14 Expirrlmen/s on the hiJUBioii of Light. 



nearcft fii!e-liiies indicated an angle of defle£\ion equal to 7' 45"; and, from eflimate 

 founded on the appearance of the lines, he judged that the other ftrcaks were defledcd in 

 angles of double and triple the fame original angle. 



After this ingenious philofophcr had proceeded fo far, and found a ready folution for tlie 

 fads he had obferved, by referring them to the Newtonian dodrine of inflection, he went 

 on, with more h.ifte than precifion, to account for the appearance exhibited by the hand- 

 kerchief. He remarks, that, if the liandkcrcliief be held before the eye, fo that its threads 

 may be difpofed horizontally and perpendicularly, and a luminous fpot be viewed, the 

 perpendicular threads will, by inilcdion, on the jrinciple explained in liis frame of hairs, 

 cauft it to appear as a row or horizontal line of fpots, for example, five; and that tiie 

 horizontal threads will, on the fame principle, convert every one of thefe into live perpen- 

 dicular rows ; by which means, a fquare figure, compofcd of twenty-five luminous points, 

 will be fcen, of which he gives the figure. 



On thcfc lafl inductions it may be fullicient to remark, that though his frame of parallel 

 hairs might have converted a real row of luminous points into fuch a fquare, yet it is in- 

 conceivable how the fame ef!c£l could have followed, with regard to rays of light wliicli 

 muft have already pafl'ed through the handkerchief in order to receive the firfl: modifica- 

 tion. But as Mr. Rittenhoufc, from his figure of 25 fpots, which are in no cafe produced 

 bv the handkerchief, fliews that he did not ftridly attend to the phenomena, but wrote in 

 this inftance from recolle£lion and reafoning, I fliall difmifs thefe particulars of the original 

 cbfervers, and mention the facls I have myfelf remarked. 



When this problem was firft pointed out to me a few months ago by a very a£live pro- 

 moter of the fciences, 1 took the earlieft opportunity of looking at a (Irect lamp through a 

 piece of muflin containing one hundred threads in the inch. Inftead of one fpot of light, 

 there appeared nine; four of which occupied the corners of a fquare, one tlie centre, and 

 four others the middle points, in a right line between the corners. Upon examining them 

 with an achromatic perfpeclive magnifying fourteen times, I faw that the fpots were tiue 

 images of the flame of the lamp, the central one being perfeft, all the others coloured, 

 with the red outermort, and the corner lights leafl luminous and diflind. It made no 

 ditFcrence in the appearance, whether the cloth was applied clofe to the objecl-glafs, or 

 at any diftance to which the arm cculd reach before it; or moved fideways in its own 

 plane ; but the apparent dinienfions of the fquare were lefs when the cloth was applied 

 between the objed-lens and its focal image, in proportion ;by eftimate) as the didance 

 of the cloth from the focus was lefs. When coarfer mufllns were ufed, the fquare was 

 fmaller according to the coarfenefs, but in what proportion was not examined, becaufe it 

 was fuppofed to follow no very fimple ratio. No elFecl of this kind took place with the ob- 

 jcd-glafs of a microfcope. 



From thefe fafts I inferred as follows : — The middle flame is formed of all the pencils of 

 light vihich pafs through the central parts of the holes in the cloth without any defledion. 

 The four flames in the middle of the fides of the fquare, are formed by the pencils of 

 light inflcSed towards the fliort threads bounding each fquare hole in the cloth, affifted by 

 the deflective power of the oppofitc threads refpeftively. The corner flames are formed by 

 the combined adion of the two threads contiguous to the angles of every hole ; the fum of 

 which force, acting in the diagonal with a power varying from the diftance of the refpedive 

 I parts 



