XVI .
Description of a reflective Goniometer .
By William Hyde Wollaston , M. D. Sec. R. S , Read June 8 , 1809 .
FrRoM the advances that have been made of late years in crystallography , a very large proportion of mineral substances may now be recognised , if we can ascertain the angular dimensions of their external forms , or the relative position of those surfaces that are exposed by fracture .
But though the modifications of tetrahedrons , of cubes , and of those other regular solids , to which the adventitious aid of geometry could be correctly applied , have been determined with the utmost precision , yet Lit has been often a subject of regret , that our intruments for measuring the angles of crystals are not possessed of equal accuracy , and that in applying the goniometer to small crystals , where the radius in contact with the surface is necessarily very short , the measures , even when taken with a steady hand , will often deviate too much from the truth to aid us in determining the species to which a substance belongs .
A means of remedying this defect has lately occurred to me , by which in most cases the inclination of surfaces may be measured as exactly as is wanted for common purposes , and when the surfaces are sufficiently smooth to reflect a distinct image of distant objects , the position of faces only z of DcccIX .
aL I an inch in breadth may be determined with as much precision as those of any larger crystals .
For this purpose , the ray of light reflected from the surface is employed as radius , instead of the surface itself , and accordingly for a radius of -5 of an inch , we may substitute either the distance of the eye from the crystal , which would naturally be about twelve or fifteen inches ; or for greater accuracy we may , by a second mode , substitute the distance of objects seen at a hundred or more yards from us .
The instrument which I use , consists of a circle graduated on its edge , and mounted on a horizontal axle , supported by an upright pillar ( Plate XI ) .
This axle being perforated , admits the passage of a smaller axle through it , to which any crystal of moderate size may be attached by a piece of wax , with its edge , or intersection of the surfaces , horizontal and parallel to the axis of motion .
This position of the crystal is first adjusted , so that by turning the smaller axle , each of the two surfaces , whose inclination is to be measured , will reflect the same light to the eye .
The circle is then set to zero , or 180o , by an index attached to the pillar that supports it .
The small axle is then turned till the further surface reflects the light of a candle , or other definite object to the eye ; and , lastly , ( the eye being kept steadily in the same place ) the circle is turned by its larger axle , till the second surface reflects the same light .
This second surface is thus ascertained to be in the same position as the former surface had been .
The angle through which the circle has moved , is in fact the supplement to the inclination of the surfaces ; but as w54 the graduations on its margin are numbered accordingly in an inverted order , the angle is correctly shown by the index , without need of any computation .
It may here be observed , that it is by no means necessary to have a clean uniform fracture for this application of the instrument to the structure of laminated substances ; for since all those small portions of a shattered surface , that are parallel to one another ( though not in the same plane ) , glisten at once with the same light , the angle of an irregular fractulre may be determined nearly as well , as when the reflecting fragments are actually in the same plane .
In this method of taking the measure of an angle , when the eye and candle are only ten or twelve inches distant , a small error may arise from parallax , if the intersection of the planes or edge of the crystal be not accurately in a line with the axis of motion ; but such an error may be rendered insensible , even in that mode of using the instrument , by due care in placing the crystal ; and when the surfaces are sufficiently smooth to reflect a distinct image of objects , all error from the same source may be entirely obviated by another method of using it .
For this purpose , if the eye be brought within about an inch of the reflecting surface , the reflected image of some distant chimney may be seen inverted beneath its true place , and by *I cannot omit mentioning , that Mr. SOWERBY had thought of employing refeclion for this purpose , nearly at the same time as myself ; but did not succeed to his satisfaction , in consequence of an attempt to fix the position of the eye .
For when the line of sight is determirjed by a point connected with the apparatus , the radius employed is thereby limited to the extent of the instrument , and the error from parallax is manifestly increased .
LI1 2g .
-turning the small axle may be brought to correspond apparently with the bottom of the house ( or with some other distant horizontal line .
) In this position the surface accurately bisects the angle , which the height of that house subtends at the eye ( or rather at the reflecting surface ) ; then , by turning the whole circle and crystal together , the other surface , however small , may be brought exactly into the same position ; and the angle of the surfaces may thus be measured , with a degree of precision which has not hitherto been expected in goniometry .
The accuracy , indeed , of this instrument is such , that a circle of moderate dimensions , with a vernier adapted to it , will probably afford corrections to many former observa- : tions .
I have already remarked one instance of a mistake that prevails respecting the common carbonate of lime , and I am induced to mention it , because this substance is very likely to be employed as a test of the correctness of such a goniometer , by any one who is not convinced of its accuracy from a distinct conception of the principles of its construction .
The inclination of the surfaces of a primitive crystal of carbonate of lime is stated , with great appearance of precision , to be lo1 e8 ' 40 " : a result deduced from the supposed position of its axis at an angle of 45 ?
with each of the surfaces , and from other seducing circumstances of apparent harmony by simple ratios .
But however strong the presumption might be that this angle , which by measurement approaches to 45 ?
, is actually so , it must nevertheless be in fact about 450o o ' ; for I find the inclination of the surfaces to each other is very nearly , if not accurately io5 ?
, as it was formerly determined to be by HUYGENs ; * and since the measure of the superficial angle given by Sir ISAAc NEWTON corresponds withi this determination of HUYGIENS , his evidence may be considered as a further confirmation of the same result ; for it may be presumed , that he would not adopt the measures of others , without a careful examination .
IN THE ANNEXED PLATE , a b. Is the principal circle of the goniometer graduated on its edge .
cc .
The axle of the circle .
d. A milled head by which the circle is turned .
e e. The small axle for turning the crystal , without moving the circle .
f. A milled head on the small axle .
g. A brass plate supported by the pillar , and graduated as a vernier to every five minutes .
h. The extremity of a small spring , by which the circle is stopped at 180 ?
, without the trouble of reading off .
ii and kk .
Are two centers of motion , the one horizontal , the other vertical for adjusting the position of a crystal : one turned by the handle 1 , the other by the milled head m. The crystal being attached to a screw-head at the point n ( in the center of all the motions ) , with one of its surfaces as HuyYGcEII Opera Reliqua , Tom. I. p. 73 -Tract .
de Lumine .
NEWTON 's Oiptics , 8vo .
p. 329 .
Qu .
25 , concerning Iceland Crystal .
nearly parallel as may be to the milled head m , is next rendered truly parallel to the axis by turning the handle I till the reflected image of a horizontal line is seen to be horizontal .
By means of the milled headf , the second surface is then brought into the position of the first , and if the reflected image from this surface is found not to be horizontal , it is rendered so by turning the milled head m , and since this motion is parallel to the first surface , it does not derange the preceding adjustment .

