XXXVIII .
A Letter from Mr. T. Melvil to the Rev. James Bradley , D. D. F. R. S. With a Discourse concerning the Cause of the different Refrangibility of the Rays of Light .
SIR , THO ' I have not the honour of your acquaintance , I have presumed to address the enclosed paper to you , relating to optics and astronomy .
The unwearied zeal which you have shown , Sir , for the improvement of these sciences , made me conclude , that whatever has an appearance of truth or novelty on such subjects , tho ' inconsiderable in itself , might not be unacceptable to you .
Besides , among all the societies of the learned in Europe~ I could think of no abler judge to consult on these matters , than the author of the ` ` Aberration of Light , and of the Variation of the Precession of the Equinoxes : ' ' the two greatest discoveries , without doubt , that have been made in astronomy for half a century .
Perhaps you may be able , Sir , from the many accurate observations you have already made , to decide the question relating to the velocity of light ; or , at least , to determine , whether the observation proposed be practicable : for I am very sensible , that many observations , which appear easy in speculation , cannot be put in execution .
Read March 1 , 1753 .
If you find any thing in the enclosed worthy of the public view , you are at liberry to lay it , in whole , or in part , before the Royal Society , or to make what use of it you please .
I am , SIR , Geneva , Feb. 2 , 1753 .
Your most obedient humble servant , T. Melvill .
Concerning the Cause of the dsfferent Refrangibility of the Rays of Light , by Mr. T. Melvill .
1 .
IN order to account for the different refrangibility of the differently-colour 'd rays , Sir Isaac Newton ~ , and several of his followers , have supposed , that their particles are of different magnitudes or densities ; but if there be any analogy between the refractive power and gravity , it will produce equal velocities in all particles , whatever their magnitude or density be ; and so all sorts of rays would be equally bent from their right lined direction .
Read March 8 , 1753 .
2 .
It seems therefore a more probable opinion , which others have advanced , that the differentlycolour 'd rays are projected with different velocities from the luminous body ; the red with the greatest , violet with the least , and the intermediate colours with intermediate degrees of velocity ; for , on this hypothesis it is manifest , that they will be differently refracted ~ Newton 's Optics , Query 29 .
refracted in the prismatic order , according to observation .
3 .
On supposition , that the different refrangibility of light arises solely from the different velocities of the rays before incidence , these velocities must be to one another nearly as their sines of refraction .
I say nearly ; for their exact proportion cannot be discovered , but by the solution of the following problem , which I take this opportunity of proposing to the learned : If two bodies fall , in equal angles of incidence , on a space terminated by parallel planes , in which any power acts perpendicularly to the planes ( according to the hypothesis in Prop. 94 , Lib. I. of the Principia ) the ratio of the sines of emergence to the common sine of incidence , and consequently to one another , being given , to determine the proportion of their velocities at the time of their incidence on the first plane .
4 .
Their velocities in any given medium ( suppose air ) being once determined , their velocities in any other may be easily discovered ; for they are to those in air as the sine of incidence to the sine of refraction , when the ray passes from air into the other medium ~ .
5 .
While the differently-colour 'd rays are supposed to move with one common velocity , any pulses , excited in the ethereal medium , must overtake them at equal ~ Principia , Phil. Not .
Prop. 95 , Lib. I. equal distances ; and therefore the intervals of the fits of reflexion and transmission , if they arise in this manner , as Sir Isaac conjectures , would be all equal : but if the red move swiftest , the violet slowest , and the intermediate colours with intermediate velocities , it is plain , that the same pulses must overtake the violet soonest , the other colours in their order , and last of all the red ; that is , the intervals of the fits must be least in the violet , and gradually greater in the prismatic order , agreeably to observation .
6 .
Let c denote the velocity of the ethereal pulses , V the velocity of red light , and U that of violet ; I and J the intervals of their fits , and D the distance between two succeeding pulses : it is plain , from the nature of Newton 's hypothesis , that I is to D , as V to C-V : and again , D to J as C-U to U : therefore , ex aequo , I is to J , as CV-VU to CU-VU , from which we have the equation C=~-JxVU/ IxU-JxV Therefore , as the proportion between the intervals of the fits in red and violet , can be assigned by experiment , and the proportion of their velocities in any medium likewise , by Art .
