NATURE 



169 



THURSDAY, JUNE 23, 1904. 



THE LIFE WORK OF A SCIENTIFIC 

 EXGINEER. 

 Original Papers by the late John Hopkinson, D.Sc, 

 F.R.S. ^■ol i., Technical Papers. Edited by B. 

 Hopkinson. Pp. lxvi+294. Vol. ii., Scientific 

 Papers. Pp. vii + 393. (Cambridge: University 

 Press; London : C. J. Clay and Sons, 1901.) Price 

 los. 6d. net each volume. 



THREE years have elapsed since the two volumes 

 of original papers by the late Dr. Hopkin- 

 son were published, and an explanation is naturally 

 required for such a protracted interval being allowed 

 to elapse before the work was reviewed in these pages. 

 Shortlv after its appearance, the writer became seri- 

 ously ill, and the diminished vigour that accompanied 

 a long tedious convalescence was marked by increased 

 requests that work should be undertaken, exemplify- 

 ing apparently the anecdote of the doctor who to 

 obtain rest doubled his fees, but only succeeded there- 

 by in doubling his practice. Hence the performance 

 of a duty had to be postponed again and again, and 

 it was not until Mr. C. S. Whitehead was kind enough 

 to bring his mathematical power to bear on a critical 

 analysis of this collection of papers and furnish the 

 substance of much which follows that this tardy 

 review came to be written. 



In these two volumes we have the collected works 

 of a man eminent not merely as a scientific investi- 

 gator, but also as an engineer; they constitute the 

 record of a worker who took a very leading part in 

 showing his fellow electrical engineers how the appli- 

 cation of scientific theory and knowledge helped more 

 than crude trial and error in solving some of the 

 numerous problems with which their industry abounds. 

 Pioneers, as a rule, have the mortification of seeing 

 their advances overlooked and neglected by their con- 

 temporaries, but Dr. Hopkinson had the happy fortune 

 of being spared this; indeed, his papers resemble the 

 writings of Shakespeare in that they appear to be full 

 of quotations. 



It might have been expected that a mathematician 

 like Hopkinson, a senior wrangler and Smith's prize- 

 man, would have freely used mathematical processes 

 of some complexity, a transformer, for example, offer- 

 ing a most tempting field for mathematical excursion. 

 But this is far from being the case ; the great majority 

 of the processes are such as can be easily understood 

 by anyone moderately well versed in the calculus. 

 But it must not be supposed that all these papers are 

 easy reading ; some of them are far from it, and it 

 often takes considerable thought to grasp their mean- 

 ing. 



The first volame contains the technical portion of 

 the papers, and opens with two dealing with light- 

 houses, one pointing out the advantages of what are 

 known as group-flashing lights, the other describing 

 the optical and electrical apparatus at the lighthouses 

 of Macquarie and Tino. 



The remainder of the volume deals almost exclu- 

 sively with dynamo electric machinery. In the 

 NO. 1808, VOL. 70] 



papers reprinted therein Dr. Hopkinson first shows 

 how the curves now universally known as the 

 " characteristic curves " are to be plotted, and 

 then how to extract useful information from 

 them ; how, for instance, to determine the lowest 

 speed at which a given dynamo can produce a short 

 arc. He then directs attention to the necessity of tak- 

 ing account of both colour and direction in 

 measurements of the brightness of the electric 

 arc. A little further on we come to two papers on 

 dynamo electric machinery, the first of which was 

 written in conjunction with his brother. Dr. E. Hop- 

 kinson, in 1S86. In these papers, which were pub- 

 lished at about the same time as an equally important 

 one on the same subject by Mr. Kapp, we are shown 

 how the characteristic may be predetermined from 

 theoretical principles based on the equations of a mag- 

 netic circuit, and on the magnetic properties of iron 

 as found by experiment. 



The paper of 18S6 is classical, for with that of Mr. 

 Kapp it laid the foundation of the design of electric 

 machinery. Previously the proper shape to give to a 

 dynamo was unknown, and it was impossible to fore- 

 see what effect on the performance of a dynamo would 

 result from altering its shape. The late Prof. Row- 

 land advocated long, lanky-legged machines, and Mr. 

 Edison told the writer that the most astonishing thing 

 in his life was finding that the dynamo which he sent 

 to the Paris Electrical Exhibition of 1S81 developed 

 about the amount of electric power that he had hoped 

 it might give. 



Next follows the ingenious and economical method 

 of testing the efficiency of a dynamo by coupling it 

 mechanically to another of approximately equal size, 

 and measuring the extra power that had to be sup- 

 plied from an external source to keep the combination 

 running when one of the machines acted as a motor 

 and drove the second, which in its turn acted as a 

 dynamo and supplied current to the first. 



In reading through these papers one cannot fail to 

 be struck with the keen insight which their author 

 displays in picking out the essential points needing 

 examination in the machine under discussion, and 

 with the beautiful methods which he employs for pre- 

 senting the results of his investigations. Thus, in his 

 papers on the dynamo, he realised that, if the machine 

 were to be improved by scientific study, it was abso- 

 lutely necessary to ascertain, not merely how much 

 power was put into it, and how much came out in a 

 useful electrical form, but also what was going on in 

 the various parts of the machine itself. Others before 

 him had considered the input and output. Dr. Hop- 

 kinson, in conjunction with his brother, took up the 

 second, and equally important, part of this investiga- 

 tion, and gave the results to the world in a manner 

 that was as simple in expression as it was novel in 

 conception. 



The papers on alternating currents are, speaking 

 generally, more important for what they suggest than 

 for what they actually prove. Thus, to take those 

 that deal with the parallel running of alternators. 

 Two lines of argument in support of the opinion that 

 they can do so are brought forward, one a purely 



I 



