506 



The Rev. T. E. Eobinson's Experimental Eesearches on the 



The second column gives A, the distance between the slide and keeper ; fx is, 

 as formerly, the ratio of the actual ^ to that which would produce the same L, 

 if the circuit were a minimum.* The next column gives the ratio of the L in 

 this last case to each of the others. In 89-92 the spirals were placed in inter- 

 mediate positions. These were the result of a separate set ; but the values for 

 z = 0"07, and z = 10-07, agree so closely with the first set that it seemed best to 

 group them together. 



It is seen at once that the value of L decreases as the length of the circuit 

 is increased : thus, for c = 32-2, it is only half what it is when c= 12-3. The 

 effect of z is similar, though less in amount, and is also dependent on c. These 

 relations are exhibited in — 



Table XVII. 



The second column gives the ratio of L, when the circuit is a minimum, to 

 that obtained with other circuits, the spirals being at the keeper ; the least 

 complicated case, as depending only on c, and giving the maximum of effect for 

 a given circuit.f The third column gives the ratio of the same L to that got 

 when the spirals are on the slide. This involves both z and c, and gives the 



* Of these /i appears best for compaTing the po-wers of helices, but the other is preferable for 

 a comparison of the magnets. In the range of these experiments they agree sufficiently for all 

 purposes of interpolation ; but the former becomes very irregular -when the maximum is ap- 

 proached, as will be ob-vious -when the progression of L is examined. I think it likely that the 

 quotient of the other by the area of the magnet's section is the best measure of the efficiency of 

 that particular form of magnet. 



t The maximum would, I think, be obtained if the spirals could be placed at the middle of 

 the keeper AB, and the minimum at the middle of the base CD. 



