December 9, 1909] 



NA TURE 



173 



are equally bright ; if they are not equally bright it cannot 

 say with any kind of proportionality what their relative 

 brightnesses are. All photometry depends on the perception 

 of an equality. 



Photometers for the measurement of illumination have 

 been mentioned earlier as coming first into notice in the 

 'eighties. One of the earliest in this country was that 

 constructed by Sir William Preece, with the assistance of 

 Mr. A. P. Trotter, for measurement of the illumination of 

 side-walks and pavements of streets. It has been sub- 

 sequently developed by Mr. Trotter, and as constructed 

 by Mr. Edgcombe is a most useful and handy instrument, 

 telling the amount of illumination directly in terms of the 

 tandle-foot. Another, by Mr. Haydn Harrison, measures 

 <he illumination, not on the horizontal, but at 45°. Almost 

 •■qually early with the Preece-Trotter illumination photo- 

 meter was the school photometer of Petruschewsky, appar- 

 ently little known in this country. Most recent of this 

 sort is the form due to Martens. 



The principles and construction of photometers are 

 matters that have interested me for nearly thirty years. 

 About 1880 I brought out a form of wedge-photometer 

 (modified from Ritchie's form), in conjunction with Mr. 

 C. C. Starling, for electric light measurements. Later I 

 gave to the Physical Society an investigation of the errors 

 arising in photometry from the almost universal assump- 

 tion that the law of inverse squares is fulfilled. In 1882, 

 when lecturing at the Crystal Palace Exhibition, I gave 

 diagrams to show the effect of the superposition of illumina- 

 tion from two or more lamps, and discussed the variations 

 of illumination in a street between the places of maximum 

 and the places of minimum illumination. Twelve years 

 ago I described a tangent photometer, which has remained 

 a mere optical curiosity. 



No one can have worked at the photometry of modern 

 lamps, or of the illumination of surfaces lit by lamps, 

 without becoming conscious how much misunderstanding 

 there is of the elementary laws of illumination. There is 

 Lambert's cosine law, admirable and simple if only it 

 were not in so many cases vitiated by the presence of 

 organised — that is, specular — reflection. There is the law 

 of inverse squares, itself a universal geometrical law of 

 action radiating from a point, so fatally and absolutely 

 misleading if applied to any other case than that of action 

 from a point. 



One subject on which more information is badly needed 

 is the specific brightness of surfaces of different kinds when 

 subjected to a standard illumination. For instance, how 

 much light is reflected, per square inch, when illuminated 

 with an intensity of i candle-foot, from such materials as 

 oak panelling, whitewash, brown paper, or the surface of 

 a red brick wall? Here in this theatre the walls are tinted 

 of a dark Pompeian red or maroon, which reflects but 

 little light. The extra annual expense on lighting that 

 might be saved had a lighter tint been used is surely worth 

 considering. 



The subject of diffuse reflection which h^re comes into 

 play has indeed been investigated partially by several 

 persons. There are Dr. Sumpner's researches of iSqd and 

 those of Mr. Trotter on white cardboard and other white 

 matt surfaces, but how few others! Again, there is the 

 subiect of diffuse refraction, which occurs in ground-glass 

 shades, ribbod and corrugated glass panes, and other 

 devices for diffusing the concentrated light of lamps. Yet 

 how little does any optical book tell us on the subiect of 

 diffuse refraction. Reflection and refraction as thev occur 

 at dull or irregular surfaces appear to be of no importance 

 to the academic writer of text-books of optics, but they 

 are of vital interest to the illuminating engineer, .'\gain, 

 there are a number of semi-physiological problems that 

 demand investigation and settlement. We all know that 

 our eyes have an automatic diaphragm which stops down 

 the entering light to protect our eves from glare, rendering 

 us relatively insensitive to bright lights. Does anyone 

 know whether the contraction of the pupil depends on the 

 total amount of light entering the eye or on the intensity 

 of theimage on local patches of the retina? 



