290 



NA TURE 



[July 30, 1896 



upon the nature of the lights to be compared. The most 

 hopeful of all the instruments described for colour 

 photometry seems to be the mixture photometer of 

 Grossc. In this instrument the lights A and B, to be 

 compared, are estimated not by directly measuring the 

 ratio A : B, but by comparing two mixtures, A+ ;« B and 

 « A + B, where m and n are constants fixed for each 

 instrument. Here the colour difficulty is got over by 

 mixing the lights, and the relative illuminations of the 

 surfaces compared is varied, partly by adjusting the 

 distances of the lights, and partly by turning the eye- 

 piece of the instrument, the working of which depends 

 on the phenomena of polarisation. As far as one can 

 judge from the description, this photometer appears to 

 be an excellent one. 



The third chapter, on photometric standards, is a full and 

 trustworthy account of the efforts which have been made 

 at different times towards obtaining a good standard of 

 light. Its usefulness is increased by plentiful references 

 to original papers, as is the case with the other portions 

 of the book. The French work was published too soon 

 to include the final results of the work of the English Com- 

 mittee on Light Standards, but almost all previous work 

 is alluded to. A short, but good, chapter on the equip- 

 ment of photometric laboratories is succeeded by one 

 entitled "electric lights," which contains a large amount of 

 photometric matter relating to electric lighting, that has 

 not hitherto appeared in book form. 



The sixth and last chapter of the book is reserved for 

 the distribution and measurement of illumination. As 

 illumination is the whole object of artificial lighting, it 

 seems strange that the consideration of it should have 

 been so much neglected. For every photometer that has 

 been designed for the measurement of illumination, there 

 are at least twenty for the measurement of candle-power. 

 The present book is, we believe, the first to take up this 

 subject as an important part of photometry ; and the 

 ordinary reader, who has not followed the subject in the 

 technical press, will find here much which is quite new to 

 him. It is a subject of increasing importance, and since 

 the issue of the French edition of this book some valu- 

 able work on it has been done by Trotter and Blondel, 

 which should be incorporated when a new edition is 

 called for. It is somewhat remarkable that no allusion is 

 made to the diffused system of lighting by inverted arc 

 lamps, or corresponding arrangements of glow lamps, 

 in which the lights themselves are hidden fiom view, the 

 rays being directed to walls and ceilings for subsequent 

 diffusion downwards. This system has been used on the 

 continent and in this country, especially for the lighting 

 of factories, and its use is extending. It is in reality one 

 of the most efficient systems of lighting, although anyone 

 seeing it for the first time is apt to think it weaker than 

 it really is. A description of it should certainly find a 

 place in a book such as this. There is, however, a large 

 amount of new matter given in this chapter, and the only 

 fault to find is a minor one, and one for which the 

 author is not directly responsible. In quoting the dif- 

 ferent writers on this subject, a number of terms have 

 been introduced which are needless, confusing, and badly 

 chosen. Such terms as "volume of illumination" and " sur- 

 face of illumination" are quite needless, while the phrase 

 " useful effect of illumination " is a very bad and mis- 

 NO. 1396, VOL. 54] 



leading name for a quantity which is of doubtful im- 

 portance. It follows from the definition that the "useful 

 effect of illumination " is always greater than the illumin- 

 ation itself, and that it increases rapidly as the distance 

 from the light is increased, until at an infinite distance 

 from the light the "useful effect" of its illumination is 

 infinite. " This result was easy to foresee," writes the 

 author, who asserts that the criticisms to which it has' 

 been subjected "are not well founded"; and calcu- 

 lations are then made by means of which " the volume of 

 useful effect," the "surface of useful effect," and the 

 "mean useful effect" may be obtained by any one 

 anxious to know. 



In spite of a few minor blemishes, the book is a 

 thoroughly good one. It is well printed, well illustrated,, 

 and contains a mass of valuable matter which is not to be 

 found elsewhere. Some useful appendices by the trans- 

 lators conclude the work. W. E. S. 



THE NOTES OF BIRDS. 

 The Evolution of Bird-song. By Charles A. Witchell. 

 Pp. X + 348. (London : A. and C. Black, 1896.) 



THIS little work will be heartily welcomed by all 

 ornithologists as the first elaborate attempt to deal 

 in a scientific spirit with the very difficult subject of the 

 utterances of buds. Considering the great amount of 

 careful observation necessary to the formation of any 

 theory on the subject, and the difficulty of recording such 

 observation correctly and intelligibly, Mr. Witchell is to 

 be warmly congratulated on his book. It is, in fact, a 

 very welcome and agreeable call-note, addressed to h 

 brethren of the craft, and urging them to come and test 

 the flavour of the food its author has discovered. Should 

 any of them be critical of details, as indeed in such a 

 subject they inevitably must be, it is to be hoped that the 

 call-note will not change into an " alarm " ; for however 

 much we may differ from Mr. Witchell in detail, we shall 

 hardly be disposed to quarrel with the main line of his 

 argument, and we shall be grateful to him for his work as 

 a pioneer. Personally I am glad to acknowledge that 

 during the present season of song I have derived the 

 greatest benefit from this book, which fortunately appeared 

 at the very time when fresh observation was most easv 

 and agreeable. 



Starting with Darwin's theory of the origin of voice,. 

 Mr. Witchell states his belief that it was first occasioned 

 by fear or anger in combat, and that, consequently, the 

 earliest cries were alarm-cries. He then proceeds to 

 show that "the first call-notes of birds were probably 

 mere adaptations of alarm-cries, the use of which was 

 induced by the influence of mutual aid among associated 

 individuals." Rapid reiteration of call-notes had a 

 tendency to produce something in the nature of song ; 

 and Mr. Witchell is able to produce several good examples 

 of this in species which are still incapable of any true 

 vocal effort, though he- gives one or two, e.g. the willow- 

 warbler, about which I feel, as yet, uncertain. Closer 

 observation on this point is much wanted, but 1 can quote 

 one case, that of the tree-sparrow (not mentioned b\- Mr. 

 Witchell), which may occasionally be heard in the spring 

 constructing a quasi-song by linking call-notes together. 

 Mr. Witchell is careful not to commit himself to the view 



