542 



NATURE 



[July i, 1920 



ventive medicine as now understood. The general 

 practitioner's part in "field" and "team" re- 

 search might well form the subject of a special 

 reference to the Consultative Council on Medical 

 and Allied Services. If the world of general 

 practice does not realise that research is of vital 

 importance to every branch of medicine, such is 

 certainly not the case with the world of science. 



Theory of Dioptric Instruments. 



Ferraris' ''Dioptric Instruments'': Being an 

 Elementary Exposition of Gauss' Theory >and 

 its Applications. Translated by Dr. Oscar 

 Faber from Prof. F. Lippich's German trans- 

 lation of Prof. Galileo Ferraris' Italian work 

 entitled "The Fundamental Properties of Diop- 

 tric Instruments." Pp. xxxi + 214. (London: 

 H.M.S.O., 1919.) Price 45. net. 



THE original of this translation was published 

 by Prof. Galileo Ferraris, of Turin, in 1876. 

 As a copy of this original could apparently not 

 be procured, the English translation was made 

 from a German one by Lippich, which appeared 

 in 1879. At the time of its appearance the book 

 unquestionably marked a great advance in the 

 treatment of its subject, and well deserved the 

 extremely favourable review with which Abbe 

 honoured the German translation in the first 

 volume of the Zeitschrift fiir Instrumentenkunde . 



Abbe himself, however, has to be credited with 

 far greater advances in the theory of image- 

 formation by optical instruments with which the 

 book before us deals, for his purely geometrical 

 treatment of the problem leads to the same 

 results without being limited to the infinitely con- 

 stricted "threadlike space around the optical 

 axis " which still plays so large a part in text- 

 books, although, with light of finite wave-length, 

 nothing of any optical interest can possibly happen 

 within tt. On the other hand, Abbe was the first 

 to deal systematically with the actual course of 

 light through instruments in accordance with the 

 limitations imposed by restricted apertures and 

 by deliberately placed diaphragms, and inasmuch 

 as the great majority of actual instruments are 

 used only at fixed or nearly fixed conjugate dis- 

 tances, the actual course of the rays so deter- 

 mined is of far greater importance and value, both 

 in the designing of instruments and in the dis- 

 cussion of" the effects produced by them, than 

 the rays referred to the Gaussian principle and 

 focal planes and points which form a convenient 

 pons asinorum in the general theory of lens 

 systems. 



Ferraris' treatment of the Gaussian theory is, 

 NO. 2644, VOL. 105] 



however,, less open to the objections just alluded 

 to than that adopted in most books, and in deal- 

 ing with the Galilean telescope he comes remark- 

 ably close to the correct treatment of the problem 

 of its field of view, which is so easily obtained 

 now by Abbe's theory of the entrance- and exit- 

 pupil of instruments. Beginners and users of 

 optical instruments desiring to acquire a general 

 knowledge of their elementary theory will also 

 welcome the numerous and frequently elegant 

 graphical solutions of the various problems which 

 are given throughout as alternatives to numerical 

 calculations by algebraical formulae. The chief 

 and decidedly regrettable omission is that the 

 simple problem of achromatism is not dealt with 

 at all. It is, of course, not a part of the Gaussian 

 theory, and the omission is therefore justifiable ; 

 but it is so closely bound up with the proper 

 explanation of the effects produced by compound 

 object-glasses and eyepieces that the book would 

 certainly have gained in value if the subject had 

 been included. 



The book is not so free from misprints as one 

 would wish, and there is a really bad muddle on 

 pp. 87-94, where the properties of thick lenses 

 are discussed. This is not a case of a simple 

 misprint or transposition of diagrams, but of 

 actual errors by either the original author or one 

 of the translators. Thus on p. 87 a thick bicon- 

 vex lens is stated to be convergent if its thickness 

 is less than one-third of the sum of the radii (both 

 taken as positive, with \L^\e^). This should be 

 three times instead of one-third. Then, on 

 p. 93 a meniscus with the shorter radius on its 

 concave face is stated to be always convergent ; 

 and on p. 94 the meniscus with the shallower 

 curve on the concave face is credited with being 

 divergent, telescopic, or convergent according to 

 thickness. The actual facts are, of course, the 

 other way about. Immediately after this the 

 properties of a concentric lens are correctly 

 stated. 



On p. 144 the strange conclusion is reached 

 that of two eyepieces of the same equivalent focal 

 length that one is to be preferred which has the 

 closer eye-point. This is directly contrary to the 

 experience of every observer. 



In the calculations of the properties of the 

 human eye, or rather of its "simplified model," 

 the author sets a very bad example by starting- 

 with data given with three significant figures and 

 undoubtedly uncertain even then in the third 

 figure, and calculating all the deduced figures with 

 six, and even seven, significant figures (pp. 71-75). 

 The idea of beginners that the percentage- 

 accuracy of observed data can be indefinitely 

 increased by putting them through the mathe- 



