1873.] Light. 419 



were it not thus confined, would be free to rotate, and which would describe a 

 circle of which its length is the radius. The first shaft, passed directly 

 through the axis of the steelyard, has fast on it, at one end, the driven pulley 

 and a bevel gear, which forms one side of a compound or box gear. The 

 bevel on the opposite side and the delivering gear are fast on a sleeve or 

 collar, which is caused to revolve in an opposite direction to the shaft by the 

 two intermediate bevels, which complete the compound, and which, being 

 driven by the first gear, rotate freely around the steelyard, when no power is 

 being transmitted, with the exception of such action as is due to the friction 

 of the shaft and gears, which friction is ascertained and deducted in making a 

 test of power. The weight used on the original machine was 1 lb., and the 

 length of the steelyard such as to move the weight at its extremity 10 feet in 

 one revolution, or 1000 feet in 100 revolutions ; therefore 1 lb. moved 1000 feet 

 represented 1000 lbs. moved 1 foot. In the improved instrument the steelyard 

 is made the base of calculation, and is graduated in inches and tenths, each 

 inch representing 100 lbs. Starting at 6$ inches from the centre to clear the 

 frame, the graduation is carried out 20 inches or to 2000 lbs., and the weight 

 obtained as follows : — The extreme length is now 26£ inches, thus being the 

 radius of a circle of 53 inches diameter, or 166*504 inches circumference. 

 This is 13*8753 feet described in one revolution, and the weight, to correspond 

 to 1 lb. moved 10 feet, is found to be 07207 lb. Now this weight is only one- 

 half of that required, for on the steelyard the action is that of supporting a 

 weight supported at one end, and the weight is therefore doubled, making it 

 1-4414 lbs., or 1 lb. 7^ ozs. to represent 1000 lbs. raised 1 foot in a second. 

 The poise or slide weight is obtained by a different calculation, as it is to 

 represent 100 lbs. for every inch it is moved, and as a circle of 1 inch radius 

 is 6-2832 in circumference, 100 times that circle is 628-32 inches, or 52*36 feet ; 

 and the weight to correspond with 1 lb. moved 100 feet, or 100 lbs. moved 

 1 foot, is found to be 1-9098 lbs., which, being doubled as before, gives 3*8197 

 lbs., or 3 lbs. 13/2 ozs. nearly for the movable weight. A worm gear on the 

 second shaft drives the clock by a pinion of 100 teeth, and at every 100 revo- 

 lutions rings a bell. The practical operation of the dynamometer is this : — 

 Having carefully levelled it and secured it to the floor in the proper line of 

 motion, a belt is brought from the driving pulley on the shaft to the first 

 pulley on the dynamometer, and one led from the second pulley on the dyna- 

 mometer to the main pulley of the machine to be weighed, and the whole put 

 in motion. The action of the bevel gears immediately raises the steelyard,, 

 which is then weighted till it remains motionless in a horizontal position. 

 This weight is noted, and also the number of seconds consumed in making 

 100 revolutions. The belt leading to the machine is then thrown off, and the 

 weight required to balance the dynamometer in motion also noted and deducted 

 from the total weight. The weight thus obtained is divided by the number of 

 seconds consumed, and the result is the number of pounds raised 1 foot in 

 i second. 



LIGHT. 



Mr. Charles Horner has contributed a paper to the "Chemical News" 

 on the " Spectra of some Cobalt Compounds in Blowpipe Chemistry." A 

 spectroscope of low dispersive power is essential for seeing distinctly the 

 bands due to these compounds ; and since the accurate position of the 

 absorption-bands had to be determined, the author had recourse to the 

 micro-spectroscope, which enabled him to measure them with precision, 

 although a small hand spectroscope, like Mr. Browning's " Miniature " in- 

 strument, is very convenient especially for examining the spectra of hot 

 beads. Mr. Sorby's interference-plate was used as a scale of reference, 

 whilst Mr. Browning's bright point micrometer served as an indicator. 

 When cobalt oxide is added to boric acid, and strongly fused for two or 

 three minutes in the inner flame, using gas as a source of heat, the cold bead 

 possesses a dull blue colour, and is almost opaque. However, by using a 

 powerful light it gives a spectrum of three very faint bands, nearly in the 

 same position as shown in Fig. 2 ; if, now, 1 per cent of sodium carbonate be 



