Researches towards establishing a Theory of the Dispersion of Light .
No. III .
By the Rev. BADEN POWELL , M.A. F.R.S. , Savilian Professor of Geometry in the University of Oxford .
Received October 20 , 1836-Read January 19 , 1837 .
Introductory Remarks .
IN two former portions of researches on the subject of dispersion * , I have discussed all the observed refractive indices for definite rays , in different media , which had come to my knowledge ; consisting of those for ten media determined by FRAUNHOFER , and those in ten other cases by M. RUDBERG , comparing them with the calculated results of the theory of M. CAUCHY ; and the agreement is sufficiently close .
In those papers , and elsewhere , I have remarked the importance of extending the inquiry , especially to media of higher dispersive power ; in which cases ( as appears from the nature of the formula ) the theory would be put to a more precise test .
In the former instances the work of determining the indices was done to my hands , and I could proceed to the theoretical computations , with the most perfect confidence in the accuracy of experimental data , furnished from the labours of observers so well known for precision and skill , and obtained , too , before the formula of theory had been deduced .
In any comparison of theory with experiment , it is , in all points of view , far more satisfactory that such comparison should be made with the observations of others rather than those of the theoretical computer himself .
In the present instance , however , this desirable condition has not been fulfilled .
Though the importance of obtaining a series of indices for the standard rays in different media had been long since pointed out and acknowledged by the most eminent philosophers , yet no observer was found to undertake the task of carrying on the work which FRAUNHOFER and RUDBERG had so successfully begun .
I was thus left to make an attempt myself to supply the deficiency ; and my observations were communicated , and printed copies distributed , to the Physical section of the British Association at the Bristol meeting in August 1836t .
From the remarks prefixed to those results , the scientific reader will , I trust , be sufficiently enabled to Judge of the nature and degree of accuracy of the observations , conceive , in general , the indices deduced may be relied on as exact , to at least three places of decimals .
Some exceptions are noticed in the tract as less certain , viz. the oils of Angelica , Cunmmin , and Pimento , and the balsam of Peru .
Such as they are , however , these results form ( as far as I am aware ) the only existing data for pursuing the comparison with theory .
But I trust they may not be thought insufficient , when we consider that in the present stage of the inquiry the object to be aimed at seems chiefly such a general comparison as may enable us to see whether the main principle of the undulatory explanation of the dispersion is applicable , with a sufficient approach to precision , to encourage us to pursue the theory , or whether it must be abandoned , and some new principle sought .
With respect to media of low dispersive power , little doubt can exist .
I have therefore not thought it worth while to go through the calculations for many of this class , but have confined my examination in the present instance chiefly to the higher cases to which my observations have extended .
Method of Calculation .
It may be necessary to premise a notice of the method of calculation adopted in the present communication , as it differs from that employed in my two former papers .
That method consisted in finding , in the first instance , by a tentative process , ( virtually equivalent to assuming the two extreme indices from observation , ) the fundamental arc 0 , from which the others were derived on dividing by A for each ray , so as to fulfil the conditions of the approximate formula It also appears from the investigations given in several consecutive papers in the London and Edinburgh Journal of Science , &amp;c. , that this formula is in fact obtained by supposing the sum of a series of analogous terms collected into one , with a common constant coefficient H. This simplified hypothesis ( from the accordances already obtained ) is evidently very near the truth for the whole range of media hitherto examined .
Professor Sir W. R. HAMILTON afterwards pointed out ( besides a direct process for performing this approximate calculation ) a method of investigating the exact expression when we do not allow the above assumption as to the coefficients , viz. ( 1 )2 Or pr sin ( -^ ) This is explained at large in the journal just named* ; and in a subsequent numbert l have stated some further particulars illustrative of the process .
It is this method which I shall use in the present investigation , and therefore here merely quote the resulting formula from the last-mentioned paper .
It is as follows ; , -SpF a , ( OF p ) + b , ( pa 2 SF + pi .
