Lifting Power of the Electro-Magnet. 



519 



keeper of the iron magnet, I could have one all iron, one in which the parts 

 excited are steel, but the induced parts iron, and one all steel. The helices 

 N, N' first prepared are rolled on wooden bobbins r-33 diameter and 7' long, 

 containing 630'4 spires in four layers, with an external diameter 1' 95. After- 

 wards, for temperature experiments, helices O, 0' were rolled on copper bob- 

 bins, 2''8 diameter by 5'-7 long, with eight layers of No. 12, external diameter 

 = 4'66, containing 677 spires. When used, they were blocked up to be in con- 

 tact with the keeper, and, in consequence of the average distance of their spires 

 from the poles being less, they are more effective than N, N', in the ratio of 

 1 : 0'96;* they are less resisting, their measures being 59'840, and 78-416. 

 They were generally used, but for the higher i^s the others were added col- 

 laterally, and the -f computed by the method given in my second paper, p. 524. 

 Small Iron Magnet. — It was necessary to begin with the temperature cor- 

 rection, for I had now learned that this depends on the size as well as the form 

 of the magnet. Two copper cylinders, with central tubes which fit on the 

 arms, are connected below by a cross tube which is coupled to a small steam 

 boiler, and are provided with means of escape for the waste steam and condensed 

 water: on these the helices O, 0' were placed, and the ^ was made 600. In 

 the first of the following Table the magnet was cooled 27° below the tempera- 

 ture of the room, by keeping a stream of ice-cold water flowing through the 

 cylinders. 



Table XXIX. 



* Determiaed by thirty observations. 



3 y2 



