154 ANNUAL, REPORT SMITHSONIAN INSTITUTION, 1923 



Here, again, the answer which we would now give depends upon 

 knowledge which has come within the last decade or two. We have, 

 even now, no direct evidence regarding the age of the stars, or the 

 sun; but we have information about the age of the earth that has 

 magnified our conception of the duration of the universe in time, in 

 as startling a fashion as the study of the globular clusters has en- 

 larged our idea of its extension in space. 



The new method of measuring time is really very simple. Ura- 

 nium is radioactive, and slowly " decays. " One by one its atoms 

 eject a part of their nuclei, and change into atoms of a different 

 element. These again break up, and so on through a long and 

 wonderful series of transformations, in which radium is one step. 

 The particle ejected from the nucleus is sometimes an electron, but 

 oftener an alpha particle, identical with the nucleus of a helium 

 atom. Finally, at the end of the list, there remains a stable atom of 

 lead — but not of ordinary lead, for its atomic weight is 206, instead 

 of 207 as usual. In the course of ages, this radio-lead must accumu- 

 late in all uranium minerals. The rate of accumulation is accurately 

 known, from a study of radioactive phenomena and we can be sure 

 that the weight of lead produced in a million years is one eight- 

 thousandth of that of the uranium which is present. By determin- 

 ing the percentages of uranium and lead now present in a mineral, 

 and applying this principle, we can find out how old the mineral 

 is — provided, of course, that it contained no lead when it was origi- 

 nally formed by crystallization in the molten rock mass. Such 

 primitive lead would, however, be ordinary lead, of higher atomic 

 weight, and a determination of the atomic weight of the lead de- 

 rived from our specimen will enable us to tell how much of it was 

 there when the mineral formed, and how much has been produced by 

 radioactivity since this time. 



In this way reliable values can be obtained for the ages of various 

 minerals, and the dates of the eruption of the rocks in which they 

 occur. The latter can often be defined in geological terms, and 

 hence we can date the various geologic periods, finding a good 

 general agreement with the geological order of succession. The 

 oldest minerals so far studied are found in rocks of Middle Pre- 

 Cambrian age. Specimens from Europe. Africa, and America agree 

 in giving ages of between 1,000 and 1,200 millions of years. These 

 individual crystals have been in the rocks for all this time. The 

 earth, as a planet, must be older. The speaker, from consideration 

 of the whole amount of uranium and lead in the earth's crust, showed 

 last year that its age is apparently less than eight billions of years, 

 and probably something like 4 billions. If, as seems most probable, 

 the planets were produced by eruptions from the sun, under the tidal 



