10 



enclosed pouring apparatus. The finished plates are then placed 

 in layers, one on top the other, on glass 'benches' under large 

 bell jars. It is remarkable what can be accomplished in many 

 cases with only limited facilities, especially without the pouring 

 apparatus. Since the beginning of the school year, I have 

 been using flat, double shells of about 10-11 cm. in diameter, 

 1-1.5 cm. high. The upper shell ser\-ed as a cover and had 

 a somewhat larger diameter. Into these dr^^-sterilized and 

 and cooled shells I poured the liquid nutrient gelatin with 

 its inoculated material. As the over lying shell is only slightly 

 raised, and the gelatin is protected by the cover, it is not 

 likely that there will be contamination: for example, the 

 germs in the air could hardly enter. The spout is flamed in 

 the usual way and cooled. The gelatin hardens very quickly 

 when poured and forms a layer one millimeter in thickness, 

 which is preserved for a considerable time, as it is protected 

 by the upper cover shell. 



"In experimenting with soil material, sand, dirt, and similar 

 substances, it is an advantage to pour the material into the 

 shell with liquid gelatin. One gradually acquires considerable 

 skill in doing this. By using short, jerky motions of the shell, 

 an even distribution of the material is obtained. If the fore- 

 going instructions are followed every particle of the poured- 

 out gelatin may be examined with an ordinary microscope. 

 Only by excessive action will the layer along the rim become 

 uneven. The gelatin dries very slowly in these shells. It may 

 be kept moist longer by putting several shells (5-6) on top 

 of each other in a larger shell on a layer of moist filter paper, 

 covered with a bell jar — mouse jar, battery jar. Such flat 

 shells are especially suitable for agar-agar plates, inasmuch 

 as agar is hard to fasten on simple plates without special 

 means. Moreover, the counting of grown colonies is simplified. 

 After removing the upper cover, a glass plate, on which the 

 usual division into square centimeters and their divisions, is 

 placed underneath. The counting lens is set up, and the count 

 is made on a black background. The surface area of the shell 

 is then calculated from its diameter." 



It does not appear from Dr. Petri's own account whether 

 the double glass "shell" which is recommended was a new de- 

 vice or shape specially prepared according to his specifications. 



