolute zero (—273° 

 C.) would not harm it. (3) Some indi- 

 viduals of each culture were able to endure 

 unharmed the temperature of liquid air 

 (—190° C), although this was often only 

 a small proportion of the whole number. 

 (4) Repeated freezings and thawings re- 

 duced this number very gradually to noth- 

 ing, but ten freezings and thawing (in 

 course of eight hours) did not kill all of 

 the individuals of P. campestris, although 

 it reduced the number in the bouillon to 

 such an extent that three one-millimeter 

 loops gave three sterile plates. (5) This 

 resistance to freezing is believed to be due 



