Yerkes Spectral Classification System

Generality

The Yerkes system constitutes a development of the Harvard system. Its main difference is due to the introduction of a second parameter, that is related to star luminosity. Essentially this system represents the work of two astronomers W.W. Morgan and P.C. Kenan and as already said in previous pages, this system is based on slit spectrograms covering a region ranging between ll 3930-4860, with a dispersion of 115 A/mm. Many years before establishment of Yerkes system, Adams and Kohlshutter at Mt. Wilson, had shown that some differences could be recognized in the spectra of giant and dwarf stars, when observed at 40 A/mm dispersion. Adamd and Kohlshutter through a careful comparison of pairs of lines not to far from each other in the spectrum and with nearly the same intensities, found that certain line ratios were strongly dependent on stars luminosity. After this discovery the logical following step was to plot average line ratios against absolute magnitudes of the stars to derive an empirical relation between line ratios and absolute magnitudes. This development yielded a great impulse in following spectroscopy researches. Therefore the simply notation giant star could be ambiguous, when it's not clear if a star could be considered giant from a determination of its absolute magnitude or from a determination of its radius or still from spectrum line ratios. Naturally from a spectroscopic point of view is preferable to use the last notation that refers only to the observed spectrum phenomena. See in the following figure a comparison between dwarf and supergiant star spectrum.

For a given chemical composition the lines in the spectra of a star are essentially determined, not only, by the conditions present in its atmosphere, conditions such as degree of thermal ionization, but also by its superficial gravity g expressed as:

g = GM/R²

The previous equation clearly relate gravity with star radius. So in supergiant stars g, density and gas pressure, will be generally quite little than in normal stars. This situation will be observable as sharpened effect of some spectral lines. Luminosity effects are considered in Yerkes system through the introduction of a series of luminosity classes from I to V. As consequence, the introduction of luminosity classes has signified also the introduction of a second dimension to the unidimensional Harvard spectral sequence. The procedure of stars classification in the Yerkes system, needs the use of standard stars both in spectral type and luminosity class, to define the system itself.

Spectral subtypes used with Yerkes system

Luminosity classes in the Yerkes System

© 2006 - Valter Arnò