XVII. RADIOACTIVE TRACERS 635 



should be devised so that the chance of such exchange is minimized 

 and interpretation of results should include a thorough examination 

 of these possibilities. 



Another example in which exchange can introduce error is in the 

 samples prepared for measurement. If a solid such as barium car- 

 bonate is adopted as the form in which C^^ will be measured, there is a 

 chance that exchange will occur between the solid and atmospheric 

 carbon dioxide. If such samples are stored for later measurement or 

 checking, low results may be obtained. This problem has been in- 

 vestigated by several authors whose results appeared originally to be 

 at variance (49,50). Study has shown that the original disagreement 

 was caused by difference in mounting methods. It appears that 

 a barium carbonate sample heated to 160°C. is fairly impervious to 

 exchange, but a similar sample prepared just by filtering and drying 

 may exchange up to 40% of its C^"* in a few hours in an atmosphere of 

 moist carbon dioxide. If it is inconvenient to prepare samples so 

 that they can be heated to such a temperature they should be stored 

 in the absence of atmospheric moisture or carbon dioxide. Other 

 problems and suggestions are discussed by a number of authors 

 {51,52). 



3. Isotope Dilution 3Iethod 



The possibility of tagging one kind of molecule by radioactive or 

 stable tracer has given rise to a new method of analysis, the isotope 

 dilution method (53). In analyzing a complex mixture of organic 

 compounds — take, for example, the amino acids liberated on hydroly- 

 sis of a protein — ^it is relatively easy to separate out a particular 

 amino acid in pure form, but much more difficult to obtain the quan- 

 titative yield that would be necessary for analysis. If a particular 

 amino acid is tagged, say wdth C^* in some nonexchanging position, 

 added in known amount to the mixture in such a way that complete 

 mixing is obtained, and then separated by the procedure that is 

 standard for that amino acid, the separated amino acid will have a 

 lower specific activity of C^^ than that added. The amount of re- 

 duction in the specific activity, or the dilution, is directly related to 

 the amount of the amino acid originally present in the mixture. 



vSamples of the added and separated amino acids could be meas- 

 ured directly or one could burn them to carbon dioxide, absorb this 

 in barium hydroxide, and measure the properly mounted barium car- 



