10 
The Astronomical Survey of the Universe [ch. i 
orbit of a star at a distance of 1000 parsecs, or 3250 light-years, is of course 
only a tenth as great; it is of the size of a pin-head held at a distance of 50 
miles. The resources of observational astronomy are strained to the utmost to 
detect even the former of these parallactic motions, and are totally inadequate 
to measure it with accuracy, the error of measurement being about equal to 
the whole quantity to be measured. It is utterly impossible either to detect 
or measure the smaller parallactic motion of a star a thousand parsecs away, 
and is likely to remain so for many centuries to come. Yet a thousand parsecs 
is only a tiny fraction of the whole size of the universe. To survey the remote 
depths of space something of wider reach than the parallactic method is 
needed. Quite recently astronomers have discovered other and more far- 
reaching methods. 
10 . Spectroscopic Parallaxes. One of the most important of these is the 
method of “spectroscopic parallaxes” discovered by Dr W. S. Adams, now 
Director of Mount Wilson, and Kohlschiitter in 1914. Two stars which are 
of exactly similar structure in all respects must necessarily emit light of 
precisely similar quality, so that their spectra must be similar in all respects. 
If the stars were at different distances, the spectra would naturally differ 
in brightness, and on measuring the ratio of their two intensities, it would 
be possible to deduce the ratio of the distances of the stars. Thus if the 
distance of one star had already been determined by the trigonometrical 
method already explained, it would be easy to deduce the distance of the 
other, even though this were so great as to render a direct determination 
of its parallax utterly impossible. The actual problem is generally far more 
complicated. When two stars have the same temperatures and the same 
chemical composition, their spectra are in general almost identical, but they 
shew minute differences if those parts of their atmospheres which emit their 
radiation are at different pressures. For reasons which will become clear later, 
stars of different sizes generally have different pressures in their atmospheres, 
and so exhibit slightly different spectra. Working backwards from this fact, 
Dr Adams discovered how to deduce the difference in size of two otherwise 
similar stars from minute differences in their spectra. As the difference in 
the intensity of their light arises jointly from differences in size and differences 
in distance, it is a simple matter to deduce the ratio of the distances of the 
two stars when once the ratio of their two sizes has been determined. This 
method is generally called that of spectroscopic parallaxes; it can hardly yet 
claim the accuracy of the trigonometrical method for near stars, but it has 
the great advantage of being successful with stars which are too remote for 
the trigonometrical method to be applicable at all. It is of course only of use 
for stars which appear moderately bright, but there is no limit to the distances 
at which it is available. 
11 . Cepheid, Parallaxes. An even more far-reaching method of determining 
stellar distances depends on the peculiar properties of a certain class of stars