96 CELESTIAL SPECTROSCOPY. 
suggests rather that they are in different stages of condensation, and 
dissimilar in density. 
It is obvious that if the orbit of a star with an obscure companion is 
inclined to the line of sight, the companion will pass above or below 
the bright star, and produce no variation of its light. Such systems 
may be numerous in the heavens. In Vogel’s photographs, Spica, which 
is not variable, by a small shifting of its lines reveals a backward and 
forward periodical pulsation due to orbital motion. As the pair whirl 
round their common center of gravity, the bright star is sometimes ad- 
vancing, at others receding. They revolve in about four days, each 
star moving with a velocity of about 56 miles a second in an orbit prob- 
ably nearly circular, and possess a combined mass of rather more than 
two and a half times that of the sun. Taking the most probable value 
for the star’s parallax, the greatest angular separation of the stars 
would be far too small to be detected with the most powerful telescopes. 
If in a close double star the fainter companion is of the white-star 
type, while the bright star is solar in character, the composite spectrum 
would be solar with the hydrogen lines unusually strong. Such aspec- 
trum would in itself afford some probability of a double origin, and 
suggest the existence of a companion star. 
In the case of a true binary star the orbital motions of the pair would 
reveal themselves in a small periodical swaying of the hydrogen lines 
relatively to the solar ones. 
Prof. Pickering considers that his photographs show ten stars with 
composite spectra; of these, five are known to be double. The others 
are: 7 Persei, € Aurigw, dO Sagittarii, 31 Ceti, and @Capricorni. Per- 
haps (6 Lyre should be added to this list. 
In his recent classical work on the rotation of the sun, Dunér has not 
only determined the solar rotation for the equator but for different par- 
allels of latitude up to 75°. The close accordance of his results shows 
that these observations are sufficiently accurate to be discussed with 
the variation of the solar rotation for different latitudes which had 
been determined by the older astronomical methods from the observa- 
tions of the solar spots. 
Spectroscopic Photography.—Though I have already spoken. inci- 
dentally of the invaluable aid which is furnished by photography in 
some of the applications of the spectroscope to the heavenly bodies, 
the new power which modern photography has put into the hands of 
the astronomer is so great, and has led already, within the last few 
years, to new acquisitions of knowledge of such vast importance, that 
it is fitting that a few sentences should be specially devoted to this 
subject. 
Photography is no new discovery, being about half a century old; it 
may excite surprise, and indeed possibly suggest some apathy on the 
part of astronomers, though the suggestion of the application of pho- 
tography to the heavenly bodies dates from the memorable occasion 
