2oU 



SCIENCE. 



[N. S. Vol. XXII. No. 556. 



rise to a condition in which the state is 

 incapable of perpetuation by transforma- 

 tion. This occurs when a savage tribe 

 nearly exterminates another tribe and 

 leads the few survivors into slavery; the 

 previous form of government then becomes 

 extinct. 



The physicist, like the biologist and the 

 liistorian, watches the effect of slowly 

 varying external conditions; he sees the 

 (juality of persistence or stability grad- 

 ually decaying until it vanishes, when 

 there ensues what is called, in politics, a 

 revolution. 



These considerations lead me to express 

 a doubt whether biologists have been cor- 

 rect in looking for continuous transforma- 

 tion of species. Judging by analogy we 

 should rather expect to find slight contin- 

 uous changes occurring during a long 

 period of time, followed by a somewhat 

 sudden transformation into a new species, 

 or by rapid extinction. However this may 

 be, when the stability of a mode of motion 

 vanishes, the physicist either finds that it 

 is replaced by a new persistent type of 

 motion adapted to the changed conditions, 

 or perhaps that no such transformation is 

 possible and that the mode of motion has 

 become extinct. The evanescent type of 

 animal life has often been preserved for us, 

 fossilized in geological strata ; the evanes- 

 cent form of government is preserved in 

 written records or in the customs of savage 

 tribes; but the physicist has to pursue his 

 investigations without such useful hints as 

 to the past. 



The time-scale in the transmutation of 

 species of animals is furnished by the geo- 

 logical record, although it is not possible 

 to translate that record . into years. As 

 we shall see hereafter, the time needed for 

 a change of type in atoms or molecules 

 may be measured by millionths of a sec- 

 ond, while in the history of the stars con- 



tinuous changes occupy millions of years. 

 Notwithstanding this gigantic contrast in 

 speed the process involved seems to be es- 

 sentially the same. 



It is hardly too much to assert that, if 

 the conditions which determine stability of 

 motion could be accurately formulated 

 throughout the universe, the past history 

 of the cosmos and its future fate would 

 be unfolded. How indefinitely far we 

 stand removed from such a state of knowl- 

 edge will become abundantly clear from 

 the remainder of my address. 



The study of stability and instability 

 then furnishes the problems which the 

 physicist and biologist alike attempt to 

 solve. The two classes of problems differ 

 principally in the fact that the conditions 

 of the world of life are so incomparably 

 more intricate than those of the world of 

 matter that the biologist is compelled to 

 abandon the attempt to determine the ab- 

 solute amount of the influence of the vari- 

 ous causes which have affected the existence 

 of species. His conclusions are merely 

 (lualitative and general, and he is almost 

 universally compelled to refrain from as- 

 serting even in general terms what are the • 

 reasons which have rendered one form of 

 animal life stable and persistent, and an- 

 other unstable and evanescent. 



On the other hand, the physicist, as a 

 general rule, does not rest satisfied unless 

 he obtains a quantitative estimate of vari- 

 ous causes and effects on the systems of 

 matter which he discusses. Yet there are 

 some problems of physical evolution in 

 which the conditions are so complex that 

 the physicist is driven, as is the biologist, 

 to rest satisfied with qualitative rather than 

 quantitative conclusions. But he is not 

 content with such crude conclusions ex- 

 cept in the last resort, and he generally 

 prefers to proceed by a different method. 



The mathematician mentally constructs 



