November 21, 1913] 



SCIENCE 



723 



amongst others, two very remarkable facts. 

 The first is that, ignoring a few, though 

 significant exceptions, if the stars of such 

 groups are arranged in the order of their 

 brightness, we find that they are at the 

 same time approximately arranged in the 

 order of the spectral classes. As an in- 

 stance, take the Scorpius-Centaur group. 

 We find that the very brightest stars are of 

 the earliest helium type; the somewhat 

 fainter ones are of the older helium type; 

 the next fainter ones are of the next stage 

 in the stellar life, or the first type. If we 

 can not follow the series further on to the 

 second and perhaps the third type, this is 

 probably due to our lack of knowledge of 

 the fainter stars belonging to the group. 

 In the Pleiades, where we have a somewhat 

 more extensive knowledge of the fainter 

 stars, we can follow the series at least until 

 in the middle of the second type stars. It 

 follows from this that in all these groups, 

 what there is of helium stars can not be 

 overlooked, for they all are of the very 

 brightest stars, and our knowledge of the 

 brightest stars is pretty complete. 



Notwithstanding this — and this is the 

 second remarkable fact, the fact that bears 

 directly on the question in hand — we find 

 not a single helium star, neither in the 

 Hyades nor in the Ursa Major group. The 

 stars in these groups show the same grad- 

 ual change of spectrum with the brightness, 

 but instead of beginning with the earliest 

 helium stars, the series begins abruptly 

 with the second stages of a star's life. In 

 the Pleiades the series begins somewhat 

 earlier; still here too there is not a single 

 star of the earliest helium type. It is only 

 in the Scorpius-Centaur group that we find 

 the complete series. 



Our second stream, therefore, behaves 

 much as do the local groups of the Hyades 

 and Ursa Major. The explanation must, in 

 all likelihood, be the same in both cases. 



How, therefore, does it come to pass that 

 in such groups as those of the Hyades and 

 the Ursa Major, the helium stars are abso- 

 lutely wanting? 



For those who, as I did in this lecture, 

 adopt the view of the order of evolution as 

 helium, first, second, third type, there can 

 be no question but that the stars which we 

 now see are first type stars, must in past 

 ages have been helium stars. 



Therefore, such a group as the Hyades, 

 which now-a-days does not contain any 

 helium stars, but which contains first type 

 stars, must in past ages have contained the 

 helium stars in great numbers. Going 

 back in time still further, these helium 

 stars must have been evolved from some 

 primordial matter, probably some nebulous 

 matter. Therefore, in a remote past the 

 groups of the Hyades and Ursa Major 

 must have been full of nebula. As far as I 

 know there is no trace of nebulosity now. 



There thus must have been an epoch in 

 the past that nebulous matter was ex- 

 hausted, had probably all gone into the 

 formation of stars. Since that time evi- 

 dently there could be formed no more 

 helium stars; and as the helium stars that 

 had been formed developed gradually into 

 first type stars we see the necessity of a 

 time in which the groups must not contain 

 any more helium stars. 



Therefore, finally, our answer to the 

 question: how does it come to pass that in 

 the second stream or cloud we find hardly 

 any heliiun stars, would be: because since 

 some time nebulous matter must have been 

 exhausted in this cloud. 



As to the first stream or star cloud, we 

 similarly conclude that the nebulous matter 

 must not yet have been exhausted, or if so, 

 only at a very recent period. 



It has been my aim to show, not that 

 much has been done, but that there is a be- 

 ginning; 7iot that we have entered far into 



