THE CULTIVATION OF MICRO-ORGANISMS 123 



together. One end is fixed, while the other is free, and when it 

 is heated it tends to bend toward the metal side, when it cools it 

 bends toward the rubber. The brass strip is 15 inches long, J 

 inch thick, and \ inch wide; the rubber strip is \ inch thick, \ 

 inch wide, and a little less than 15 inches long. In the diagram 

 the end is fixed at d and is free at b. When it is heated, the free 

 end travels away from the set-screw at 6; when it cools, it moves 

 toward the set-screw. Rogers also recommends a regulator 

 made of invar and brass instead of hard rubber and brass. Where 

 invar is used instead of the hard rubber the dimensions for the 

 two metals are the same as those given for the brass strip in the 

 hard-rubber-brass regulator just described. As is evident from 

 the description, the circuit controlling the magnet is closed when- 

 ever the free end of the regulator comes in contact with the set 

 screw at b. When this circuit is closed the magnet attracts the 

 armature, and the heating circuit is closed by the contact formed 

 at c between the armature and the set-screw. In the diagram 

 this point of contact is put to one side for the sake of clearness, 

 but as a matter of fact in the instrument in use, the set-screw is 

 above and between the ends of the horse-shoe magnet, and 

 comes in contact with the armature which is extended upward 

 in the shape of a tongue. From the description just given it will 

 be noted that the thermoregulator does not control the heating 

 directly, but indirectly through the electro-magnet. 



Certain precautions have been found necessary in practice 

 in order to obtain satisfactory results with this incubator. The 

 set-screw against which the armature strikes at c should be so 

 set that the armature does not come in contact with the magnet. 

 In the apparatus described above there is a space of about J inch 

 between the armature and the magnet when contact takes place 

 between the set-screw and the armature. If the set-screw does 

 not project far enough to prevent the armature from coming in 

 contact with the magnet, the armature may adhere to the 

 magnet even after the current is broken at b, and when this is 

 the case of course the lamp remains lighted, and the temperature 



