May 25, 1883.] 



♦ KNOWLEDGE ♦ 



r.o.j 



waves emitted from a self-luruLnous surface should diminish 

 in intensity as the sine of the angle between them and the 

 surface. For let A B, A' B', be two spaces on the self- 

 luminous surface, represented at two positions at 8 »• and 

 S' «', viewed by the eye at E and E' respectively, and 

 appeariny as so seen to be equally large circles, the real 

 circle, A B, being viewed squarely, the oval, A! T>\ being 

 viewed obliquely. Then, obviously. A' B' is larger than A B 

 in the same proportion that A' B' exceeds a B'. Hence, if 



Fig. 1. 



waves proceeded as freely from every point in A'B' towards 

 E' as from every point in A B towards E (a pencil from a 

 point in each space is shown in the figure), the eye, E', 

 would receive more light from the space A' B' than the 

 eye E receives from the space A B ; and as the two spaces 

 appear equally large to these respective eyes, the space 

 A' B' from which most light came would appear the 

 brightest, in the same proportion that it exceeds the space 

 A B. But as it only appears equally bright, the light 

 waves proceeding from each point of it towards E' must be 

 smaller than those from each point of A B towards E in 

 the proportion that the space A B is less than the space 



A' B' : that is, in the ratio B' re' to B' A'. But 



B' A 



is the sine of the angle B' A' a, which may be taken as 

 the average angle which the course of light from the space 

 A' B' to E makes with the plane S' s. 



The second point to be noticed in self-luminous surfaces 

 is that they appear equally bright at all distances, unless 

 seen through an absorptive medium. This law may be 

 demonstrated experimentally ; but in reality it may be 

 regarded as resulting directly from the nature of light. 



Thus, let E and E' (Fig. 2) be two eyes at different dis- 

 tances, viewing the same self-luminous surface represented 

 at S .s and S' s. Then, obviously, the light pencil from any 

 point P in S .s to the more distant eye E is smaller than 

 the light pencil from a point P' in S' *' to the nearer eyeE' 

 in the proportion that the square of P' E' is less than the 

 square of P E. For the cross-section of the former pencil 



at e, such that P e is equal to P' E', is less than Uic crons- 

 section at E (the pupil of the eye), that is, than the crosb 

 section of the pencil P' E' at E', as the square of P« 

 ii less than the square of P E. In other words, 

 the cross-sections of the two pencils, at equal dis- 

 tances from the luminous surface, are inversely as the 

 squares of the distances from wiiich the eye views the 

 surface. But it is equally obvious that the apparent ei?.e 

 of the surface is reduced in precisely the same degree with 

 distance. Thus, the total quantity of light received from 

 the surface is reduced with increase of distance, in pre- 

 cisely the same ratio that the apparent area of the surface 

 is reduced. But this amounts to saying that the surface 

 appears equally bright at ail distances. However, the 

 matter can be readily demonstrated by experiment. Thus, 

 if two globes of iron of unequal si>:e be heated to the 

 same degree of redness, and if they be then placed in a 

 dark room at such distances from the eye as to appear 

 equally large, it will be found that they appear absolutely 

 alike in all respects. Again, if two candles of the same 

 kind be placed at unequal distances in clear air, it will be 

 seen that the flames, though they seem unequal in size, 

 remain apparently equal in intrinsic luminosity. 



Next as to opaque bodies shining by reflected light. 



First, both the laws just stated for self-luminous bodies 

 hold also for illuminated opaque bodies. The second law 

 admits of being tested more thoroughly in the case c£ 

 illuminated than in that of self-luminous bodies. Thus, 

 from observations of the planets, it has been found that 

 the quantity of light received from the same planet at 

 different distances, but in other respects similarly situated 

 (as from Mars when in opposition near perihelion and near 

 aphelion), varies inversely as the sciuare of the distance, 

 which is, of course, precisely the ratio in which the ap- 

 parent dimensions of the disc vary. 



The two laws may be combined into one simple and 

 easilj'-remembered statement : — A surface, tc/ielher self- 

 luminous or illuminated, appears in all positions and all 

 distances Just as brir/ht as it is iu realiti/. Of course, this 

 law does not take into account the absorption exercised 

 by any medium through which a luminous object may be 

 viewed. 



(To ie continuod.) 



Zinc Co.vtixg for Iron. — Attention li.is recently again. 

 been drawn to MM. Neugean and Delaite's process of pro- 

 tecting iron surfaces against rust. A very tine powder of 

 metallic zinc is mixed with oil and a siccative, and applied 

 to the iron by means of an ordinary brush. In many cases 

 one coat is sufficient ; two coats are at any rate guaranteed 

 to secure a protection against the corrosive action of the 

 atmosphere as well as of sea water. The zinc coating gives 

 the iron a steel-grey appearance, and it does not interfere 

 with subsequent painting. !MM. Neuj;ean and Delaite 

 received a diploma at the Paris Electric Exhibition of 1881, 

 and now recommend their process for iron structures, 

 bridges, lamp-posts, itc, and also for iron ships. If this 

 process really affords the piotection it claims, nothing need 

 be said in recommendation of it, since it can hardly be 

 surpassed in simplicity and cheapness, and is capable of 

 application in cases where galvanising, the Bower-Barft', 

 and similar processes would hardly be piacticable. A good 

 nii.xtiire, of which only the necessary quantity ought to be 

 prepared, consists of S parts by weight of zinc, 71 of oil, 

 and 2 of a siccative. — Enyineerimj. 



Copies of the Index to Vol. I. Knowledge wanted. Apply or 

 address, the Publisher, 7-1 to 76, Great Queen .Street, London, ^V.C. 



