Dr. Lamont on the most advantageous Form of Magnets. 373 



From this it is cleducible that the augmentation of the breadth is 

 also to be considered as disadvantageous, but in a more trifling 

 proportion than we have found that of the thickness in the 

 second series of experiments. 



4th Series of Experiments. — Four needles (fig. 7) contracting 

 from the middle to sharp points at the two ends (rhomboids) 

 were cut out of an iron plate. They had all the same length 

 = 59'"'6 ; the breadths in the middle were very nearly in the pro- 

 portion of 1, 2, 3, 4, and amounted in the broadest needle to 

 19"'-5. The observation gave the following numbers : — 





Magnetic moment. 



Mass. Proportion to the mass. 



A . 



. . 4-304 



4-95 0-870 



B . 



. . 5-313 



9-84 0-539 



C . 



. . 5-914 



14-45 0-412 



D . 



. . 6-595 



19-45 0-339 



It appears hereby that the proportion of the magnetism to 

 the weight is the more advantageous the more sharply the 

 needles are pointed, that is, the smaller the breadth is in the 

 middle. 



5th Seizes of Experiments. — Three equal needles (fig. 8) 

 were made in form similar to those of the fourth series; length 

 46'"'0, breadth in the middle 13'" 3 ; from two of them a part 

 was taken out of the middle, so that they had the appearance of 

 perforated rhomboids, and the part cut out was similar to the 

 whole figure. The magnitude of the part cut out amounted in 

 B to one-third, and in C to two-thirds of the whole figure. The 

 observation gave — 





Magnetic moment. 



Mass. Proportion to the mass. 



A 



. . . 3-46 



1-02 3-39 



B 



. . . 3-47 



0-85 4-08 



C 



. . . 317 



0-52 604 



It is therefore very advantageous to take out a part of the 

 mass in the middle. 



6th Series of Experiments. — In the fourth and fifth series of 

 experiments the needle contracted to a point from the middle 

 towards the two ends ; in the present series of experiments it is 

 to be ascertained what difference depends on the circumstance 

 whether the breadth begins to diminish directly from the middle 

 or nearer to the ends. For this purpose flat pieces of steel were 

 employed of 43'"- 1 length, l"'-0 thickness and 10"'-0 breadth 

 (in the middle), whose figure is represented in fig. 9 ; the part 

 a b amounted in B to a sixth, in C to a third, and in D to a 

 half of the length. The results were — 



