'9'-] OF THE GLOBULAR CLUSTERS. 153 



spaces still large, and remember also the vast interval of time re- 

 quired to produce the ijiz'isiblc state of close compression, it will 

 become evident that the chances of our living at the epoch of a 

 cluster conflagration totally disappears, and the most we can hope to 

 recognize is the resulting giant star such as Canopus. 



XIV. Proof that ]\Iatter Actually is Lost from our Sun 

 Supplied by the A^ery Straight Tails of Comets Devel- 

 oped IN Close Perihelion Passage. 



In view of the recent development of the doctrine of repulsive 

 forces in nature, it becomes important to have readily at hand specific 

 illustrations of these forces adequate to meet any demand that may 

 be made on the new doctrine. Now the tails of comets and the 

 streamers of the corona, as explained by Arrhenius in Lick Observa- 

 tory Bulletin 58, give abundant evidence of the operation of repulsive 

 forces directed from the sun ; but every case does not show a repul- 

 sion sufificientl)^ strong to carry the particles away from our solar 

 system on parabolic or hyperbolic paths. The question thus arises : 

 Are there any known cases of repulsion sufiiciently powerful to carry 

 particles away from our sun to the other stars, and thus cause a 

 secular decrease in the sun's mass? We may answer this question 

 in the affirmative, for the following reasons : 



1. Those comets which have had a very small perihelion distance, 

 as the great comets of 1680, 1843 ^^'^ 1882, have all had also very 

 straight tails, which were found by calculation to be of immense 

 length near perihelion passage. It is well known that this extreme 

 straightness of tail indicates very powerful repulsion of the particles 

 composing it. 



2. By actual calculation I have established the fact that the veloc- 

 ity of the particles in the tails of the above comets, at perihelion, 

 exceeded the parabolic velocity of a body driven away from the sun. 

 The matter in these tails therefore was not only diffused over the 

 solar system, but also carried away to other fixed stars. 



3. Now if this repulsion with more than parabolic velocity could 

 happen for vaporous matter developing in a comet's tail near peri- 

 helion, but remaining of sufficient density and luminosity to be visible 



