July 4, 1889J 



NATURE 



235 



attention states this amount for the different colours. If we 

 use a piece of quartz so thick that it rotates any particular tint 

 just 90°, that tint will be cut off by the crossed analyzer, and all 

 others will — in greater or less proportion — be transmitted, so 

 that the resulting tint will be complementary to that cut off. 

 For example, a slice just so thick as to twist yellow waves round 

 90° must be 375 millimetres thick. (I may remark, for the 

 benefit of those who think it easier to express this exact thickness 

 in fractions of a British inch, that the quartz which rotates 

 yellow light 90° must have a thickness equal to one-eighth, plus 

 three-sixteenths of an eighth, plus one sixty-fourth of an eighth of 

 an inch.) When such a quartz is placed between the crossed Nicols, 

 the light shown is yellow ; but if placed between parallel Nicols 

 (i.e. in the bright field), it shows a rich purplish-violet colour, 

 the complementary of the yellow. This particular tint Biot 

 found to be excessively sensitive, the smallest inaccuracy in 

 adjustment between the prisms at once producing a change, the 

 colour appearin'4; too red or too blue, according to the direction 

 in which the analyzer has been turned out of exact adjustment. 

 This tint is accordingly known as the "transition tint" or 

 "sensitive tint," its accurate definition being due to the fact 

 that the human eye is more sensitive to the presence or absence 

 of the complementary yellow than to any other tint in the whole 

 spectrum. If we take, however, a quartz plate twice as thick as 

 this — namely, 1\ millimetres thick — this will give the yellow 

 light a torsion of 180". Hence this gives the purple transition 

 tint in the dark field, and is yellow in the bright field. A quartz 

 plate 115^ millimetres thick gives again a transition tint in the 

 bright field. I shall recur presently to the question of the 

 transition tints of the several orders. 



One of the familiar facts in this subject is that there are two kinds 

 of quartz crystals, optically alike in every other respect, differing 

 only in this, that one kind produces a right-handed twist, the 

 other kind a left-handed twist. All the pieces of quartz I have 

 so far employed are right-handed specimens. I now introduce 

 two small slices of crystal, each 34 millimetres thick, giving the 

 yellow tint when the Nicols are exactly crossed, but you will 

 notice that when we are using the right-handed crystal, the tint 

 grows red<lish as the analyzer is turned towards the left, and 

 greenish when the analyzer is turned towards the right ; whereas, 

 when I substitute the left-handed slice, the tint grows greenish 

 as the analyzer is turned toward the left, and reddish when it is 

 turned toward the right. If the analyzer is turned through an 

 exact right-angle, we get an extinction of the yellow light, the 

 remaining blue and red rays combining to give us the purple 

 transition tint. 



You will have noticed that the way in which we have 

 (approximately) measured the angle of rotation has been first to 

 set the analyzer to extinction, then to introduce the substance 

 which has the property of rotating the beam, then to turn the 

 analyzer again to extinction, and read off its angle. For, of 

 course, the angle through which the analyzer is turned measures 

 the angle through which the plane of polarization has been 

 turned. 



It is possible, however, to show in the lantern something like a 

 more obvious rotation of the light by introducing between the 

 Nicols a crystal star, built up of radial pieces of mica, twenty- 

 four in number (Fig. 8). You see in the bright field a white 

 cross with black sectors at 45°. Or, in the dark field we have a 

 black cross with vertical and horizontal arms, the sectors next to 

 those that are black seeming dusky. If now I put in a quartz 

 plate between the star and the analyzer, you see the cross shift 

 round, and it shows colours, because the blue rays are twisted 

 round more than the green, the green than the yellow, the yellow 

 than the red. Repeating the experiment with the 375 milli- 

 metre quartz which turns yellow waves round just 90°, we get 

 this gorgeous radiation of colours, and our black cross is turned 

 into a yellow one. With the 7*5 millimetre quartz, the black 

 cross is replaced by one of " transition " tint. 



The black crosses seen in certain sections of natural crystals, 

 sphseroliths, sections of stalactites, crystallizations of salicme 

 and of Epsom salts, may also be used instead of the 24-rayed 

 star of mica. But best of all I find to be the beautiful black 

 cross which is seen by polarized light in the prepared crystalline 

 lens taken from the eye of a fish. You notice how, when the 

 fish lens is projected and the quartz introduced, the cross turns 

 round. 



This is, however, a rough-and-ready way of displaying the rota- 

 tion, and it is of vast practical importance that precise methods 

 of measuring the angle of rotation should be available — of vast 



importance, because in several large industries this optical method 

 is applied as a species of handy analysis. I have named a solu- 

 tion of sugar as being an " active " substance. In the industry 

 of sugar-refining, as in that of brewing, the strength of sugar in' 

 the liquids is directly measured by measuring its optical effect. 

 Consequently there has been developed a special instrument, the 

 polarinieter, for this express purpose. 



I have here examples of several practical forms of polari- 

 meters ; there are diagrams of several more upon the walls. 



The problem of finding the best polarimeter naturally leads to 

 the inquiry what special means are there for making the 

 observation of the angle more precise than by merely observing 

 the extinction of the light, its restoration when the active substance 

 is interposed and the subsequent renewal of extinction when the 

 analyzing prism is turned. 



Biot considered that much greater accuracy could be attained 

 by watching for the restoration of the sensitive tint than by 



Fig. 8. — Mica disk of twenty-four rays, showing black cross in the dark field. 



watching for the mere restoration of extinction of the light. 

 Accordingly we will use the plate of quartz 7*5 millimetres thick, 

 giving the purple tint, to enable us to measure the rotation 

 produced by the tube of sugar solution which is now inserted 

 in the beam of polarized light. You notice how the tint has 

 changed. But 1 have only to turn the analyzer to an amount 

 equal to that to which the light has been twisted by the sugar, and 

 again I obtain the sensitive transition tint. 



The eye is not always, however, alive to minute changes of 

 colour in a single coloured patch ; it much more readily distin- 

 guishes a minute difference between two tints when both are 

 present at once. Hence Soleil devised the well-known biquartz 

 arrangement, consisting of two pieces of crystal, equal in thick- 

 ness but possessing opposite rotations. You will notice how the 

 slightest inaccuracy in placing the analyzer causes the two halves 

 of the field to differ in tint. This is especially marked when the 

 tint chosen is the transition purple. 



( To be conlitttced. ) 



THE PLANET URANUS. 



n^HAT anomalous section of the solar system, Uranus and its 

 -*- satellites, offers yet a wide field of investigation to astro- 

 nomical specialists of all kinds. Its figure, its rotation, satellites, 

 and physical constitution altogether, are moot points, which 

 should be more or less settled with the increased optical, power 

 now at our disposal. 



The circumstances attendant upon the discovery of Uranus- 

 in 1781 (Phil. Trans., 1781, p. 492) are matters of common, 

 knowledge. The planetary nature of the supposed comet seems 

 to have been first suspected by Maskelyne, and it was this sug- 

 gestion that induced Lexell to calculate for it a circular orbit in 



1 781 (Grant's " History of Physical Astronomy," p. 274). The 

 elliptic elements of the planet were first calculated by Laplace in 



1782 {Mem. Acad, dcs Sciences, Paris, January 1783). 



The first measures of the Uranian diameter were singularly 

 incongruous. Herschel found it in March 1781 to be 2"*53 ; 

 whilst in the following month he measured it as 4" "5 2 and 5" '2, 

 (Phil. Trans., 1781, p. 494). A like discordance occurred in 

 the results obtained by other astronomers. Maskelyne fixed the 

 magnitude of the apparent diameter as 3", whilst Mayer, of 



