iNCRKASINt; PkiUoD OK ClOIt'lAIN Cl.O.SK iJiNAItV StaKS. .'{3 



one of whitli would l)i> ;i sulHciciit rxplanatioii of tlu- plienomeiKjii 

 described. 



(1) If V^ Puppis ov f-i Lyme exhibited a variation in period of a 

 secular character, now increasing, now decreasing, or rather increasing- 

 and tiien decreasing through l(»ng intervals of time, we might suppose 

 either star to circle round some remote but massive central orb. 



There arises at once many arguments against this view. There is 

 nc> indication in the case of either \' Puppis or /3 Lyrae of any such 

 central star. Then, again, the probability that V Puppis, /5 Lyrae, or 

 any of the other stars of this class would decrease in period is as great 

 as that they should increase, on the assumption that eacli system moves 

 round sinne central body. But there is no evidence with regard to any 

 st^r of this class of a decreasing period. There are other arguments 

 against this explanation, over which we may not tarr}'. 



(2) In the paper already referred to — The Obsn-i^atori/, No. 367, 

 p. 100— I said: — 



'' It seems reasonal)Ie to consitler the progressi\e change in the 

 pei'iod of /3 Lyrae as due to the slow recession of the component stars 

 from i>ne another. A very slight recession — -one-thousandth part of 

 the radius of the orbit — would account for the augmentation in time, 

 30 min. in a century. Thus if we suppose the bipartition of the stars 

 to have taken place in 1780 (this is simply a tlieoretical assumption), 

 the ratio of the radius of the orbit of the system to the sum of the radii 

 of the two components will be a i^atio of unity in 1780. This ratio in 

 succeeding years will be : — 



1800 ... ... 1-00061 



18-JO ... ... 1-00099 



1840 ... ... 1-00123 



1860 ... ... 1-00144 



1880 ... ... 1-00163 



1900 ... ... 1-00180" 



This, with regard to the case of /3 Lyrae. When we come to the 

 increasing period of V Puppis this line of argument would 3'ield us 

 the following conclusions. A recession of one ten-thousandth part of 

 the radius of the orbit is quite sutiicient to account for the whole 

 of the increase of V Puppis during the twentieth century— a very 

 small recession truly I 



A very minute recession during each period — less than a mile — 

 would account for the uniform acceleration of one-thousandth of a 

 second every circuit of the component stars. 



This explanation of the cause of the increasing period of V Puppis 

 and 13 Lyrae is, also, in accord with the deductions of theoretical 

 research. Darwin in his masterly arguments has conclusively proved 

 that if two stars circle round one another, thereby raising tides, the 

 component stars are bound to recede from one another. It is not a 

 questioii of they may. They must ! 







