90 ABSTRACTS 



microscopical counts, due to other factors than possible inaccuracy 

 in measuring the quantity of milk under examination, a better means 

 of testing is by comparing the weights of the amount of milk delivered. 

 Twenty samples, weighed as discharged by one pipette, had a variation 

 of 3 per cent, while under the same conditions the weights of the milk 

 discharged from a loop varied 51 per cent. 



Poorly formed tips and faulty calibration cause difficulty in securing 

 the amount of milk desired. The tip should be a truncated cone with 

 the flattened end a circle of about 2 mm. in diameter. Pipettes should 

 be calibrated so as to deliver (not contain) 0.01 cc. of milk (about 0.0103 

 grams). This may be done by weighing the amount of milk dis- 

 charged into the hollow of a clean, hollow ground microscope slide, 

 covered to prevent evaporation. A pipette should deliver 4 to 5 per 

 cent in excess of 0.01 cc. of mercury if this is used in calibrating, as 

 mercury does not adhere to the walls of the pipette as do water and 

 milk. 



Pipettes should be clean, but sterilization is unnecessary. Sterilizing 

 does not remove bacteria, and growth is stopped in quickly dried smears. 

 Smears from ten samples of milk made by individual, sterile pipettes 

 averaged 6150 bacteria per cubic centimeter. Smears made from the 

 same samples by individual, clean pipettes averaged 4200 per cubic 

 centimeter while those made from the same samples by one pipette 

 which was recontaminated between each delivery in a high count milk 

 and cleaned by rinsing in clean water, averaged 4600 per cubic centi- 

 meter. Smears made from the same samples by one pipette which was 

 recontaminated between each delivery in a high count milk and rinsed in 

 the sample of milk from which the next smear was to be made, averaged 

 338,900 per cubic centimeter while when smears were made in the 

 same way but with the pipette unrinsed the average was 17,723,000 

 per cubic centimeter. These results indicate that for quantitative 

 milk work, individual, sterile pipettes have no measurable advantage 

 over individual, clean pipettes and that neither have any advantage 

 over the use of one pipette for all samples, provided it is cleaned by 

 rinsing in clean water. Dirty pipettes or pipettes handled carelessly, 

 cause measurable errors. 



A Simple Apparatus for Isolating Anaerobes. Zae Northrup. 



A simple apparatus for isolating anaerobes consists in a 25 cc. burette 

 complete with rubber tubing, glass tip and pinchcock. This is filled 

 with any desired liquid nutrient medium, plugged with cotton, the de- 

 livery tube protected from contamination by inserting it in a small 

 test tube and the whole apparatus set up on a ring stand and sterilized. 



This tube may be inoculated with any material containing anaerobes. 

 In a comparatively short incubation period the various classes of micro- 

 organisms will adjust themselves to their optimum oxygen require- 

 ments which will be noted sooner or later by a clear zone at the upper 

 portion of the liquid varying in depth from a few millimeters to 8 or 

 9 cc. in some instances. 



