570 Mr. Arthur Stanley Eddington [March 26, 



move under the same law so far as can be ascertained. But the tail 

 particles of the comet do not seem to recognise this force, or, at least, 

 with them it is more than counterbalanced by a repulsion ; for them 

 the sun behaves as a centre of repulsion and urges them away. It 

 was for long a great puzzle to understand how the sun can thus play 

 a double role, but various plausible explanations are forthcoming, some 

 of which I shall consider presently. The difficulty is now not so much 

 to suggest an explanation as to decide between rival explanations. 



A great deal of work has already been done on this subject in 

 the case of previous comets. Of modern workers the name of 

 Bredichin stands out pre-eminently. Much of his work was done 

 thirty years ago— of course, derived from visual observations — but 

 it is still of fundamental importance. By studying how much the 

 tail lagged behind the radius vector (by the radius vector is meant 

 the direction from the sun to the comet, which is, of course, the 

 direction in which the solar repulsion acts), and also the curvature 

 of the tail, Bredichin was able to form a measure of the amount of 

 this repulsive force. It is convenient to compare this repulsion 

 with the ordinary attractive force of gravitation of the sun in the 

 same position (the force which the head of the comet is experiencing) 

 and use that as a unit. It was found that the tails of comets could 

 be divided into three classes, for which the repulsive force was 

 respectively 18, 2*2 to 0*5, and 0-8 to in terms of this imlt. 

 Very often a comet would have two or three distinct tails, so that 

 all three classes might be represented in a single comet. 



The inti'oduction of photography has naturally improved the 

 methods by which the forces may be determined. Perhaps the most 

 remarkable researches are those of Jaegermann ; he has, in the case 

 of several comets, proved the existence of repulsive forces in just 

 the' same way that the existence of the ordinary attractive force on 

 a planet is proved — namely, by calculating the orbit in space of the 

 matter acted on by the force. To show how this may be done. [Two 

 photographs of the present comet taken three hours apart were shown 

 combined on one plate, so as to show the motion amongst the stars 

 in that interval.] The head of the comet is moving in one orbit ; 

 but there is a detached mass, which was evidently projected from 

 the head some hours before, which is moving in a different orbit. 

 Now from four observations of the position of this head, we can 

 calculate the orbit and the force exerted on it by the sun (I say four 

 observations, and not three, because I want to determine the force 

 exerted by the sun, and not assume it as known). Similarly, from 

 four observations of this detached piece we might calculate its orbit 

 and the force exerted on it by the sun. Jaegermann has examined in 

 this way detached portions of former comets, and shown that the 

 detached portions pursue hyperbolic orbits convex to the sun, showing 

 that they are repelled from it. 



The results of determinations of the repulsion, whether found in 