: where the three indices of refraction are assumed from observation for the respective rays B , F , and II , in the particular medium ; while A is the index sought , any other of the four remaining of the seven standard rays , corresponding to which a , and b , are constants for the ray , independent of the particular medium , and which have been found from the values of A in the paper last referred to , as follows : log a , = 1P95433 log be = 2'65253 log a , = 1P80441 log bD = 1P06281 log aE = 1-49646 log bE = 103196 log a , ,= 1.74027* log b. = 1,62954* .
The nature of the process of computation is obvious from an inspection of the formula ; and the data here stated will suffice for those who may wish to verify the calculations .
I proceed to give the results in a tabular form , and then to offer such general conclusions as I think may be safely derived from them .
Comparison of Refractive Indices from CAUCHY 's Theory andfrom observation .
1 .
Nitric Acid .
3 .
Sulphuric Acid .
Ray .
Index Index Difference .
Ray .
Index Index ifference .
observed .
calculated .
observed .
calculated .
B 1'3988 B 1^4321 C 1'3998 1'3998 '0000 C 1'4329 1'4329.0000 D 1'4026 1'4024 *0002 D 1'4351 1'4349 *0002 E 1'4062 1'4058 0004 E 1'4380 1'4375 '0005 F 1'4092 F 1'4400 G 1'4155 1*4156 + *0001 G 1'4440 1'4448 + '0008 H IK4211 H 14468 2 .
Muriatic Acid .
4 .
Oil Angelica .
B 1'4050 B 14836 C 1'4065 1*4061 '0004 C1 4863 1'4849 *0014 D 1-4095 14089 *0006 D 1'4887 1'4882 '0005 E1 4130 1'4125 '0005 E 1'4932 1'4924 '0008 F 1'4160 F 1-4963 G 1'4217 1'4222 + '0005 G 1'5049 1'5035 *0014 H 1'4265 H 1'5099 Oil Cumnmi .
11 .
Oil of Anise .
General Mean .
Ray .
Index Index Difference .
Ray .
Index Index Diference .
observed .
calculated .
observed .
calculated .
B 1'5023 B 1'5469 C 1'5043 1'5039 '0004 C 1'5489 1'5490 +.0001 D 1'5075 15079 + *0004 D 1'5553 15551 *0002 E 1'5130 1i5136 + *0006 E 1'5642 1'5635 '0007 F 1i5196 F 1'5726 G 1-5326 1'5317 -0009 G 1'5904 15909 + ?
0005 H 1-5432 H 1-6082 6 .
Oil Sassafras .
12 .
Oil of Anise .
Temp. 15'-2 .
B 1-5252 B 1'-5486 C 1-5267 1i5268 -'0001 C 1-5506 1-5509 + '0003 D 1P5312 1'5315 + '0003 D 1'5574 1'5570 0004 E 1'5380 1P5376 '0004 E 1i5661 1'5656 0005 F 1'5441 F 1-5747 G i-5568 1-5569 + '0001 G 1-5926 ]'5932 + '0006 H 1i5687 H 1'6103 7 .
Ofl Pimento .
13 .
Sulphuret of Ca bon .
Temp.220 .
, ~~~7 Oil Pi _ en. .
B 1'5281 B 1'6145 C ]'5317 1-5300 '0017 C1 6176 1'6178 + '0002 D 1'5347 1'5348 + '0001 D 1'6272 1'6269 '0003 E 1e5422 1'5412 '0010 E1 6405 1'6392 '0013 F 1-5478 F 1'6521 G 1P5599 1i5605 + '0006 G 1'6763 1'6776 + '0013 H 1-5719 H 1'7009 __ ~ ~_ B ___ ___ __ 8 .
Kreosote .
14 .
Sulphiu ret of Carbon .
Temp. 1 2 ' , B 1'5327 B 1'6234 G 1'5341 1-5344 1 '0003 C 1'6258 1-6268 + '0010 D 1-5387 1'5392 ?
0005 D 1'6369 1-6361 '0008 E 1'5457 1-5455 0002 E| 16496 1-64S5 0011 F 1'5520 F 1-6616 G 1'5645 1 5646 + -0001 CG 1'6864 1-6870 + '0006 H 1-5762 H 1'7103 9 .
