414 G. S. WILSON 



have been repeated from day to day, the acquirement of a stock 

 of 200 or 300 plates would have been necessitated, had these been 

 employed. (2) There is considerably less risk of contamination 

 in the case of tubes. Absolute sterility is obtained by autoclav- 

 ing the agar. Contamination, in fact, does not occur. In deal- 

 ing with several thousand tubes during the course of this work, 

 only one single case of contamination has been encountered. 

 With plates, on the contrary, the risk of contamination is by no 

 means negligible, especially if incubation has to be continued for 

 three days. (3) Less media is required in the case of tubes than 

 with plates; for the former 2 cc. are sufficient, for the latter — if a 

 5-inch plate be employed — 16 cc. or so must be prepared. (4) 

 Tubes are easier to count than plates. They possess no corners; 

 one is dealing with a relatively flat surface throughout, whereas 

 with plates there is frequently an undesirable amount of spread- 

 ing of the colonies at the junction of the bottom with the sides 

 which renders counting a difficult and uncertain procedure. (5) 

 Tubes can be incubated immediately after rolling, instead of 

 having to be kept immobile for twenty minutes till the agar is 

 set, as in the case of plates. Where time is a consideration the 

 practical value of this point will be appreciated. 



After finding that tubes were perfectly well suited for the pur- 

 pose of performing the viable count, it was necessary to compare 

 the results obtained with those obtained by means of plates. For 

 this purpose two distinct sets of experiments were undertaken, 

 differing in the way in which the plates were poured. In the 

 first series viable counts were made on various broth cultures 

 employing both tubes and plates. The tubes were put up and 

 rolled in the way already described. In preparing the plates the 

 emulsion to be counted was delivered into a test tube containing 

 15 cc. of melted agar which was then poured into a plate and 

 allowed to soUdify. Each was incubated for three days before 

 counting. In all cases the conditions were strictly comparable. 

 During the course of several exijeriments of this nature 69 tubes 

 and 69 plates were examined. The total number of colonies in 

 each set were added together : the results were as follows : 



