296 ANNUAL, REPORT SMITHSONIAN INSTITUTION, 1920. 



and Williamson that the deepest interior is not entirely a rigid solid, 

 but more in the nature of a very viscous, thick liquid, which damps 

 the transverse earthquake vibrations. The possible factor of the 

 disintegration of the elements of highest atomic weight must be taken 

 into account, but more can not be said here on this topic. 



It is of interest to note that this idea, that the elements of higher 

 atomic weights, the metallogenic elements, occupy for the most 

 part the deepest portions of the earth's interior, is in harmony with 

 Abbot's view as to the distribution of the elements in the sun. 23 

 He points out that the elements showing the most intense spectrum 

 lines are those of low atomic weight, with the exception of the 

 negative elements, none of which (with the possible exception of 

 oxygen), for some unknown reason, show solar spectral lines. It is 

 interesting to compare Abbot's table of intensities (p. 91) with the 

 elements as presented in our Table III of the periodic arrangement. 

 It will be seen that the first 22 elements showing the most intense 

 lines are all terrestrially petrogenic elements, and that (apart from 

 the negative elements) all the terrestrial petrogenic elements are 

 among those that show the more intense lines, with the curious 

 exceptions of glucinum, cerium, and especially potassium, which 

 show but very weak lines. The order is not the same, but the first 

 10 elements in order of spectral intensity include calcium, iron, 

 hydrogen, sodium, magnesium, silicon, aluminum, and titanium, 

 which, with oxygen, potassium, and phosphorus, are the first 11 

 elements in order of abundance in the earth's crust. On the other 

 hand, the metallogenic elements show the least intense or no solar 

 spectrum lines. Thus in Abbot's intensity tables ISTos. 23 to 36 (the 

 last) include in order palladium, copper, zinc, cadmium, germanium, 

 rhodium, silver, tin, and lead. The metals of the platinum group, 

 with tungsten, bismuth, mercury, thallium, and one or two others, 

 give extremely feeble or doubtful lines. As Abbot shows, taking 

 the elements in groups of order of intensity, this diminishes with 

 increase in the mean atomic weight of the group. 



Abbot explains this distribution, to which the only real excep- 

 tions are cerium, glucinum, and potassium, by the supposition that 

 "the explanation of the decrease of intensities with increasing 

 atomic weights seems to depend on the depth of these gases below 

 the sun's surface," and this supposition is confirmed by the spectrum 

 observations of displacements of the lines of various elements due to 

 pressure and those that show in the " flash " spectra during eclipses. 

 The coincidence between the occurrence of the elements in the earth 

 and in the sun, as regards relative abundance and depth, is ap- 

 parently so very close and detailed as to be suggestive of a similar 



M Abbot, C. G., The Sun, 1911, pp. 91, 94, 99, 104, and 252 ff. 



