BACTERIA IN AIR, EARTH, AND WATER. 391 



are prevented from running together again, for a time at any rate, by 

 the solid gelatinized medium. Where it is necessary to isolate and 

 obtain pure cultures of any special organisms, the method described as 

 used in separating any colony of organisms should be used (see page 155). 

 Salomonsen recommends a cooling apparatus made of an ordinary plate 

 on which a shallow glass dish with a ground rim rests. This is filled up 

 to the surface, but not to overflowing, with water and lumps of ice, a 

 plate of ground glass is fitted on to this, and then a sterilized glass cover 

 or bell jar is used to cover the glass plate ; the further procedure is then 

 much as above. The organisms may be counted as a whole, but it is more 

 convenient to use a plate of glass marked into squares, which is sup- 

 ported over the gelatine plate, a low power lens being used to define the 

 smaller colonies. The colonies in a number of squares in different positions 

 are counted, then the number of squares that cover the gelatine plates 

 is determined, and an average is obtained from which the whole number 

 of organisms may be reckoned. It must be remembered, however, that a 

 small proportion of the gelatine still remains in each tube, therefore it is 

 necessary when making the plate cultivation to spread the remaining 

 quantity of gelatine over the walls of the tube by keeping the tube rotating 

 in water until the gelatine is fixed in position ; the organisms left in the 

 tube also give rise to colonies. These should be counted and added to 

 those found on the plates. In place of plates von Esmarch recommends 

 the use of test tubes. The inoculation is made, exactly as above, into 

 wide test tubes containing from eight to ten cc. of nutrient gelatine, a 

 second tube may be inoculated from the first, and a third from the second 

 by means of the sterilized pipettes. The plug is then thrust well home 

 after being singed, and a tight-fitting indiarubber cap is placed over the 

 mouth of the test tube, or melted paraffin is run in to protect the wadding. 

 The mixture is then effected by rolling the tube rapidly between the 

 hands, keeping it in a vertical position. When the fluids are sufficiently 

 mixed the tube is placed into ice-cold water and, in a horizontal position, 

 is kept rotating on its longitudinal axis until the gelatine is "set" in a 

 thin layer all over the walls of the tube ; the tubes are then put aside 

 and kept under observation. Should the presence of a large quantity 

 of oxygen be necessary, the indiarubber cap may be removed and 

 the plug may be pierced at one or two points with needles that have been 

 carefully heated, these needles passing through both paraffin and gelatine. 

 These tube cultivations may be made at once and on the spot, are readily 

 carried about, and serve as control experiments even when plate cultivations 

 are made in the ordinary manner. In place of glass plates and large moist 

 chambers, glass capsules are sometimes used for making plate cultivations, 

 and Petri has devised a shallow glass tray with a lid which answers the purpose 

 admirably. The method of procedure is much the same as in the above 

 cases, but is much simplified from the fact that the gelatin* is allowed to 

 cool on the floor of the capsule, the space above serving as a moist chamber. 

 Before sterilization these boxes are wrapped in a sheet of the hard tissue 

 paper, they are then subjected to dry heat and are ready for use at anytime. 

 In working with gelatine there is the disadvantage that certain organisms 

 peptonizing it cause it to liquefy exceedingly rapidly ; it has the advantage 

 that it is exceedingly clear, is readily melted, and solidifies rapidly. Agar- 

 agar on the other hand is not liquefied by the peptonizing organisms, but it 

 is not nearly so clear as the gelatine, and requires a much higher tempera- 



