METHODS OF INVESTIGATING BACTERIA. 799 



colonies to make it doubtful whether or not they have 

 arisen from accidental impurities, air germs, &c. If at 

 least 10, or at most 500 colonies grow on a plate, it is 

 in most cases a useful plate ; if we only wish to count 

 the colonies, and not to investigate them further, the 

 number on the plate may be as great as 5,000. If in 

 all three plates the numbers are greater or less than 

 those given it is well to repeat the experiment. 



In investigating various questions it is of great im- Enumeration 

 portance that we should be able, by counting the colonies ^iSiies 

 which have grown on the plate, to estimate the number 

 of bacteria which are present in a mixture of organisms. 

 In order to do this we must add a measured quantity 

 of the material to the jelly, and afterwards count the 

 number of colonies accurately. This is most easily done, 

 when the colonies are numerous, by means of a glass 

 plate divided into squares. In such a case we only count 

 a few of the squares, strike an average from the figures 

 so obtained, and calculate the number in the whole plate. 



Similar plates can also be made with nutrient agar. Agar plates. 

 This material is first liquefied in the tubes by boiling, 

 and then cooled to 40 C. in a water bath ; the material 

 is then added to the liquid agar, and the mixture poured 

 out on the plates. We must employ agar, which be- 

 comes fluid at 40 C., but solid at 38 or 39 C. The 

 agar afterwards often expresses water, and as a result 

 the whole mass may subsequently glide over the glass 

 plate. It is therefore of importance to line the vessels 

 in which the agar plates are to be kept only with dry 

 filter paper; and thus the water as it is expelled from 

 the agar evaporates so quickly that it cannot collect 

 between the material and the glass plate. 



A modification of the plate cultivations which has Esmarch's 

 been made by Esrnarch* is very useful for many prac- method> 

 tical purposes. Instead of flat glass plates he employs 

 wide test tubes, and the surface of the glass plate is 

 replaced by the equally large internal wall of the test 

 tube; while the gelatine is still fluid, and after it has 

 been mixed with the material under investigation, the 



* Zeitschr.f. Ihjyienc, rol. i., Part 2. 



