

NATURE 



3 6i 



! HURSMS, R# u/1919. 



OPTICAL RESEARCH \XI> DESIGN. 



blified Method of Tracing Rays through any 



ses. Prisms, and Mirrors. 



B) Dr. Ludwik Silberstein. With diagrams. 



Pp. ix + 37. (London: Longmans, Green, and 



[918. 1 P ice 55. net. 



THERE are two 1 riteria for estimating the value 

 of Di tein's latest work. On one 



hand it t tin 1 al interest, yet on 



the oilier ii lays claim to practical usefulness. One 

 must attempt to do justice to both aspects. 



From the theoretical point of view the author's 

 suggestion for the tracing of rays through optical 

 ems is deserving of the highest commenda- 

 tion. Progress in the application of mathematics 

 to problems in physical science is largely bound 

 up with the art of economy of symbolical and 

 menial effort, by means of the employment of the 

 briefest possible notation and the most powerful 

 analytical processes. The introduction of the 

 vector notation into geometrical optics is therefore 

 an important step in advance, and Dr. Silberstein 

 IS eminently the man to briny this about. He has 

 already done much to initiate English students into 

 the methoi ir analysis, and to convince 



them of its merits as a tool for the discovery of 

 new results, and as a medium for the brief formu- 



1 of discoveries both old and new. In par- 

 ticular he has published some beautiful results of 

 the a oi vector analysis to optical re- 



search. The latter portion of the present book 

 is a sufficient vindication of the claims of vector 

 methods, and there can be no doubt that they 

 repp is! powerful yet devised for attack- 



ing problems on reflection and refraction at 



ms ol plane surfaces. 

 The optical computer is, of course, interested 

 in plane surfaces; his main task, however, con- 

 sists in the tracing of rays through systems of 

 spherical surfaces, generally coaxial. In the first 

 half of the book the author develops on vectorial 

 principles the theory of ray tracing through such 



ms, and here, too. the superior merit of the 

 vector notation is made evident. One must never- 

 theless remark that his strictures on the old 

 trqnsfi r formulae are not fully justified, for it is 

 possible i" represent the transfer by means of a 

 comparatively short and simple set of equations, 



without the aid of .vectors. 

 When, however, we come to the consideration 

 of the practical aspect of the question, some doubts 

 arise as to the merits of the formulas thus derived. 

 1 1 is a great pitv that exact schedules for compu- 

 tation by logarithms or by the calculating machine 

 are not included ; they would have greatly en- 

 hanced the value of the book. Even one or two 

 actual examples, worked out in full, would have 

 I ome. It may be that the arithmetical 

 processes contemplated by the author do not 

 justify the suspicion, hut one cannot help wonder- 

 ing whether 1 he number of table-entries required 

 by his formulae does not exceed that required by 

 NO. 2567, VOL. I02] 



the well-know .inula'. Further, it has 



'i pointed out to the present writer that Dr. 

 Silberstein's method 



using the cosines of many of 



small angles and in optica] systems most of the 

 .: igles "i incident e, reft ... ergence 



are only a few degrees — cosines define the angles 

 with very little accuracy. Thus in the case of 

 angles of about io° the sines are six limes as 

 .1. .urate as the cosines, whilst logarithmic sines 

 are about thirty times as accurate as logarithmic 

 cosines. The further discussion of the vector 

 method in the light of these two criticisms would 

 be very desirable, and students of geometrical 

 optics will welcome any further contributions from 

 i>i. Silberstein's pen. S. BRODETSKY. 



PETROL AND PETROLFAM. 



(1) Petroleum Refining. By Andrew Campbell. 

 With a Foreword by Sir Boverton Redwood, 

 Bart. Pp. xvi + 297. (London: Charles 

 Griffin and Co., Ltd., 1918.) Price 25s. net. 



(2) Petrol and Petroleum Spirits. A Description 

 of their Sources, Preparation Examination, am! 



By Capt. W. E. Guttentag. With a 



Preface by Sir John Cadman. Pp. xi + 135. 



(London : Edward Arnold, 191S.) Price 10s. 6d. 



net. 

 'T'HF petroleum industry has developed enor- 

 ■*• mously during the last few years, and pro- 

 cesses of production and manufacture of products 

 have so changed from the old " rule-of-thumb " 

 methods which were in vogue in the early days 

 that it is only natural for the literature on the 

 subject to have increased in proportion. In the 

 mass of literature in the English language which 

 has been published on petroleum there has hitherto 

 been no publication on petroleum refining, 

 or one solely confined to petrol and petroleum 

 spirits, both of which are of particular interest at 

 the present time, when we are emerging from the 

 throes of a world-war in which petroleum and its 

 products have played such a prominent part. 



Mr. Andrew Campbell, the author of "Petrol- 

 eum Refining," is one of the pioneers of the 

 British petroleum industry, and his views on the 

 practical side of the subject are the outcome of 

 experienced knowledge. It is rare that technical 

 knowledge, experience, and clarity of diction are 

 found together, but when so found they render 

 the work of their possessor of inestimable value, 

 and it is the possession of these qualities which 

 is strikingly emphasised throughout Mr. 

 Campbell's book. 



The volume is divided into nine chapters as 

 follows: — (1) Examination of the crude oil; 

 (2) general departments: (3) storage of crude oil 

 J and liquid products; (4) distillation; (5) paraffin 

 extraction and refining; (6) candle manufacture; 

 (7) chemical treatments; (8) distribution of pro- 

 ducts; (i)) engineering specifications ; an appendix 

 1 onsisting oi a bibliography of the literature (either 

 published in English or of which extracts have 

 been published in English) appertaining to petrol- 

 eum refining, together with a name and subject 



U 



