152 REPORT—1868. 
On some further Results of Spectrum Analysis as applied to the 
Heavenly Bodies. By Witu1am Hueeins, F.R.S., Hon. Sec. to the 
Royal Astronomical Society*. 
[Plate V.] 
Two years ago, at the Meeting of the British Association at Nottingham, I 
had the honour to give, in an evening discourse, a summary of the results 
of Spectrum Analysis as applied to the heavenly bodies, which had been 
obtained partly by myself and partly as the conjoint work of myself and 
Dr. W. Allen Miller, Treas. and V.P.R.S. 
I beg now to offer to the Association a brief account of some of the prin- 
cipal results of the observations which I have made since August 1866. 
These observations may be arranged according to the classes of the hea- 
venly bodies to which they relate. 
§ I. On the Fixed Stars. § IY. On the Spectra of Sun- 
§ II. On the Nebule., spots. 
§ IIL. On the Light of Comets. § Y. On the Planets. 
§ I. On tHe Fixep Srairs. 
A. Observations to determine whether the Stars are moving towards or from 
the Earth. 
The determination of the proper motions of stars from observations of 
their angular motions among the stars apparently near them, gives to us 
information of that part only of their motion whick is at right angles to a 
ray of light coming from the star to the observer. It would be a rare acci- 
dent only if the motion thus obtained represented the whole of their motion. 
The other part of the star’s motion, namely, the movement of the star in the 
direction of the visual ray towards or from the earth, seemed to be beyond 
the reach of our means of observation; for photometric estimation of the 
increase or diminution of the light of the star was obviously too coarse and 
uncertain a method for so delicate an investigation. 
Now it is precisely information on this point which appeared to be inacccs- 
sible to us, which has been brought within our reach by means of observa- 
tions with the prism. Supposing waves to be coming in upon the shore, a 
ship leaving the harbour would encounter a larger number of these waves in 
a given time than a ship would at anchor; and further, the increased velo- 
city of succession of the waves which would strike its prow could be deter- 
mined if the velocity of the ship and that of the waves were known. Con- 
versely, if the period and the velocity of the waves had been ascertained, the 
captain, by counting the number of waves which met the ship in a given 
interval of time, could determine therefrom the motion of his vessel. <A little 
consideration will make it evident that a similar effect would take place if 
the vessel were at rest, and the source of waye-motion were supposed to 
approach or recede from the vessel; in this case the velocity of the source 
of wave-motion could be determined, if the initial period of the waves and 
the velocity of their propagation were known. This illustration sets forth 
the principles on which is founded the method of investigation which is now 
to be described. 
The idea that a change of period in luminous or sonorous waves would 
arise in consequence of a motion of the observer, or of the source of the light, 
* A communication ordered to be printed in cxienso. 
