PHYSICS. 103 



other scale-pan is a vessel containing glycerin. This appa- 

 ratus is placed at the point where the temperature is to be 

 determined, the glycerin -cup being connected by a siphon 

 and tube with the distant registering apparatus, and the 

 beam being electrically connected with it. This apparatus 

 is simply a differential wheel-work actuated by an electro- 

 magnet, which causes a carriage carrying a pencil to traverse 

 horizontally a prepared paper. Should the temperature rise, 

 mercury would flow into the cup, cause the balance to de- 

 scend on that side, make electric contact cause the wheel- 

 work to move in one direction, carrying the pencil with it. 

 At the same time the glycerin surface would be raised, and 

 the liquid would flow through the tube into the reservoir, 

 and lift a float. But the action of the clock-work at the 

 same time depresses this float, raising the level of the liquid, 

 causing it to flow back to the balance again, and thus to 

 restore the equilibrium. 



An Italian optician in Paris, says Nature, has constructed 

 a very sensitive metallic thermometer on a new principle. 

 The dilations of a small sheet of platinized silver are ampli- 

 fied by means of a system of levers, and the motion is com- 

 municated to a needle on a dial on which degrees are mark- 

 ed. The motion of the needle is almost instantaneous. 



Fawcett has suggested a ready means of obviating the 

 deposition of moisture which often takes place in the interior 

 of minimum thermometers exposed on the grass. A piece 

 of cork, about a quarter of an inch long, is cut so as to fit 

 tightly around the neck of the thermometer tube, and then 

 this tube with the cork packing is inserted into the glass 

 case. The exposed end of the cork is covered with two 

 or three coats of asphalt varnish, and when this is dry 

 the protection is complete. 



2. Expansion. 



Reusch has described a simple form of apparatus for meas- 

 uring coefficients of expansion as a lecture experiment or for 

 students' use. Upon a horizontal axis, capable of rotation, a 

 mirror is fixed at one end, while near the middle, but out of 

 line witli the axis of rotation, is an abutting screw, against 

 which one end of the bar to be measured presses, the other 

 end being supported by a similar screw in the base of the 



