November 30, 1^93] 



NA TURE 



I r 



this method in the study of a variety of allied phenomena. 

 After the publication of Joubert's papers the method seems to 

 have come into common use in the physical laboratories, par- 

 ticularly in the exploration of the fields of continuous current 

 dynamos and motors. 



In i888 it was applied by Duncan, Hutchinson, and Wilkes 

 ("Experiments on Induction Coils," ^/t'cVr/ca/ World, vol. ii. 

 p. 160, 18S8) to the study of induction coils and transformers. 

 To them we owe the first set of complete diagrams relating to the 

 performance of this class of alternating current apparatus. In 

 the same year Meylan("Sur les Appels Magnetiques," La 

 Lumicre Ehctrique, xxvii. p. 220, 1888) used an interesting 

 modification of the method in the investigation of the vibratory 

 magnetic call-bell of Abdank. 



In the same year appeared the first definite data with reference 

 to the Westinghouse alternating dynamo, at the hands of Messrs. 

 Searing and Hoffmann (" Variation of the Electromotive Force 

 in the Armature of a Westinghouse Dynamo," yoiiriial oj the 

 Franiliti Institute, vol. cxxiii. p. 93), of Stevens Institute. 

 Then followed in the order named the researches of Ryan and 

 Merritt ("Transformers," Trans. Am. Inst. Electrical Engineers, 

 vol. vii. p. I, 1889),' Humphrey and Powell (" Efficiency of the 

 Transformer," Ibid, vol. vii. p. 311), Tobey and Walbridge 

 (Ibid, vol. vii. p. 367), of Marks (Ibid, vol. vii. p. 324), of 

 Herschel {Ibid, vol. vii. p. 328), of Fortenbaugh and Sawyer 

 (Ibid, vol. vii. p. 334). 



In all these investigations the methods under consideration 

 liave been used with varying accessories in the problem of the 

 transformer. 



In 1890 it was applied under much more difficult conditions 

 to the analysis of the "ball and point effect" by Archbold and 

 Teeple (see Nichols, " On Alternating Electric Arc between a 

 Ball and Point," American J ottrnal of Science, vol. xli. p. i). 



In 1891, Thompson ("Study of an Open Coil Arc Dynamo," 

 Trans. Am. Inst. Electrical Engineers, vol. viii. p. 375) deter- 

 mined the intricate changes of induction in open coil arc lighting 

 machines by means of the same method, and Ryan (" Relation 

 of the Air Gap and the Shape of the Poles to the Performance 

 of Dynamo-electric Machinery," Ibid, p. 451) utilised it in his 

 investigations of the influence of the air gap upon the perform- 

 ance of dynamos and motors. In 1892, Duncan ("Note on 

 some Experiments with Alternating Currents," Ibid, vol. ix. 

 p. 179) described modifications of the method of instantaneous 

 contacts by means of which the rapidity of reading is greatly 

 enhanced. 



During the present meeting, you will doubtless have the 

 pleasure of listening to a description of the applications of the 

 same device to the study of electrostatic hysteresis. (Reference 

 is here made to the paper presented by Messrs. Bedell, Ballan- 

 tyne, and Williamson: "Alternate Current Condensers and 

 Dielectric Hysteresis," Physical Review, vol. i. p. 91. Subse- 

 quent note.) 



Such has been, in brief, the history of a method by means of 

 which in greater degree than of any other we have been able 

 to extend and complete our knowledge of alternating current 

 phenomena. 



To the practical electrician and to the theorist alike, the do- 

 main has been one of the most attractive of those which have 

 been developed in recent years. To the electrical engineers of 

 ::he younger generation the very complexity of alternate current 

 iheory has proved a benefit. It has forced them to increased 

 mathematical proficiency and to more rigorous thinking; it has, 

 ndeed, served as an excellent source of discipline. What the 

 3roblems of submarine telegraphy did for the English electricians 

 jwho served their apprenticeship during the early days of the 

 ':able-laying industry, compelling the development of those 

 turdy qualities which have been so highly serviceable in every 

 Jranch of electrical progress since, the intricacy of alternate 

 :urrent practice is unquestionably doing for the younger school ' 

 vhich is growing up to-day in this country. The difficulties 

 vhich have to be met and overcome in this field of work will 

 ave an excellent influence upon the manner in which the > 

 iioblems of the future will be approached. 



I Another investigation, which owes its existence to a most in- 

 enious application of this same principle of instantaneous 

 ontacts periodically repeated, is well known to all of you. I 

 efer to Prof. E. H. Hall's ^ study of periodic heat- flow in the 



J^ "A Thermo-Electric Method of Studying Cylinder Condensation in 

 team-Engine Cylinders." Trans. Am. Inst. Electrical Engineers, vol. 



