Part III. Relations between the Refractive Indices and 

 Compositions of the Minerals of the Calcite Group 



Some twenty-nine analyses of these minerals have been col- 

 lected, with which the indices of refraction are recorded. A 

 short study of these showed that the refractive indices varied 

 directly with the composition and that given the percentages of 

 the dififerent carbonate groups present it was possible to calculate 

 the indices that any specimen should show. In order to do this, 

 it was first necessary to establish the refractive indices of the 

 various pure carbonates. Calcite was apparently the only mem- 

 ber of the group whose indices had been determined on theoreti- 

 cally pure material. Iceland spar, which is practically loo per 

 cent CaCOy, has been repeatedly and very carefully tested. In 

 the cases of the other molecules it was necessary to study the 

 analyses and make some trial calculations before fixing upon their 

 proper refractive indices. The values, as they were finally 

 determined for the pure carbonates, were as follows ; 



w e 



CaCOs 1.658 1.486 



MgCOs 1.700 1.509 



MnCOs 1.817 1.597 



FeCOa 1.875 1-633 



By using these numbers the theoretical indices for each anal- 

 ysis were calculated and are given in the columns headed "Calc." 

 in Table VII, below. It will be seen that the agreement between 

 these values and those determined by actual measurements is, 

 in the majority of cases, very close and usually within the proba- 

 ble limits of error. There are two cases in which it seems clear 

 that some error must exist in the determinations. The presump- 

 tion is strong that the indices of the ordinary rays in Analyses 

 19 and 29 are wrong, since they show marked dift'erences from 

 the values given for other specimens of similar composition. 

 With these two exceptions there are only four cases that show 

 a difference between the measured and calculated indices for the 

 ordinary ray greater than .004 and the same number in the case 

 of the extraordinary ray that have a difference greater than 

 .005. The average difference, still omitting Analyses 19 and 

 29, for the ordinary index is .004 or if the -\- and — signs are 



