Current interest in stellar evolution is concentrated on the life of a star after it has left the main sequence. Of particular interest are the red giant or supergiant periods during the hydrogen and helium shell burning phases. Convective mixing during these stages can mix nuclear processed material to the surface where it may be viewed by spectroscopic methods. It is imperative that this rare chance to view processed material be exploited fully to increase our knowledge of stellar evolution.

The observation and interpretation of cool star spectra has its own particular set of problems and advantages. A particular difficulty is the formation of molecules at the low temperatures which occur in the atmospheres of late stars. Not only must the particularly complex spectra of molecules be dealt with but the problem of chemical equilibrium in the atmosphere must be solved accurately before quantitative analysis may be performed. The formation of molecules, however, has one advantage in that it very dramatically separates those stars with carbon to oxygen ratios greater than one from those with ratios less than one. It is the very high dissociation energy of 11.1 eV for the CO molecule which performs this separation. If carbon is less abundant than oxygen all of the carbon is tied up in CO and only oxides are formed in the stellar atmosphere which produce typical M star spectra. If, however, carbon is more abundant than oxygen then carbon compounds such as C2 are formed in place of the oxides and a carbon star spectrum is formed. One of the great advantages of infrared stellar spectra is that it is the only ground based technique for observing CO in stellar atmospheres.