﻿556 Mr. F. L. 0. Wadsworth on a very 



be connected either in series (resistance 86 ohms) or in 

 parallel (resistance 21+ ohms) ; or by changing two con- 

 nexions inside the case, all four coils could be put in parallel. 

 Total (5 ohms) as desired. 



The case was unusually large and heavy, with brass frame 

 and glass sides, with all joints as air-tight as it was possible 

 to make them. It had the usual doors and levelling-screws. 



The constant of the galvanometer with the magnet system, 

 already described*, was for all coils in series, C = 4xl0~ 11 , 

 where C, as before, is the current in amperes required to 

 produce a deflexion of 1 millim. at a distance of 1 metre 

 for a time of single swing of 10 sec. This is about the 

 degree of delicacy recently attained by Snow f, the constant 

 of whose instrument reduced to the units above used was 

 C = 4'5 x 10 -11 for a somewhat higher resistance (140 ohms). 



More recently, Paschen J has attained a degree of delicacy 

 considerably exceeding this by the use of an excessively light 

 magnetic system, which did not appear to be stable enough 

 to be actually used with a 10 sec. period in measurement. 



The question of the influence of the mass of the system on 

 the delicacy is one of importance. It was first pointed out 

 by Boys, and later by Paschen, that if a fixed time of vibra- 

 tion be considered, the sensitiveness of the galvanometer 

 (other things being equal) will increase as the mass of the 

 system decreases. This has been experimentally verified by 

 Du Bois and Rubens, who, by the use of light magnetic 

 systems, have recently produced commercial galvanometers 

 having a sensitiveness nearly equal to that of Snow's instru- 

 ment §. 



But the assumption of a fixed time of vibration is unfair 

 to the heavier systems, for it conditions them to a degree of 

 more and more imperfect astaticism as the weight increases. 

 If, instead of the condition of a fixed time of vibration, we 

 impose that of a fixed degree of astaticism, viz., if we make 

 the residual magnetic moment of the systems as a whole 

 constant, then the sensitiveness of the galvanometer will 

 increase with the mass of the magnetic system, supposing 

 always that the coils of the instrument are suitably propor- 

 tioned to the needle, A better understanding of these points 

 is afforded by their analytical expression. We will suppose 



* As has already been stated in a previous footnote, this system could 

 be very sensibly improved, but as the sensitiveness of the galvanometer 

 even with the present system is greater than necessary, and as my time 

 has been fullv occupied with other work, this has not yet been done. 



t Wied. Ann. vol. xlvii. p. 218 (1892). 



X Ibid. vol. xlviii. p. 284 (1893). 



§ Wied. Ann. vol. xlviii. p. 236 (1893). 



