404 



SCIENCE. 



[N. S. Vol. XII. No. 298. 



considerations, and the remainder is devoted to 

 practical suggestions and directions. 



In the discussions of photometric apparatus, 

 such types have been selected as have been 

 shown by experience to be really <}seful. 

 Among these, the Bunsen screen holds easily 

 the first place, from actual use, convenience, 

 and sensitiveness, though attention might well 

 have been called to its two notable weaknesses : 



1. That it violates a fundamental principle 

 of photometric construction, namely, that the 

 portions of the photometric screen which are 

 used for comparison should be illuminated each 

 by one only of the lights to be compared, and 

 not by both. The violation of this principle 

 renders it possible, as is shown in the analytic 

 discussion, to make settings in anyone of three 

 ways, which may give quite different readings, 

 so that agreement is only obtained (and not 

 surely even so) by reversing the instrument. 

 How many users of the Bunsen screen for in- 

 dustrial purposes habitually reverse their pho- 

 tometers ? 



2. That the ordinary binocular use of this in- 

 strument is attended by the possibility of a 

 considerable constant error. This is indeed 

 pointed out on page 210, but is of sufficient im- 

 portance to deserve mention in the description 

 of the photometer itself. 



It is questionable also whether the old shadow 

 photometer is not too hardly dealt with. The 

 illustration on page 54, though similar to that 

 generally given in books on the subject, affords 

 no idea of the proper use of the instrument. 

 When arranged in the most advantageous man- 

 ner this photometer becomes convenient in use 

 to an extent hardly approached by any other 

 form, and sufficiently sensitive for most work. 



The bolometer, as a photometer, is dismissed 

 with a few lines, yet it is worth noting that 

 while energy measurers — like the bolometer — 

 which can be made to register their results 

 mechanically, do not measure the physiological 

 sensation of light, yet for certain purposes they 

 may be most useful. The variation in bright- 

 ness of a light, within not too large limits, 

 takes place generally without changing materi- 

 ally the character of the light, and hence is 

 proportional to the corresponding change in 

 energy. Such questions as the steadiness of a 



standard can be investigated by means of a bo- 

 lometer with far more precision than by any 

 photometric arrangement. No photometric in- 

 dictment against the standard candle has ever 

 approached in severity the curves obtained by 

 Nichols and Sharp, in the work referred to by 

 the author. 



The method is recommended in the chapter 

 on arc light photometry, of calibrating an in- 

 candescent lamp at white heat, by comparing 

 in succession lights of higher and higher in- 

 candescence, starting with the ordinary yel- 

 lowish standard, until through a series of steps 

 the required limit is reached. This is a ques- 

 tionable method in practice. As the change of 

 color in the successive steps is always in the 

 same direction, from yellow toward white, 

 errors made on account of the differing colors 

 of the lights are likely to be always in the same 

 direction, and therefore cumulative. I have 

 found it very difficult to make a series of 

 measurements of this kind tally in their final 

 results with a direct comparison between the 

 limits of the series made with a flicker pho- 

 tometer. 



But these are small questions and affect but 

 little the value of a book which may be recom- 

 mended to students of the subject as a safe and 



efficient guide. 



Frank P. Whitman. 



LIVERPOOL MARINE BIOLOGICAL COMMITTEE'S 

 MEMOIRS. 



Numbers II. and III. of the Liverpool 

 Marine Biological Committee's memoirs have 

 recently come to hand. It was hardly to be 

 expected that the standard of scientific excel- 

 lence set by No. I. of the series, on Ascidia 

 (see Science, January 19, 1900), written bythe 

 most experienced ascidiologist living, could be 

 reached by all succeeding numbers. If, how- 

 ever, the two now under review may be ac- 

 cepted as establishing the quality of those that 

 are to be prepared by specialists less distin- 

 guished than is Professor Herdman, the writer 

 of the first number and editor of the series, a 

 set of very valuable little books is to be the 

 outcome of this unique undertaking. 



Their usefulness will be by no means re- 

 stricted to English laboratories of elementary 



