254 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1937 



erally accepted that ages by helium would be only an uncertain frac- 

 tion of those obtained by lead ratios. Since then work by Holmes 

 and Dubey and Paneth and Urry has given results which seemed 

 more reliable and to lit the lead scale more closely. 



But they are not beyond question, and more recent work by Good- 

 man and Keevil under Evans seems to show that while the helium 

 determmations and those of thorium are fairly close, all the previous 

 work on the amount of radium in the rocks may need substantial 

 correction. Specimens from the same dike and therefore of the same 

 age since consolidation, but sho\ving considerable difference in uran- 

 ium and thorium, have, nevertheless, heUum just right to give the 

 same age, within the limit of error of the work. Contamination with 

 atmosphere can be told by the presence of neon. The process requires 

 great skill, involving as it does determination of uranium in quantities 

 of grams per ton, of radium in ten-thousandths of a milligram per 

 ton, helium in cubic centimeters per cubic meter of rock, and thorium 

 also in grams per ton. Methods first developed by Paneth have been 

 perfected by Urry.^^ 



If, therefore, we find Devonian traps giving ages between 265 and 

 300 milhon years by the helium method, while minerals in Devonian 

 pegmatites with isotopes determined give ages from 277 to 300 million 

 years, since the errors by the helium /method are likely to make the 

 ages too low and those by the lead method to make them too high, 

 we may be pretty confident that 275 million years ago was in Devonian 

 time, unless there is a general systematic error not yet located. Thus 

 the ages given by Stille in figure 2 must be somewhat changed. 



Thanks to Urry, we thus have the geologic column rouglily outhned 

 as far back as there are fossils. In fact, in the Noranda mine we find 

 rocks that seem to bo twice as old. It remains to test the country 

 rock where the minerals in the pegmatite seem to be 1,800 million 

 years old, as near Winnipeg. In this region where Harwood writes 

 of very old "Prekeewatin tonalite," the recent work of Hahn and 

 Mattauch on a rubidium-bearing mica and the change from Rb (87) 

 to Sr (87) leads to an age of 1,700 to 2,000 million years,^° which agrees 

 with the age determined by analyses hj Ellsworth, Hecht, and Kroupa 

 of uraninite and monazite, all of which indicate an age of about 1,800 

 million years. 



Thus, in spite of diflBculties, progress is being made, as results are 

 obtained from smaller and smaller quantities, and as the methods of 

 separating and determining isotopes are improved. 



•« Urry, W. D., Helium and the problem of geologic time. Chem. Rev., vol. 1, pp. 306-343, 1933. Ages 

 by the helium method. Bull. Oeol. Soc. Amer., vol. 46, pp. 1105-1120, 1935; vol. 47, pp. 1217-1234, 1936. 

 » Personal communication. See also Die Naturwissenschaften, vol. 25, No. 12, p. 189, 1937. 



