57- 



NATURE 



[Oct. 12, 1882 



resembling the folds of a curtain passed rapidly across the 

 northern sky. The light was colourless, with occasional flashes 

 n on. H. Mellish 



ock Priory, October 9 



Newton, Wollaston, and Fraunhofer's Lines 



In most of the current treatises on spectrum analysis, on the 

 'pe, and on optics generally (Lloyd's works being 

 exceptions), injustice is done to Newton's care, and scientific 

 insight in his optical experiment-, when Wollaston's discovery 

 of the dark lines in the solar spectrum is alluded to, by most 

 positive statements to the effect that Newton never used the 

 slit, or that Wollaston was the first who ever made observations 

 011 the pure spectrum. 



That the statements are erroneous may be seen by a com- 

 parison of the following extracts from Wollaston's paper in the 

 Philosophical Transactions for 1802, p. 378, and 'Newton's 

 ''Opticks" edition of 1704. 



Wollaston concludes from his experiments that " the colours 

 into which a beam of white light is separable by refraction, 

 appear to me neither 7, as they usually are seen in the rainbow, 

 nor reducible by any means (that I can find) to 3, as some 

 persons have conceived, but that by employing a very narrow 

 light, four primary divisions of the prismatic spectrum 

 may be seen with a degree of distinctness, that, I believe, has 

 not been described nor observed before." He describes the 

 experiment as follows : — 



"If a beam of daylight be admitted into a dark room by a 

 crevice l-ioth of an inch broad, and received by the eye at a 

 distance of 10 or 12 feet, through a prism of flint glass free from 

 veins [italicised by Wollaston], held near the eye, the beam is 

 seen to be separated into the four following colours only, red, 

 yellowish, green, blue, and violet." He then describes four 

 lines marking these divisions, together with two others for which 

 he does not offer any explanation. 



Compare with this Prop. 4 of Book I. of the "Opticks," 

 which is "To separate from one another the heterogeneous 

 rays of compound light." Newton, after showing at some 

 kngth why he uses a lens to "diminish the mixture of the rays," 

 describes Experiment 11, p, 47, as follows : — 



" In the sun's light let into my darkened chamber through a 

 small round hole in my window-shutter, at about ten or twelve 

 feet from the window, I placed a lens, by which the image of 

 the hole might be distinctly cast upon a sheet of white- paper. 

 . . . Then immediately after the lens I placed a prism, by 

 v. Iiich the trajected light might be refrae'ed either upwards or 

 sideways." The "oblong image" thus formed he received 

 upon paper placed "at the just distance where the rectilinear 

 sides of the image became most distinct." By varying the size 

 of the hole, he made "the mixture of the rays in the image to 

 be as much or as little as I desired." For this purpose he 

 caused the breadth of the image to be sometimes sixty or seventy 

 times less than its length. 



"Yet instead of the circular hole, '/is better to substitute an 

 oblong hole shaped like . long parallelogram, -,i<itli its length 

 parallel to the prism. For if this hole be an inch or two long, 

 and but a tenth or twentieth part of an inch broad or nar- 

 rower ; the light of the image will be as simple as before, or 

 simpler, and the image will become much broader, and therefore 

 more fit to have experiments tried in its light than before." 



For the purpose of comparing the simpler light with the more 

 compound, he used also a hole of the shape of an isosceles 

 triangle, whose base was "about the tenth part of an inch, and 

 its height an inch or more" (the width of which, therefore, at 

 a quarter of an inch from the vertex, would be one-fortieth of 

 an inch). The refracting edge of the prism was parallel to the 

 perpendicular of the triangle. The images would therefore be 

 ''equicrural triangles," "a little intermingled at their bases, 

 ot at their vertices." 



He is very emphatic as to the precautions in making the 

 experiments. lie was always careful to have the image in the 

 position of minimum deviation — all foreign light must be care- 

 fully excluded from the chamber. The lens must be good — the 

 prism being made of " glass free from bubbles and veins," with its 

 sides truly plane and its polish elaborate." "The edges also 

 of the prism and lens, so far as they make any irregular refrac- 

 tion, must be covered with a black paper glued on." "It's 

 nit to get glass prisms fit for this purpose." 



lie did not, as is sometimes supposed, always receive the 



images on paper, for in Expt. 4, Prop, ii., p. 22, he says: "I 

 looked through the j>rism upon the hole." 