NO. 1257, VOL. 49] 



cylinder walls of the steam-engine by means of therrno-elements 

 embedded within the metal and connected momentarily during 

 a selected time-element in the course of each stroke with a sen- 

 sitive galvanometer. To my mind no more interesting example 

 of the indirect method of studying the phenomena of the time 

 element could be found than this suggestive memoir. 



The method of instantaneous contacts has been a fruitful one, 

 and productive of high results, but it does not yield a knowledge 

 of any individual time-element, nor the picture of any single 

 completed cycle. Numerous attempts to record single cycles 

 have been made, the results of which are of considerable 

 interest, because they deal with the more direct study of the 

 time-infinitesimal. 



The device which lay nearest to hand, and which by its per- 

 formance seemed to promise success in this direction, was the 

 magneto-telephone. The investigations of Mercadier [jourjial 

 j de Physique, vol. ix. pp. 217 and 282) had already paved the 

 way to some extent, when P'roehlich described his experiments 

 i upon the optical representation of the movement of the 

 i diaphragm of the telephone, followed almost at once by 

 j Thomson. 



I Froehlich 1 reported his preliminary results to the Electro- 



technische Verein of Berlin, in 1887. Elihu Thomson ("An 



: Indicator for Alternating Circuits." La Luinicre Electriqtie, 



\ vol. xxvii. p. 339 (1888) brought out his indicator for alter- 



j nating circuits, an instrument in which the movement of a 



i diagram was amplified by levers, and then made visible by 



optical means (or photographed) in the same year. Froehlich's 



method in its complete form, including the photography of the 



images from the involving mirror,-' was first described in the 



year 1889. Some of the curves published in the papers just 



cited, and particularly the experinients shown in the exhibition 



• of the method at the Frankfort Electrical Exposition of 1891, 



are most striking, but considering the method by which they 



are produced, the question inevitably arises as to the part played 



by the ineitia of the moving masses. 



Froehlich himself points out the necessity of great care in the 

 matter of the adjustments, and of distinguishmg the natural 

 oscillation of the plate, which are frequently superimposed upon 

 those to be recorded. Some experience with Froehlich's method 

 has convinced me that not only is extraordinary skill necessary 

 in order to obtain, by means of a mirror attached to the dia- 

 ' phragm of a telephone, curves which should represent, even 

 with a fair approximation, the law of whatever periodic changes 

 we may desire to record, but that the attainment of the proper 

 i adjustment is a matter so entirely fortuitous, and its maintenance 

 I so uncertain, as to deprive the method of much of its useful- 

 ness. One may indeed hope to get, by means of successive ad- 

 justments, curves which correspond to a known type, but 

 whether in passing to new and unknown types the apparatus 

 retains its faithfulness, is always a question. 



By way of illustration, I introduced three of an extended 

 series of curves obtained by this method with a telephone in 

 circuit with an alternating current dynamo. The character of 

 the cycle had been determined by the method of instantaneous 

 contacts. The true cycle was represented by a curve of sines, 

 but with the apparatus under consideration complex curves of 

 the kinds shown in figures 2 and 3 were the rule ; curves even 

 approximating to simple sinuosity were the rare exception. 



The difficulties of the method lay not merely in the tediousness 

 of adjustment, but rather in the tendency to revert to complex 

 forms under changes of condition so slight as to be entirely 

 beyond control. The remedy clearly consists in the elimination 

 of mechanism and the reduction of inertia of the moving parts. 

 Following the suggestion of an assistant, Mr. E. F. Northrup, 

 I tried the following experiments : — 



A mercury stream flowing from the contracted nozxie of 

 a funnel (Fig. 4) was made to pass between two metal terminals 

 which were attached to the poles of a large Holtz machine. A 

 portion of the falling column of mercury within the electrostratic 

 field was illuminated by means of an arc lamp, and so much of 

 it as could be seen through a horizontal slit was photographed by 

 transmitted light. The sensitive plate was given rapid vertical 

 motion through the field of the camera. When the machine 

 was out of action there resulted a vertical trace running the 

 length of the developed plate. As soon as the machine was put 



1 "The Optical Representation of the ]Movements of a Telephone Dia- 

 phragm." La LuiniLfc Elei:triquc, vol. xxv. p. 180 (1S87). 



- Froehlich : " Ueber eine neue Methode zur Darstellung \on Schwing- 

 ungskurven." Electrotechnische Zcitschrift, bd. .x. pp. 345, 369 (1SS9). 



