VOL. LXXXVI.] PHILOSOPHICAL TRANSACTIONS. ,fo 



periodical changes in the lustre of the stars, so satisfactorily, that we certainly need 

 not look out for any other cause. Let us, however, take a review of any ob- 

 jections that might be made. 



The periods in the change of the lustre of Algol, (3 Lyrae, S Cephei, and * An- 

 tinoi, are short; being only 3, 5, 6, and 7 days respectively: those of o Ceti, the 

 changeable star in Hydra, and that in the neck of the Swan are long, amounting 

 to 331, 394, and 497 days. Will not a doubt arise whether the same cause 

 can be admitted to explain indiscriminately phenomena that are so different in their 

 duration? To this it may be answered, that the whole force of the objection is 

 founded on our very limited acquaintance with the state of the heavens. Hitherto 

 we have only had 7 stars whose periodical changes have been determined. No 

 wonder then that proper connections between their different periods were wanting. 

 But let us now place a Herculis among the list, which is not less than 60 days in 

 performing one return of its changes. Here we find immediately, that the step 

 from the rotation of « Herculis to that of Ceti, is far less considerable than that 

 from the period of Algol to the rotation of « Herculis; and thus a link in the 

 chain is now supplied, which removes the objection that arose from the vacancy. 



There is however another instance of a slow rotatory motion ; and it is doubly 

 instructive on this occasion. In a former paper it has been shown, that the 5th 

 satellite of Saturn revolves on its axis in 79 days; this not only shows that very 

 slow rotatory motions take place among the celestial bodies; but from the argu- 

 ments that were brought to prove its rotation, which I believe no astronomer will 

 oppose, we are led to apply the same reasoning to similar appearances among the 

 fixed stars. A variation of light, owing to the alternate exposition of a more or 

 less bright hemisphere of this periodical satellite, plainly indicates that the similar 

 phenomenon of a changeable star, arises from the various lustre of the different 

 parts of its surface, successively turned to us by its rotatory motion. The rota- 

 tions of the sun and moon, and of several of the planets, become visible in a 

 telescope by means of the spots on their surfaces; the remote situation and small- 

 ness of the 5th satellite of Saturn leave us without this assistance; but what we 

 can no longer perceive, with our best optical instruments, we now supply by rational 

 arguments. The change in the light of the satellite proves the rotation; and the 

 rotation once admitted, proves the existence of spots, or less luminous regions on 

 its surface, which at setting off were only hypothetical. In the same manner a 

 still more extended similarity between the sun and the stars offers itself, by the 

 spots that now must also be admitted to take place on their surfaces, as well as on 

 that of the sun. 



To return to the difficulty which has been started, it may be further urged, that 



there are some reasons to surmise that the 34 Cygni is a periodical star of 18 years 



return * ; and that other stars seem very slowly to diminish their lustre, and may 



probably recover it hereafter. In answer to this, I remark that it will not be neces- 



* Phil. Trans, for the year 1786, page 201.-— Orig. 