4 .
the velocity of the ethereal pulses may be easily computed .
The velocities of the red and violet in air are nearly as 78 and 77 .
In the celestial spaces they are less , but almost in the same proportion ; the intervals of their fits are by experiment as 100 and 63 ~ , from whence , by the canon now laid down , the velocity of the ethereal pulses in the celestial space is found to be to that of red ~ Newton 's Optics , Lib. II .
Part I. Obs. 14 .
red light as 79763 to 78000 .
As light moves from the sun to us , by Dr. Bradley 's accurate estimation , in 8~ 12~ ( a ) , the pulses of the ethereal fluid must be propagated thro ' the same space in about 8~ 1~ .
7 .
Hence also may be determined , in known measures , the distance between two succeeding ethereal pulses for D= CI-VI/ V. 8 .
Upon the hypothesis of the different velocities of different colours , we may understand , at least in general , the reason of the strange analogy , discovered by Sir Isaac , between the intervals of the fits , and the spaces occupied by the several colours in the spectrum ( a thing hitherto unexplained ( b ) ; since , from the velocities of the several rays , upon which depend the intervals of the fits , as has been now explained , arise likewise their several degrees of refrangibility .
9 .
But , as it is of great consequence in philosophy , to distinguish between facts and hypotheses , however plausible , I observe , that the various refrangibility , reflexibility , and inflexibility , of the different colours , and their alternate dispositions , at equal intervals , to be reflected and transmitted , which are the whole ground-work of the Newtonian system , are to be considered as undoubted facts , deduced from experiment ; but that the velocities of different rays are different in the manner now described , is no more than probable conjecture : and tho ' this point should be decided , by a method that we are now to propose , it ( a ) Eames 's Abridge .
Transact. Vol. VI .
p. 157 .
( b ) Compare Newt .
Opt .
Book I. Part 2 .
Prop 3 .
with Book II .
Part 3 .
Prop. 16 .
it would still remain uncertain , whether the fits of reflexion and transmission consist in an alternate acceleration and retardation of the particles of light , or in something else .
For instance , it might be supposed , that every particle of light has two contrary poles , like a loadstone ; the one of which is attracted by the parts of bodies , and the other repelled ; and that , besides their uniform rectilineal motion , the particles of differently-colour 'd rays revolve in different periods round their centers ; for thus their friendly and unfriendly poles being alternately turned towards the surfaces of bodies , they might be alternately disposed to reflexion and transmission , and that at different intervals , in proportion to the periods of their rotation .
Lastly , tho ' it were proved , that the fits do proceed from an alternate acceleration and retardation of the particles of light , it would still be no more than probable conjecture , that this is brought about by pulses excited in the ethereal medium .
Nay there are some circumstances in these phenomenon , that seem hardly intelligible by that hypothesis alone : as , Why the intervals of the fits are less in denser mediums ( a ) ; and , Why they increase so fast , and in so intricate a proportion , according to the obliquity of incidence ( b ) .
10 .
By Dr. Bradley 's beautiful theory of the aberration of light , the stars appear to be removed from their true place to a certain distance , according to the proportion which the transverse motion of the spectator 's eye bears to the velocity of light .
It is plain therefore , ( a ) Newt .
Opt .
Book II .
Part 3 .
Prop. 17 .
( b ) Prop. 15 .
ibidem .
therefore , that , on our hypothesis , any star must have a different apparent place for every different colour ; that is , its apparent disk must be drawn out by the aberration into a longitudinal form , resembling the prismatic spectrum , having its red extremity nearest its mean place .
In the stars situated about the pole of the ecliptic , its length should continue always the same , tho ' directed along all the different secondaries of the ecliptic in the course of a year : but in those which lie in or near the plane of the ecliptic , it should be greatest at the limits of the eastern and western aberrations , the star recovering its colour and figure , when the true and mean places coincide .
