578 FRED M. UBEK 



The technical skill required for determining isotope ratios with a 

 mass spectrometer is not beyond that of most college-trained labora- 

 tory workers, assuming that the apparatus has been properly ad- 

 justed initially. Where the batteries shown in Figure 2 are replaced 

 by electronic rectifying circuits, the services of someone skilled in 

 electronics will be needed from time to time. Such a person should 

 be available locally and on short notice if reasonably continuous 

 operation is desired. 



Complete installations of currently available mass spectrometers 

 will probably cost in excess of ten thousand dollars (34,37,38). An 

 instrument of the Nier type probably could be built in most uni- 

 versity instrument shops for somewhat less. But it is almost neces- 

 sary, even then, to have the complete ion tube assembly of Figure 2 

 made by someone who has had construction experience wdth them. 

 The remainder of the apparatus consists of less critical items of 

 either an electronic or glass-blowing nature. (See also 3a.) 



4. Preparation of Samples for Analysis 



Isotopic analyses of the complex organic substances that occur in 

 nearly all biological studies must be made indirectly, since only gases 

 or volatile liquids possessing suitable characteristics can be introduced 

 into an ordinary mass spectrometer. Furthermore, it is often neces- 

 sary to degrade a complex molecule stepwise so that an analysis can 

 be made for isotopes occurring in the various positions or radicals. 

 The final measurements, however, in all biological investigations to 

 date have been carried out on simple gases. 



The ultimate selection of a particular test gas rests, for each ele- 

 ment, on several basic considerations. Low molecular weight is 

 favored owing to the better spectral separation obtained. Mole- 

 cules with a tendency to adhere to the inner walls of the evacuated 

 spectrometer system are avoided since difficulty is experienced in 

 removing them completely following each analysis. Water vapor 

 falls in this category. Some substances decompose too readily upon 

 contact with the hot filament; others furnish a varying yield of ioni- 

 zation products; still others produce ionized particles subject to 

 interference from other particles of the same mass-to-charge ratio 

 (Table II). 



Recommended test gases for each of several is 3t opes, based on the 

 considerations just outlined, will now be briefly discussed. For 



