Chemistry and Physics. 405 



)niis to -001772 grains, or about an eighth of a grain per square 

 ot. From this it would api)ear that the light of the sun would 

 i equivalent to 32 grains per square foot or 57 tons per square 

 ile. Mr. Crookes further applies this instrument as a photometer 

 id suggests its application to observatories to determine the 

 tal amount of sunlight received during the year. The number 

 ' revolutions could be counted by attaching a magnet to the 

 diometer which should act on a magnetic needle moving a 

 unter outside of the glass case. Similarly the power of the in- 

 rument might V>e transmitted through the glass without the 

 ual loss by friction. 

 Numerous other articles appear on the same subject. Poggei 



that the repulsion is due to convection currents. Several articles 

 appear also in ' Nature,' on the radiometer ; on p. 891, Mr. Crookes 

 shows that the repulsion is inversely as the square of the distance 

 and compares the elfect of rays of various wave lengths. On page 

 324, Mr. Hutchinson states that a radiometer with mica vanes on 

 metallic supports revolves more rapidly with dark heat than with 

 light, but Mr. Crookes replies that pith should be used as the 

 absorbing substance, since metals give erratic results.] e. c. p. 



4. The Grant Magneto-electric machine. — M. Tresca has made 

 a careful measurement of the power required to diive a large and 

 a small gram machine and compared the result with the light gen- 

 erated. A photometer disk was used, of which one portion was 

 illuminated only by the electric light and an adjacent portion 

 only by a careel burner consuming 40 grams of oil per hour. 

 Much trouble was experienced from the difference in color of the 



the equality was best obtained by interposing t 

 of light 



nd the oth( 

 egularities in^the carbons the light contimni 

 egularities sensible only to the photometer. 'H 

 arger machine was placed 40 meters from the dh 



burner moved until the square of their dis 

 1850: 1, which was about the mean ratio of the two lights. When 

 the two portions of the disk appeared equally bright the observer 

 gave a signal and instantly the power and velocity were observed. 

 The larger machine had a length of 80 cms., width 55 cuhs., and 

 height 58o cms. The average number of turns per minute was 

 1274, and the work 576 killograra meters or 7-68 horse-power. 

 The light being 1850 burners, would equal -415 of a horse-power 

 per 100 burners, or -31 kgms. per burner. 



The smaller machine had a length of 65 cms., breadth of 41 cms., 

 and height of 50-6 cms. It made 872 turns per minute, and gave 

 a light of 302-4 burners. This required 211 kgms., or 2-8 horse 

 power, equivalent to '92 of a horse power per hundred burners, or 

 '69 kgms. per buraer. 



The consumption of oil to produce a light equal to that of the 

 larger machine would be about 71 kgs. per hour or 194 cubic 



Am, Jour. 8ci.— Third Series, Vol. XI, No. 65.~Mat, 1876. 



