486 



NATURE 



[September 22, 1892 



As to the experiments which are " new and important," 

 there is no doubt that there is one which will prove to 

 be highly important if it can be always repeated to give 

 the same results under conditions which bear rigid 

 examination. The experiment is described as follows : 



"Very many years ago I showed experiments with 

 rotating discs, which proved that ' persistence ' does not 

 (at any rate, wholly) take place in the way previously 

 supposed ; in the retina, or in the individual parts sensi- 

 tive, as I maintain, variably sensitive, to light and colours. 

 In one of the experiments I refer to, a brightly coloured 

 disc was covered by a black disc having sector- shaped 

 openings, such as to render the entire disc area half black 

 and half coloured. When the discs are rotated at a 

 suitable speed, under a strong light, the entire rotating 

 disc appears more brightly coloured than an entire disc 

 placed near the rotating disc. Thus, the colour effect of 

 a disc, half of which is covered by black sectors, is by 

 rotation made equal to or greater than that of an entirely 

 uncovered disc of the same colour." 



This one experiment would have caused a good deal of 

 anxiety to those who have been at work at the general 

 theory of vision had they known of it. Fox Talbot, 

 Plateau, and others would have had to amend their 

 papers — for the " persistence of vision " would evidently 

 not obey the law which they adopted after submitting it 

 to such experimental proofs as they could devise. Other 

 experiments which the author brings forward as confirma- 

 tory of the 5-colour theory, it seems to the writer can 

 be equally well explained by the 3-sensation theory, and 

 probably by that of Hering. 



In conclusion, it seems safe to say that these two 

 last rival theories have not been overthrown by the work 

 under review. Both have their weak points, but the 

 number of them has not been increased by the exponent 

 of the 5-colour theory. 



ELECTRICAL RULES AND TABLES. 

 A Pocket-book of Electiical Rules and Tables. By John 

 Munro,C. E. , and Andrew Jamieson, M.I.C.E., F.R.S.E., 

 &c. Eighth Edition, Revised and Enlarged. (London : 

 Griffin and Co., 1892.) 



JUST eight years ago we reviewed the first edition of 

 this electrical vade mecum. The fact that we have 

 now to notice the eighth edition is abundant proof that it 

 has been found of service by the electrical public. That 

 it deserved well of those for whom it was compiled there 

 can be no doubt. The authors have been most active in 

 collecting information from all sources, and in extending 

 the work so as to keep the information contained in it 

 fairly representative of the current state of industrial 

 electricity. Since the first edition it has been almost 

 doubled in size, and much very important matter has 

 thereby been added. 



A special feature of the book as it now stands is the 

 short accounts of various branches of electrical engineer- 

 ing which have been contributed by specialists. Such 

 are Dr. Thompson's chapter on dynamo machinery, 

 Mr. Kapp's account of transformers, and Prof Ewing's 

 sketch of magnetic measurements. These are very valu- 

 able, and add much to the authoritative character of the 

 work as a guide to engineers more especially concerned 

 with electric lighting and transmission of power. 

 NO. I 195, VOL. 46] 



When the first edition appeared we noticed a number 

 of points in which we thought the book required amend- 

 ment. Looking over the present edition, we have been 

 struck with the very considerable improvement which has 

 been effected in point of precision and accuracy. But 

 we have again met with some passages in which we fancy 

 the work may be still further improved. 



First, on p. 10, we were not able to see before, and we 

 do not see yet, what the fact that the dimensions of 

 resistance in electromagnetic units are those of velocity,^ 

 has by itself to do with the velocity v, which is the ratio 

 of the electromagnetic to the electrostatic unit of quan- 

 tity of electricity. 



At p. 15 it might have been well to mention the con- 

 vergence of all the latest absolute determinations of the 

 ohm upon something like io6'3 cms. as the length of the 

 column of mercury, representing it according to the usual 

 specification. 



At p. 42 definitions of the pound avoirdupois and stan- 

 dard kilogramme are given, in which the precise tem- 

 perature and pressure of the atmosphere at which the 

 weights of the standard lumps of platinum are a pound 

 and a kilogramme respectively are carefully specified \ 

 We should like to know why there is not a reference to 

 the hygrometric state of the atmosphere as well ! 



With regard to the statement with respect to density, at 

 p. 43, it is not usual, we think, to define density otherwise 

 than as mass per unit of volume. It is therefore a quan- 

 tity of dimensions ML"*, whereas specific gravity is a 

 mere numeric. In cases in which specific gravity and 

 density are numerically the same, there is still this essen- 

 tial difference in nature between the two quantities. 



Of course, the same word density is used in a peculiar 

 sense, frequently, when applied to gases, and our experi- 

 ence shows that nothing has a more confusing effect 

 in the mind of an elementary student of physics and 

 chemistry than this double use of the word. It would 

 be well to insist, as is often done by careful teachers, 

 that it is relative density that is here meant, and not 

 density in the ordinary sense. 



With respect to the velocity of sound in air (p. 50) it 

 might be as well to notice that it depends upon the 

 temperature of the air. 



At p. 127, under the heading " Impedance," impedance 

 in its proper technical sense as sj'^^ + «-L^ is not de- 

 fined. The definition is given elsewhere in the book, but 

 there is no clue to it in the inde.x. In the last formula 

 the exponential e has fallen out from before its exponent. 

 Here we might remark that in a book of this kind, where 

 space is of great importance, and especially with such 



RT 

 lumbering exponents as — --— the use of the solidus no- 

 tation would be a great improvement. 



The authors will not think us inappreciative in making 

 these remarks. In a work dealing with such a multi- 

 farious set of topics it is difficult even in several editions 

 to completely eliminate error, and we have made these 

 notes (and some others) in case the authors may care ta 

 make use of them. As we have said, the book is a 

 useful and handy synopsis of electrical information of all 

 kinds, and is very worthy to take the place which it 

 seems is being accorded to it, of the electrical Moles- 

 worth. G. 



