148 ANNUAL KEPOET SMITHSONIAN INSTITUTION, 1915. 



Almost contemporaneous with the work of Herschel was that of 

 M. C. S. M. Pouillet, a record of which, on the determination of the 

 sohir constant, appears in Comptes Rendus, Vol 7, 1838, pages 24-G5. 

 His value of the solar constant was 1.763 calories per square-centi- 

 meter-minute. 



Carl Giintner w as at work experimenting with reflectors in Lai- 

 bach in 1851, and in 1873 he exhibited one at the Vienna Exposition. 

 Giaitner wrote in the Scientific American Supplement of May 2G, 

 1906, pages 25, 409-412 : 



This reflector possessed, however, the disadA'antages common to all sheet-metal 

 reflectors — that to maintain the surface in proiDer condition when exposed to all 

 sorts of weather requires careful and costly attention. 



Being convinced, however, that the exploitation of solar heat will come more 

 and more into vogue, even in spite of the disadvantageous periodicity of this 

 source of warmth, I have taken the trouble to put aside the evil mentioned above 

 and overcome it by an entirely new method of reflector construction. * * * 

 This plane reflector consists of a large number of long, narrow mirrors placed 

 at suitable distances from one another, and which, when upon level ground, lie 

 parallel with each other horizontally, extending either from north and south or 

 from east to west. 



Each one of these mirrors revolves about a horizontal axis, and by means of 

 a simple parallelogram motion may be made to follow the sun in such a manner 

 that all the sun's rays falling upon the plane mirrors may be reflected on the 

 surface of u tube or boiler, the long axis of which lies also in the plane of the 

 mirror axis * * *_ jjy ^ simple movement of a hand lever, all the mirrors 

 may be simultaneously turned through an arc of 180", which means that all the 

 mirrors may thus be made to look toward the ground and be in this way pro- 

 tected from the destructive fiction of sudden falls of hail. 



He claimed that the reflector coidd be made at a cost of 8s. 6d. per 

 square yard of reflecting surface, and that it required " but 200 square 

 feet of surface to generate steam sufficient for 1 horsepower." He 

 proposed to construct the reflectors of thin corrugated steel plates, 

 faced with lead and then coated with tin. 



Hence it is necessary to discover the value of e (the amount of nseful heat 

 dispensed per unit of surface per minute) which affords the unit of heat that 

 can be made available for effective service from a square foot of catching sur- 

 face per minute. 



Being deprived of the experience of any former experimenter in this direction, 

 I myself made appropriate trials with reflectors * * *. The two opposite 

 sides, each 3 feet long, of a wood right-angular frame, having a width of 1 foot 

 and a length of 3 feet, were hollowed out to correspond with a previously de- 

 signed parabolic template, and upon the parabolic curve thus established two 

 sheets of white tin were nailed. Four sui>ports, which were fastened to the 

 sides of the frame, carried a SJ-inch tube erected in such a manner that its axis 

 coincided with the burning axis of the reflector * * *, The catching surface 

 presented a superficial area equal to 3 square feet * * *. The boiler was 

 not lagged with glass or anything. 



