CARBON- 14 AGE DETERMINATION — F. II. H. ROBERTS 337 



purest form. In running tests the samples to be dated are burned, 

 treated in the separation unit, and then measured in the radiation 

 counter. The measurements are given on the basis of the carbon-14 

 disintegrations per minute per gram of carbon. For present-day living 

 samples the specific activity is 15.3, for samples 5,568 years old it is 

 7.65, and for samples 11,136 years old it is 3.83. The disintegration 

 rate is such, however, that the proportion of radiocarbon remaining 

 after 20,000 years is so small that accurate counting is very difficult 

 and the effective range may be considered somewhat less than that 

 age. There is a method for enriching samples which may make possi- 

 ble the obtaining of dates as far back as 30,000 years, but that at 

 present appears to be the maximum. The errors in the dates now 

 being obtained are considered to range from 5 to 10 percent. 



When the laboratory equipment was ready Dr. Libby and his 

 associate. Dr. James R. Arnold, ran a series of tests on samples whose 

 ages had been fairly accurately established by other means but which 

 were unknown to them. Material, ranging in age from 1,300 to 4,600 

 years, from Egyptian tombs, from archeological sites in our own 

 Southwest, and from redwood trees was provided by different muse- 

 ums, and it was found that the carbon-14 dates obtained for them 

 agreed with the known ages within the calculated error of the method. 

 In making the preliminary tests it was found that the most useful 

 materials are plant fibers and wood, charcoal, antler, burned bone, 

 shell, dung, and peat. 



Arrangements were then made with a committee representing the 

 American Anthropological Association and the Geological Society of 

 America to obtain samples for testing and a grant was made by the 

 Wenner-Gren Foundation for Anthropological Research (the Viking 

 Fund) to assist in the support of the program. Archeologists and 

 geologists began sending in the necessary materials and the series of 

 tests got under way in the spring of 1949. Announcements of the 

 dates obtained were made informally from time to time, but it was 

 not until October 1950 that a lengthy series was made public. That 

 list was printed in February 1951 in Science, vol. 113, No. 2927, pp. 

 111-120. An additional list appeared in the same journal, vol. 114, 

 No. 2960, pp. 291-296 in September of that year. Most of the infor- 

 mation contained in those articles was also published, with discussions 

 of its significance, in the Memoirs of the Society for American Arche- 

 ology, American Antiquity, vol. 17, No. 1, pt. 2, 1951. 



The method developed by Dr. Libby and the results obtained indi- 

 cated so many potential applications for radiocarbon dating that 

 new laboratories for making carbon-14 measurements have been estab- 

 lished at Yale University, the University of Michigan, Columbia 

 University, and the United States Geological Survey. Others are 



