( 525 ) 



is expressly menlioned. 11" the molioii of dineren 1 classes of' slurs 

 relative to the solar system is diiFereiit, the absolute motion must be 

 (lifTerent too. 



How can we imagine such a different motion for our star-classes? 



Are we to imagine, that instead of the two star-clouds which were 

 assumed up to the present, there are six — I confine myself to the 

 three spectral classes really investigated — of which three move 

 permanently in slightly different directions, and three others approxi- 

 mately in another? Of the three former of which the first would 

 contain only B stars, the second only A stars, the third only G stars, 

 whereas for the latter set of three there is a similar distribution? 

 Six wholly independent streams between which there would be 

 differences in velocity, number and amount of the peculiar velocity ^) 

 gradually changing with the age of the stars. 



All this seems hardly admissible. It seems infinitely simpler to 

 assume, as I did in my April lecture, that originally there were 

 only two star-clouds. That in course of time the stream-direction 

 and the stream-velocity have slightly changed in such a way, that 

 the oldest stars have deviated most, the youngest least, but all in a 

 higher or lower degree from the original direction and velocity. 

 In that lecture I also tried to sho^v how we can imagine such a 

 change in motion to have taken place. 



I do not wish to enter again into this explanation, but I wish to 

 draw attention to the fact that, unless w^e adopt the hypothesis of 

 the six permanent streams, alluded to just now, we have to admit 

 that the matter which originally composed the two star-streams is 

 expanding. 



For a better understanding of this expansion, it is necessary to 

 know the elements of the two streams separately for the stai'S of 

 the different classes of spectrum. 



The most reliable elements, which up to the present I have been 

 able to derive, are as follows: (see p. 526). 



In this table the galactic latitudes and the velocities relative to 

 ths sun of the two streams have been denoted by the letters (ij i\ {i.^ i\. 



The reliability of these elements is very different and for the 

 greater [)art rather small. Those for the G- and B-stars of stream I 

 are the most trustwoj'thy. For the second stream of the B-stars, of 

 which we still know but a few members, the position of the vertex 

 is [)articularly uncertain, the stream velocity jiractically unknown. 



For the change of the second stream \n ith time, we have therefore 



'j See the leclure quoted above. Congres of physicistö and physicians April 1 UU. 



