222 FROM NEBULA TO NEBULA 



sun as much as the latter outweighs the earth. Suppos- 

 ing, then, the star to have come to within 46,000,000 miles 

 of the sun (i. e. one-half the earth's mean distance, and 

 one-third way between Mercury and Venus) it must, un- 

 der Newton's rule of cubes, have possessed a mass some 

 40,000 times the solar mass and a diameter of 18,000,000 

 miles! It goes without saying that a body of such im- 

 mensity would have swallowed up our pygmy sun in short 

 order, nor left a vestige behind. Moving the star out 

 farther does not mend matters; if to the earth's distance, 

 its diameter would have to be increased to 36,000,000 

 miles, and if to Neptune's, more than a billion. All this 

 calculation is based on the exaggerated supposition that 

 a tide upon the sun comparable to that the latter pro- 

 duces upon the earth would have sufficed to pry open a 

 crust which, until then, had been able to resist the inces- 

 sant straining of a jinnee a million times stronger. That 

 Doctor Chamberlin himself is thinking of a compara- 

 tively weak tide sufficiently appears from his mild ex- 

 pressions: "For its partner in action let a more massive 

 star be chosen" * * * "only a quite distant approach," 

 and "let it be so dense and inert that its response to the 

 reaction of the sun upon it may be neglected. ' ' 



In one of his veiled allusions to Professor Bicker- 

 ton's theory, our author seeks to emphasize the greater 

 probability of a "near approach" over actual collision, in 

 which contention he is clearly justified. But, though the 

 argument against the probabilities of his own hypothe- 

 sis is thereby relatively weakened, it still remains strong 

 enough to overthrow his, too. In order to facilitate the 

 calculation, let us assume for the maximum field of "ef- 

 fective approach" a diameter of 200,000,000 miles, then 

 the area of our original postulated diaphragm would be 

 to it in the ratio of 200 2 to 26,000,000 2 , or as 1 to 16,900, 

 000,000, which latter number being squared and then mul- 

 tiplied by 50, as previously explained in the discussion of 

 Professor Bickerton's hypothesis, yields the probabilities 

 against even such an approach as 14,265,500,000,000,000, 

 000,000, to one ! 



Next to the supreme problem of the origin, mainte- 