Balsam of Peru .
15 .
Oil of Cassia .
B 1-5846 B |15885 C 1-5s69 1-5869 '0000 C 1-5918 1-5913 '0005 D 1-5931 1'5935 + '0004 D 1-6017 1-6000 '0017 E 1|G029 1'6028 |0001 E 1'6155 1-6135 '0020 F 1'6130 F 1'6295 CG 1'6336 1i6336 '0000 C-G 16607 1-6646 + '0039 H 1P6533 H 1'7002 10 .
Oil of Anise .
Temp. 210-2 .
B 1-5452 C 1'5473 1'5473 '0000 D 1'5533 115533 '0000 E 1-5623 1-5615 '0008 F 1'5705 G 1'h5883 1-5887 V+ '0004 H1 16o61 General Conclusions .
With regard to some of the first media above compared , the agreement of observation and theory is sufficiently near ; but they are low in the scale of dispersion .
The cases before mentioned as being somewhat uncertain do not allow of a close comparison .
The balsam of Peru ( though only from .
approximate data ) and the lreosote , present higher numbers , and the agreement is therefore the most satisfactory .
This is also the case with oil of anise seed , though perhaps not to the same degree ; but the differences are probably not greater than what may reasonably be allowed .
In all these cases , however , we trace some uniformity in the character of the differences ; theory being always in defect for the ray E , and in excess for G. When we proceed to the sulphuret of carbon this becomes more apparent ; and the differences increase for the more refrangible rays .
And lastly , in oil of cassia the same regular order of deviation is more marked , and the differences much greater , especially in the rays E and G. This regularity of character , as well as increase in the amount of the discrepancy , at once shows that it is at least partly due to some other cause than errors of observation .
Yet we must remark , that even in these cases there is a sort of general accordance preserved between observation and theory , the agreement being still accurate to the second place of decimals .
The following considerations , then , suggest themselves : I. It is to be observed that in all these cases closer accordances might be obtained if we took slightly different values for the assumed indices of B , F. and H-1 ; such as would be consistent with the probable errors of observation .
In oil of cassia , however , I have found that the errors , even when thus distributed among all the indices , are still too great .
2 .
The constants a and b are derived solely from the values of X. for each ray , taken from the well-known determinations of FRAUNHOFER from interference .
It is doubtless possible that these determinations may be affected by errors .
In computing , by the method here used , some of FRAUNHOFER'S indices , in which small discrepancies were found , Sir W. R. HAMILTON undertook to investigate what amount of alteration in the values of X would account for those differences .
He communicated his researches in a letter to niyself , with the values of , _ and consequently those of a and b , thus altered .
I have repeated the calculation for oil of cassia , but find this change quite insufficient to remove the discrepancy ; and that , in fact , a much larger alteration than could for a moment be allowed in FRAUNHOFER 's data , must be supposed , in order to produce any sensible effect .
3 .
The entire method of computation here followed , though founded on the exclusion of those approximate suppositions which are allowed in the simpler formula , is yet dependent on the omission of terms in the series for p beyond the three first*a Since in this analysis we know nothing of the values of the coefficients of this series , but proceed solely by the elimination of them , it is still possible that in the neglected terms we may find the means of explaining the observed discrepancy .
In a word , it remains an important subject for future research whether a further prosecution of the analysis , either by this method or by that which M. CAUCHY has himself recently pointed out , or that of Mr. KELLAND , may not lead to more successful results .
For the present I will only remark , by way of recapitulation , that upon the whole I conceive the formula , as already deduced from the undulatory theory , applies sufficiently well to the case of media whose dispersion is as high as that of oil of anise seed .
It also represents , with a certain general approximation to the truth , the indices of some more highly dispersive bodies .
It is therefore extremely probable that the essential principle of the theory has some real foundation in nature .
While , looking at the regularity of the deviation , it seems likely that the formula only requires to receive some further development or extension , in order to make it apply accurately to the higher cases , while it shall still include the simpler form which so well accords with the lower .
October 10 , 1836 .

