Abstract
The cosmic abundances of the elements were estimated by Goldschmidt (1937) from a study of terrestrial and meteoritic abundances and a comparison of these with Russell's data on the sun. More recently Brown has prepared a table weighting the proportions of the iron and silicate phases according to an estimated proportion of these phases in the earth. The writer has recently proposed that the chondritic meteorites themselves may represent an average sample, and has shown that this assumption is consistent with the density of the moon, which on the basis of his recent discussion in regard to the origin of the solar system should also be approximately a sample of nonvolatile materials. A table of abundances is prepared assuming that these meteorites do represent such a mean sample. This table does not differ markedly from Goldschmidt's, but is distinctly different from Brown's, both because of a different weighting of the phases in meteorites and because of a different choice of data from the literature. Iron is much less abundant than estimated by Brown and somewhat less abundant than Goldschmidt's estimate. There is some indication for markedly low abundances of selenium and tellurium and bromine and iodine, which may indicate some escape of these elements during the formation of the meteorites. Mercury is low, almost certainly because of its loss as a volatile substance during the formation of the meteorites.

This publication has 15 references indexed in Scilit: