S. O. MAIZEL, Light and Vision, GTTI, Moscow–Leningrad, 1932, 123 pp., price 3 rubles 20 kopecks.

As the author, one of our best photometrists, states in his preface, “the proposed book is not a textbook, on the one hand, nor, on the other, a monograph claiming to cover the subject fully. Its aim is to set forth the most fundamental questions of light and vision, knowledge of which is necessary for every lighting engineer.” The author provides in it much valuable factual and numerical material, for the compilation of which all those working in this field will be grateful to him.

In its content the book is divided into two parts. The first part is devoted to the basic photometric quantities and concepts, the second to our visual apparatus. In comparison with the first edition (1925), a large chapter has been added on the color characteristics of luminous flux. With regard to this chapter the following should be pointed out. Prof. Maisel, being a photometrist, also regards colorimetry as a part of photometry. This, however, is incorrect. In his classical memoirs on the metric of color, E. Schrödinger quite clearly and sharply distinguished the domains of the lower metric of color, the foundations of which were formally given by Grassmann and which deals only with settings for complete identity, and of the higher metric, to which belong all the problems of heterochromatic photometry and which deals with settings for greatest similarity, barely noticeable difference, or greatest contrast. Schrödinger writes that “the strict separation of these two domains” he considers “exceptionally important,” since otherwise there is the danger that “many ambiguities in the fundamental concepts and the unreliability of experimental data, which we encounter at every step in the higher metric, may firmly take root also in the domain of the lower metric and introduce confusion (Verwirrung) there, whereas it has been soundly and reliably grounded (fest und sicher in sich ruht).”

Schrödinger rightly indicates that the bridge between the two domains is laid by the theory of fundamental sensations. Since, however, König’s old data are known to be inaccurate, and the new curves of the “fundamental sensations” have not yet been obtained,* we do not yet possess such a bridge. The replacement of the eye by three receivers with spectral sensitivities \(r_\lambda, g_\lambda, b_\lambda\), on which, by the way, S. O. Maisel bases his entire scheme of colorimetry, is therefore in practice still impossible, although in principle there is nothing incorrect in such a replacement. It should also be noted that the author’s following assertion is incorrect: “Two persons obtain identical and at the same time important results on a colorimeter only in the case that their eyes are identical in the sense of the coincidence of the course of the function \(V_\lambda\) with its course for the average eye” (p. 112). In fact: from the formula \(V_\lambda = \alpha r_\lambda + \beta g_\lambda + \gamma b_\lambda\), where \(\alpha, \beta\), and \(\gamma\) are constants, it follows that individual differences in the course of the functions \(V_\lambda\) may depend both on differences in the functions \(r_\lambda, g_\lambda, b_\lambda\), and on differences in the coefficients \(\alpha, \beta, \gamma\) with identical \(r_\lambda, b_\lambda, g_\lambda\). Conversely: from the identity of the functions \(V_\lambda\) one still cannot conclude that the color-mixture curves are identical. Bearing this in mind, as well as certain other circumstances, one of the best modern colorimetrists, Guild (London), writes as follows: “... it is not proper to determine the quality of color by methods that include any photometric determination whatever.”

And, finally, Fig. 22 (on p. 115) should be replaced by another, since the curve of “visibility” for peripheral vision ends considerably earlier than is shown there (already at 620 mµ, and not at 700 mµ).

N. Fedorov

* At present they are being determined by various methods in London by Wright and in Moscow by us.

N. F.