20 ESSAYS SCIENTIFIC AND PHILOSOPHICAL. 



whole, there are two points which seem to be of 

 special importance. 



i. The first is that Weismann's theory makes 

 a definite advance in attempting to account for 

 variability. Hitherto evolution has been com- 

 pelled to postulate variability, and has neither set 

 limits to the possible variations nor accounted for 

 their origin. Given a practically infinite number 

 of " accidental " variations, evolution sets to work 

 to show why, under the law of natural selection, 

 supplemented by the supposed inheritance of 

 acquired characters, certain forms rather than 

 others survived. But Weismann's theory attempts 

 to explain the existence of variation, and to show 

 that the variations, though innumerable and prac- 

 tically infinite, are neither really infinite nor acci- 

 dental, but from first to last are subject to law. In 

 the lower asexual organisms variation is brought 

 about by external influences acting on the indi- 

 vidual, these individual characters, though acquired, 

 being as yet transmissible, because reproduction 

 takes place by simple fission. But when the dis- 

 tinction between reproductive and somatic cells 

 arises, the direct action of environment ceases, or 

 rather it affects the individual, not the species. 

 /Sexual reproduction now appears as a new method 

 of multiplying variations. For sexual reproduc- 

 tion, by combining the variations already in exist- 

 ence, increases them in geometrical ratio, and thus 



