50 MORPHOLOGIC VARIATION 



With this organism all of the cell counts were made with the same 

 counting chamber and cover glass, so that whatever absolute error 

 there may have been, there is no error in the course of the growth 

 curves due to this technique. Wilson has concluded that when the 

 number of cells is lower than fifty million per ex. the error of the 

 counting chamber method becomes too great to permit accurate 

 results, but it is evident that this depends entirely on the number 

 of counts made and averaged for the final results, i.e., on the total 

 number of cells counted regardless of their concentration. With the 

 agar slant cultures the cells were actually concentrated during the 

 early hours of growth until the liquid showed some turbidity, which 

 generally indicates about a million cells per c.c, so that the count 

 is the average of a number of tubes. This method involves a certain 

 amount of error in that not all of the cells are washed off in the 

 collecting process. With the broth cultures no such concentration 

 was possible, and the counts during the early hours of growth had 

 to be made by patiently going over one preparation after another 

 until a sufficient number of cells was counted; in general a thousand 

 cells were chosen as the limit, but in the very early hours of the 

 lightly seeded broth cultures this was quite impractical, and the 

 counts here are admittedly not of a high degree of accuracy. 



With the other organisms, however, because of their small size 

 and their tendency to remain floating in the suspension, the counting 

 chamber method was too difficult and too prone to error to be 

 useful. For all of these other organisms, therefore, I have used a 

 direct microscopic counting method devised by myself, but in prin- 

 ciple essentially the same as that introduced by Breed and Brew 

 for counting bacteria in milk. The suspensions of organisms were 

 mixed with an equal volume of a 2 per cent solution of Congo red, 

 and .01 c.c. of this mixture was spread as uniformly as possible over 

 an area of four square centimeters on clean glass slides. During 

 later phases of growth, where the number of organisms is larger, 

 eight square centimeters were sometimes used. The Congo red 

 preparations, after drying, were then placed a moment in a solution 

 of 1 per cent hydrochloric acid in alcohol, which serves to fix the 

 film and also to change the Congo red to a deep blue. The count- 

 ing of the organisms is then accomplished by counting the number 