That with good prisms, and the great variety of experiments 

 which he must have tried, he did not see the dark lines by look- 

 ing through the prisms, seems remarkable. It may possibly be 

 explained by the fact that in the very class of experiments in 

 which he was most likely to discover the lines (and in which 

 Wollaston actually discovered them), he found himself obliged 

 to rely on the observation of an assistant. This is mentioned on 

 p. 92 in Prop. iii. of the second part of Book I. The proposi- 

 tion is "To define the refrangibility of the several sorts of 

 homogeneal light, answering to the several colours." In this he 

 says: " I delineated therefore in a paper the perimeter of the 

 spectrum" . . . "and held the paper so that the spectrum 

 might fall upon this delineated figure, and agree with it exactly, 

 whilst an assistant, whose eyes for distinguishing colours were 

 more critical than mine, did by right lines drawn across the 

 spectrum note the confines of the colours." 



Alexander Johnson 



McGill College, Montreal, Canada, September 19 



The Spectroscope and Weather Forecasting 



I MUCH regret that Prof. Smyth should have taken the word 

 prognostic, applied to the rain-band as a depreciatory epithet, 

 when it was only intended as a term of classification. In com- 

 mon parlance any particular " look " of the sky is called a 

 progi ostic, and it is a natural extension of the idea to call the 

 "look" of the sky absorption a spectrum prognostic also. 



The question at issue is this. Assuming that the 1 

 is a quantitative measure of the amount of vapour in a section 

 of the atmosphere, why is it of more use in forecasting than the 

 numerous sky and other prognostics which also indicate an exces- 

 sive amount of vapour, as, for instance, sweating walls, or a soft sky. 

 Like them, it precedes rain in certain cases, and for the same 

 reason ; like them, it fails in numerous cases where rain falls 

 without being preceJed by excessive vapour quantity ; and, like 

 them, it cannot compare in forecasting value with synoptic 

 observations over a large area, which correlate moist currents 

 w ith isobaric lines. 



But there is one case in which the rain-band may give valuable 

 information — when we have a vapour-laden upper current over 

 a dry surface wind. This often occurs in winter, with a warm 

 south-west current over an area of frost and an east wind. In 

 practice this almost invariably makes itself visible by the long 

 converging stripes of cirrus which so often precede a rainy thaw, 

 but still cases may occur when no cirrus is formed, or it is other- 

 wise invisible. Here is a case in point. One spring morning 

 in London there was a thick fog, with a south-west wind. About 

 1 a.m. the wind shifted to east, the pavement remaining white 

 and dry ; when, to my surprise, the ordinary spectrum of a fog 

 was crossed by a strong rain-band. Two hours afterwards a 

 few big drops of rain fell, which soon ceased, and the wind- 

 returned to the south-west. 



The 8 a.m. chart showed London to be then on the northern 

 edge of an anticyclone, with a small secondary cyclone over 

 Devonshire ; this moved on during the day, bringing rain with 

 it, which soon passed. 



We can now estimate the different values of the several indi- 

 cations. Cirrus, if there is any, tells that a moist south-wesfc 

 upper current has set in, but not if it is specially vapour-laden.* 



The rain-band tells us simply that there is an abnormal amount 

 of vapour somewhere, and roughly measures it ; by inference^ 

 from the dryness of the ground, we know that the vapour mus( 

 be overhead : in very rare cases the band would speak before! 

 the cirrus, and in any case would show unusual vapour, which 

 the cirrus could only suggest by looking softer than usual. Oib 

 the other hand, the spectroscope would be silent in numberlcst 

 cases where cirrus would indicate rain correctly, and neither 

 could tell their story till the vapour-bearing current had set in. 



'ihe forecaster, who used synoptic charts, would know that a 

 damp south-west wind always blew over the north-east wind in 

 front of a cyclone, and in a case like this could say that as the 

 secondary approached, the moist upper current would set in some 

 hours before it arrived, and would be so far ahead of anjj 

 prognostic ; but he would have no means of saying if the current) 

 was extra vapour-laden or not. 



Hut in most cases a knowledge of the fact would be but of. 

 little service to him. Suppose that in this case there was rain at 

 Plymouth, cirrus or rain-band at Portsmouth, and blue sky over 



