ON TECHNIQUE 49 



the results, which could only be explained as being due to a lack 

 of uniformity in the various tubes. This was most noticeable in 

 the early hours of growth. 



In all of the work standard beef extract agar and bouillon were 

 used as culture media, adjusted to pH 7. 



The cell counts and microscopic preparations for purposes of 

 measurement were made from the growth on the above mentioned 

 agar slant cultures suspended in sterile water. For the later stages 

 of growth, where the number of cells was large, generally one or two 

 tubes were removed from the incubator and the growth washed off 

 with water and used for preparations for counting and measuring. 

 Samples for counting were immediately killed by adding formalin. 

 In the early stages of growth, however, the number of cells is so 

 small that it is necessary to remove a number of tubes in order to 

 obtain a sufficient number of cells for reasonably accurate counting, 

 and particularly to obtain a sufficient number of cells for micro- 

 scopic preparation and measurement. 



The procedure followed was to remove a number of tubes and 

 add to the first 2 or 5 c.c. of water. The bacteria were then scraped 

 off into this water, which was poured into a second tube, and this 

 process repeated from tube to tube until enough bacteria had been 

 removed to give a faint but perceptible turbidity to the liquid. The 

 number of tubes used and the amount of water used was varied 

 according to the stage of growth and the number of bacteria actually 

 present in the cultures, the final calculations being made so that the 

 number per c.c. is expressed as though the growth from one tube 

 had been suspended in 2 or 5 c.c. of water, as the case may be. 



Of the various methods which offered themselves for obtaining 

 the cell counts necessary for plotting growth curves, one of two 

 microscopic methods was used in each case. There were various 

 reasons for this, the most important being that with some of the 

 organisms studied the organisms tend to clump or to form long 

 chains, and the degree of clumping and chain formation varies with 

 the age of the culture. Plate cultures, therefore, would introduce a 

 serious source of error, which can be more or less obviated by the 

 microscopic method. With Bacillus megatherium, which is relatively 

 large, counting was done by the use of the Helber counting chamber. 



