20 EVOLUTION OF LIVING ORGANISMS. 



mentation, infusoria, and suchlike forms, were supposed 

 to develop de novo in liquids of appropriate composition. 

 But Pasteur and others in the last century definitely 

 proved that such organisms really arise from germs or 

 spores originally present in the liquids ; that neither 

 putrefaction nor fermentation will take place, nor organ- 

 isms of any kind appear, in substances which have been 

 thoroughly sterilised, or from which all living germs 

 have been rigidly excluded. So far as we know, living 

 organisms at the present day do not develop spontane- 

 ously, but are all derived from pre-existing organisms. 



We must, however, suppose that at some period in the 

 earth's history, when conditions were favourable and 

 perhaps very different from those of the present time, 

 living protoplasm made a first appearance. Possibly 

 these conditions will never be repeated, either in nature 

 or in the laboratory, and the first stages in the evolution 

 of life may never be discovered. The temperature, 

 moisture, pressure, and other conditions must have been 

 such as to allow of the formation of high compounds of 

 various kinds. Many of these would be quite unstable, 

 breaking down almost as soon as formed ; others would 

 be stable, and merely persist and accumulate. Still 

 others might, possibly with the help of some katalytic 

 substance, tend to reform as fast as they broke down. 

 Once started on this track such a self -repairing compound 

 or mixture would inevitably tend to perpetuate itself, and 

 might combine with, or " feed " on, other compounds less 

 complex than itself, as was long ago suggested by 

 Lankester. For any chemical action will continue so 

 long as the conditions are favourable ; a trail of gun- 

 powder will inevitably explode from end to end provided 

 it be continuous, and will go on burning so long as 

 powder is supplied. 



The principle of the survival of the fittest applies with 

 all its force to such initial steps in the evolution of life. 

 The more completely self-regulating mixtures would out- 



