130 THE CANADIAN NATURALIST. [Dec. 



besides others which give negative evidence only. Alpha 

 Orlonis has been carefully examined, and contains most of the 

 above-named elements with the exception of hydrogen. The 

 presence of hydrogen has been noticed in the sun, and in almost 

 forty fixed stars, and is eminently characteristic, showing that its 

 presence belongs to the atmospheres of the luminous bodies them- 

 selves, and not merely to our own atmosphere. 



These investigations have confirmed and demonstrated, beyond 

 the shadow of a doubt, that all the planets shine by light reflected 

 from the sun, and that any variety differing from the solar 

 spectrum may be attributed to the peculiar properties of the 

 atmospheres that surround the planets themselves. 



One of the most important and interesting deductions to be 

 drawn from these researches, is in connection with the origin of 

 the colour of the stars. That a difference of colour in the stars 

 does exist, is too well known to require any comment : for " one 

 star differeth from another star in glory." And it is now no longer 

 a matter of conjecture that the brightest stars at least are, like 

 our sun, giving energy and life to systems of worlds like our own, 

 adapted for the abode of intelligent life. While yellow and red 

 stars are the most frequent, in double stars the contrasted 

 colours are green and blue. The source of the light of the stars 

 must be a solid or liquid body in a state of incandesence, as only 

 such bodies, when raised to a high temperature, give out a con- 

 tinuous spectrum. In the case of the fixed stars and the sun, 

 this continuous spectrum becomes crossed by dark bands, which 

 are produced by the absorbing power of the constituents, held in a 

 vaporous form in the investing atmospheres. These atmospheres 

 vary in chemical constitution, according to the elements composing 

 the star ; and the dark lines are produced by the absorptive power 

 of the vapours forming the stellar atmospheres. They correspond 

 to the bright lines they would form in an incandescent state, aiad 

 would be the strongest and most numerous in the more refrangible 

 portions of the spectrum, consequently a star would have a red or 

 orange tint should that part of the spectrum suffer least absorp- 

 tion : while, on the contrary, should the red and yellow portion 

 have most lines, the blue and green rays would then predominate 

 in the colour of the star. 



In Sirius, the < dog star,' which is of a brilliant white, there 

 are no lines sufficiently intense, in any particular part of the spec- 

 trum, to interfere with our receiving the light in about the same 



