948 



SCIENCE 



[N. S. Vol. XXX. No. 783 



substance daily as well as loses, but so far 

 as present knowledge goes, its gain is 

 greatly inferior to its loss. So long- as the 

 heat of the sun was supposed to be de- 

 pendent on ordinary chemical changes, or 

 on the fall of meteorites, or on self -contrac- 

 tion, an activity adequate for terrestrial 

 life could only be estimated at a few mil- 

 lion years. But recent discoveries in 

 radio-activity have revealed sources of 

 energy of an extremely high order. In 

 the light of these the forecast of the sun's 

 power to energize the activities of the at- 

 mosphere dependent on it and to warm 

 the earth is raised to an indeterminate 

 order of magnitude. 



If we may thus find grounds for a com- 

 placent forecast in reciprocal actions 

 on the earth and in reciprocities between 

 the earth and the sun, are we free from 

 impending dangers in the heavens without ? 



Present knowledge points to one tangible 

 possibility of disaster; collision with some 

 celestial body, or close approach to some 

 sun or other great mass, large enough to 

 bring disaster by its disturbing or dis- 

 ruptive effects. Within the solar system, 

 the harmonies of movement already es- 

 tablished are such as to give assurance 

 against mutual disaster for incalculable 

 ages. Comets pursue courses that might, 

 theoretically at least, bring about collision, 

 but do not appear usually to possess masses 

 sufficient to work complete disaster to the 

 life of the earth even should collision occur, 

 whatever local disaster might follow at the 

 point of impact. The motions of the stars, 

 however, lie in diverse directions, and col- 

 lisions and close approaches between them 

 are theoretically possible, if not probable, 

 or even inevitable. There are also in the 

 heavens nebulaj and other forms of scat- 

 tered matter, and doubtless also dark 

 bodies, which may likewise offer possibili- 

 ties of collision. The appearance of new 



stars flashing out suddenly and then gradu- 

 ally dying away suggests the actual occur- 

 rence of such events. It has been even 

 conceived that the close approach of suns 

 is one of the regenerative processes by 

 which old planetary systems are dispersed 

 and new systems are brought into being. 

 One phase of the planetesimal hypothesis 

 is built on this conception and postulates 

 the close approach of some massive body 

 to our ancesti'al sun as the soiirce of dis- 

 persion of a possible older planetary sys- 

 tem and the generation of the nebulous 

 orbital condition out of which our present 

 system grew. However this may be, it 

 must be conceded that in collision and close 

 approach lie possibilities, if not probabili- 

 ties, of ultimate disaster to the solar system 

 and to our earth. But here, as before, the 

 vital question lies in the time element. 

 How imminent is this liability? The dis- 

 tances between stars are so enormous that, 

 though they move diversely, the contingen- 

 cies of collision or disastrous approach are 

 remote. Nothing but rough computations 

 based on assumptions can be made, but 

 these make disaster to a given sun or sys- 

 tem fall on the average only once in bil- 

 lions of years. There is no star whose 

 nearness to us, or whose direction of mo- 

 tion is such as to threaten the earth at any 

 specific period in the future. There is 

 only the general theoretical possibility or 

 probability. While, therefore, there is to 

 be, with little doubt, an end to the earth 

 as a planet, and while perhaps previous to 

 that end conditions inhospitable to life may 

 be reached, the forecast of these contin- 

 gencies places the event in the indetermin- 

 ate future. The geologic analogies give 

 fair ground for anticipating conditions 

 congenial to life for millions or tens of 

 millions of years to come, not to urge the 

 even larger possibilities. 



But congeniality of conditions does not 