But there is no hope of discovering , whether our system be true or false , by this consequence of it : for the greatest length of the dilated disk being to the whole aberration , as the difference of the velocity of the red and violet to the mean velocity of light , i. e. but about a 77th part of it , cannot much exceed the fourth part of a second .
11 .
The time which the extreme violet light takes in arriving from any distance to the eye , will be to that which the extreme red takes in coming from the same , as 78 to 77 .
If Jupiter be supposed in a quadrate aspect with the sun , in which position the eclipses of his satellites are most commodiously observed , his distance from the earth being nearly equal to his distance from the sun , light takes about 41~ in passing from him to the earth ; therefore the last sensible violet-light , which the satellite reflects before its total immersion into Jupiter 's shadow , ought to continue to affect the eye for a 77th of 41~ ; that is , about 32~ of time after the last sensible red light is gone .
It is therefore a certain consequence of our hypothesis , that a satellite , seen from the earth , ought to change its colour about half a minute before its total immersion , from white to a livid greenish colour ; thence into blue , and at last evanish in violet .
I need hardly observe , that the same phenomenon must take place in the time of emersion by a contrary succession of colours , beginning with red , and ending in white .
12 .
If this phenomenon be perceived by astronomers , we shall have a direct proof of the different velocities of the differently-colour 'd rays , and , consequently , a mechanical account of their different degrees of refrangibility ; for I see not , to what other cause such an appearance could be reasonably ascribed .
If it be not , we may conclude , that rays of all colours are emitted from the luminous body with one common velocity .
Geneva , Feb. 1 , 1753 .
T. Melvill .
This paper was delivered to Mr. Short , in order to attend to the particular observation of Jupiter 's satellites recommended by Mr. Melvill , who after some time made the following report to the Society .
EVER since this paper of Mr. Melvill 's was put into my hands , I have carefully attended the emersions of Jupiter 's first satellite thro ' a reflecting telescope of four feet focal length , and with a proper magnifying power ; but I have not perceived the least alteration in the colour of the light reflected by the satellite , except in quantity .
It may indeed be observed , that these emersions are seen sooner or later thro ' telescopes of different lengths , and by eyes of different goodness : and it may therefore be alleged , that there is a certain quantity of time elapsed between the very first emersion of the satellite , and the instant when it is perceived by the very best eye , assisted by the best telescope ; and that , during this interval , the succession of colours , above-mention 'd , is performed .
But our author , in consequence of his hypothesis , says , that this succession of colours may be perceived for the space of 32 ' ' after the first emersion of the satellite ; and I am fully satisfied , from repeated observations , that the quantity of time elapsed from the very first emersion of the satellite , till it is perceived by a good eye , assisted by a good telescope , can amount only to a very few seconds .
So that , upon the whole , we may conclude , that it does not appear , by the observations of the emersions of the first satellite of Jupiter , that the rays of different colours move with different degrees of velocity .
But our author 's conclusion , that , if the rays of light emitted from Jupiter 's satellites , at the time of their immersion and emersion , should not be found of different colours , the rays of all colours emitted from luminous bodies will have one common velocity , seems only to hold good , on a supposition , that light is propagated by a continued motion , in the manner of a projectile .
Dr. Knight , in his treatise on attraction and repulsion ( Prop. 69 .
) has considered the propagation of light , as performed by vibrations in an elastic fluid , in the same manner as sound is produced by vibrations in the air : and he thinks , that it is as easy to conceive how the velocities of the particles of light may be different , and yet take up equal times in propagating their motion from one to another through a given space , as to explain how sounds of different tones move with equal velocities .
In accounting for both , he shows , that , in a series of particles , which mutually repel each other , the greater their velocity , the nearer they will approach other , in communicating their motions from one to another ; and consequently each of them must move thro ' a greater space in so doing : wherefore the same time may be spent in propagating a successive motion thro ' a series of particles , whose velocity is greater , if each particle has to move thro ' a greater space , as is spent , where the velocity of each particle is less , but is continued thro ' a less space .
The dilemma , to which our author 's reasoning seems to have reduced the doctrine of refrangibility , may therefore be considered as a probable argument for adopting this hypothesis of the propagation of light thro ' an elastic medium .