.Again, we all know how an unsh.aded arc-lamp, or even 

 clow-lamp, " cuts " the eyes by the very concentration of 

 its beams, even when it mav be many feet away, while 

 vr>. 2693, VOL. 82] 



the same actual amount of light, if diffused over a greater 

 apparent surface, as by a surrounding globe of ground 

 glass, is quite readily endured, and does not produce the 

 same painful sensation. Does anyone know how great is 

 the specific brightness of surface that the eye will tolerate 

 without experiencing this discomfort? We can look at a 

 white cloud or at the blue sky without pain. Can we 

 endure a specific brightness of so much as one-tenth of a 

 candle per square inch? 



Our eyes are provided by nature with a most exquisite 

 and automatic iris diaphragm which opens in the dark 

 and closes in the light, thereby shielding us partially 

 against the evil effects of glare. Putting it in the language 

 which the photographer uses to describe the stopping- 

 down of a camera-lens, the automatic iris of our eye can 

 close the pupil so that while in a comparative darkness 

 the aperture opens to //2 or //2.5, it closes, amid a 

 biilliant surrounding illumination, to about //20. Suppose 

 we are looking out in relative darkness, and are confronted 

 with a brilliant patch shining with a specific brightness of 

 one-tenth of a candle per square inch, we shall feel a 

 certain amount of discomfort from its glare, and if we 

 regard it steadily for a second or two will, on closing our 

 eyes or turning away, see those persistent coloured images 

 that trouble us after looking at any very bright light ; 

 but now let the same brilliant patch be placed against a 

 bright background. Far more light will enter the eye ; 

 the automatic iris of the eye will in a few moments have 

 contracted, stopping down the lens of the eye so that it 

 will be far less sensitive. In these circumstances, will the 

 patch that has a specific brightness of one-tenth candle per 

 square inch pain or dazzle the eye? I ask the question, 

 but I do not know the answer. Does anyone know what 

 the answer ought to be? It is a simple question, and a 

 few experiments would soon settle it. Of course, 

 one must admit that the automatic action of the iris 

 diaphragm, important as it is, does not by any means 

 account for the whole of the facts about the want of pro- 

 portion between the intensity of a stimulation and the 

 intensity of the resulting sensation. Fechner's logarithmic 

 law of psvchophysics gives a clue, but even this does not 

 seem capable of expressing, much less of explaining, the 

 facts about the observed want of proportionality. Why 

 should a light of ten-fold brilliancv not produce a sensation 

 ten times as intense? And why should a greater brightness 

 of the general surroundings relieve us of the annoyance 

 of those coloured after-images? After-images can be seen 

 even under extremelv feeble illumination, as I have again 

 and again found. Has anyone discovered any exact law 

 govni-ning their occurrence? 



.All these Queries show that ther" is plenty of work 

 awaiting us, even in tho mere collection and completion of 

 such scattered information as is already available ; but 

 there are ev»n more important questions before us, more 

 important, not in science, but in their relation to the public 

 welfare and the economics of the community. 



Now that we have a standard of illumination and simple 

 portable instruments that will measure it, there can be no 

 excuse for inaction or ignorance in .loplyln" that knowledge 

 to securing proper illumination for public and private 

 buildings. 



Let me begin with school buildings. They are the most 

 important ; for whatever bad results flow from bad lighting 

 cf churches, factories, or railway stations, those which 

 result from the bad illumination of schools are far more 

 to be deplored — they imperil the eyesight of the next 

 generation. 



All ophthalmic surgeons agree that the cause which 

 forces the children into increasing shortsightedness is pro- 

 tracted poring over books under an insufficient illumina- 

 tion. Even in what an inspector might call a well-lit 

 school the illumination at the surface of the desk may be 

 quite insufficient if the desks are badly placed, or the 

 windows insuRiciently high, or the lamps badly distributed. 

 .All educational authorities ought henceforth to insist on 

 rational requirements as to lighting. Hitherto they have 

 had nothing definite to specify ; now that illumination 

 photometers are available, they ought to require a mini- 

 mum of \\ candlerfoot at the worst-lighted seat In the 

 schoolroom, and not depend on purely architectural rules 



