BAXTER AND CHAPIN. — ATOMIC WEIGHT OF' NEODYMIUM. 229 



von Welsbach from his own material. ^ The last were taken from the 

 drawings which were Auer's only published record, and since they are 

 subject to two personal errors are less exact than the others. Being 

 obtained, however, from very carefully prepared material, they should 

 represent at any rate the number and the general location of the bands 

 in the neodymium spectrum. Rech's measurements, since his material 

 was less carefully prepared, include lines belonging to praseodymium 

 and possibly to other elements. 



While some of these bands are sharp, and appear in the same posi- 

 tions in solutions of various concentrations, others fade away gradually 

 on the edges, and sometimes have their maxima shifted by changes in 

 the concentration. Consequently the disagreement as to the exact 

 positions of bands is due in many cases merely to differences in the 

 conditions of observation. Furthermore, Forsling and Rech used chlo- 

 ride solutions, whereas Auer von Welsbach and the writers used nitrate 

 solutions. On the whole the agreement between the different sets of 

 values is very satisfactory. The bands between X 587 and X 571 are 

 too close together to be distinguished from one another under any but 

 the most favorable conditions. The double band X 575 and X 574 

 in particular separates only under rather limited conditions. The band 

 at X 579 in one of Auer's drawings may be intended to represent two 

 lines at X 580 and X 577, and is thus interpreted by Kayser. Auer's 

 band X 567.5 appears very faintly in but one of his four drawings. 

 There is a complete absence of shadow at this point in another drawing 

 which represents a more concentrated solution, and which one would 

 expect consequently to show the band. It is possible therefore that 

 its appearance in the first print was merely accidental. The band X 525 

 in concentrated solutions merges with X 522 and in dilute it be- 

 comes invisible, so that it is not surprising that it was not observed 

 by Forsling and Holmberg. The band X 514.5, given by Rech alone, 

 is coincident with no prominent band of any other rare earth, nor could 

 it be detected in any of our preparations. The band X 488 which both 

 Forsling and Rech observed, as has already been stated, can be seen 

 only with very strong absorption. This is doubtless the reason that it 

 was not observed by Auer. 



The presence or absence of the predominant impurities, praseodym- 

 ium, cerium, lanthanum and samarium, in our best mateiial has already 

 been discussed. It is of some interest to consider other earths having 

 prominent absorption bands in the visible spectrum. ^ Since none of 



1 Sitzungsber. Akad. Wiss. Wien., 112, 1037 (1903). ~ 



2 The following data are obtained from drawings by Urbain, Jour, de Chim. 

 Phys., 4, 31, 105, 232, 321 (1906). 



