412 G. S. WILSON 



it is not proposed to enter into here as the space required would 

 be greater than would be warranted by their relative importance. 

 The results arrived at may be set out in the following order: (1) 

 In making any dilution not less than four drops should be used, 

 as with a smaller number sufficient accuracy is difficult to obtain. 

 With four drops, however, or more, the error of dehvery does not 

 exceed 1.2 per cent. (2) The bacterial content of drops delivered 

 in succession appears to be uniform; no difference between indi- 

 vidual drops could be substantiated. (3) So long as the interior 

 of the pipettes is clean- -and all pipettes should be washed through 

 with alcohol and ether — only a comparatively small number of 

 bacteria remains adherent to the walls; this error of chnging does 

 not appear to be greater than 1 per cent. (4) In making serial 

 dilutions a separate pipette should be used for each emulsion. 

 Under these circumstances the deviation of the actual from the 

 calculated dilution was found to be not greater than 3 per cent. 



The employment of roll tubes for the estimation of the liable count 



Before entering on a complete description of the method em- 

 plo3^ed in the estimation of the viable count, it is thought ad\-is- 

 able to give a brief resume of some of the technical details which 

 had to be worked out in order to insure the greatest degree of 

 accuracy. In the first place Petri dishes have been discarded in 

 favor of roll tubes. These are prepared in the following manner: 

 Test tubes measuring 6 inches by f inch are selected; after 

 sterilization in the hot oven, about 2 cc, of nutrient agar is 

 placed in each. They are then autoclaved for twenty minutes 

 at 120°C. When required for use the agar is melted and allowed 

 to cool to 45°C., in a water bath; the inoculum is dehvered directly 

 into the tube, the contents of which are niked bj- gentle shaking. 

 The tube is then rolled between the fingers — as in the case of the 

 Esmarch roll tube — the agar being allowed to pass about halfway 

 up the tube. As a rule sohdification is complete in half to one 

 minute, and the tubes are then incubated in an inverted position 

 for three days at 37°C. At the end of that time, when they are 

 removed, it will be found that the agar is studded with colonies 

 varjdng from round to lenticular in shape, evenlj^ distributed 



