THE FACTORS OF EVOLUTION 205 



we thereby justified in exploiting this fact as a proof of the 

 ineffectiveness of selection ? In my opinion this would prove 

 an entirely wrong conclusion. 



By artificial selection we are generally able to increase at any 

 rate one defined characteristic. In consequence there arises in 

 the organism a discord, a disturbance of the equilibrium of 

 the correlated parts. The more extreme the development of a 

 characteristic, the greater the disturbance, until finally Nature 

 imposes a strict halt to any further advance. Man has no regard 

 for the well-being of the affected animals, but solely seeks to 

 increase those qualities which are of advantage to him. It is, 

 therefore, entirely wrong if someone as for instance, H. St. 

 Chamberlain regards our thoroughbreds as specific samples of 

 perfection, and seeks to derive from their genesis hints for the 

 improvement (Veredlung) of the human race. But we should 

 rather regard the animals with a one-sided development as cripples 

 which are barely capable of independent life. Would it not 

 sound ridiculous to speak, for instance, of a dachshund as 

 'improved' (Veredelt)? It is only the uniform, harmonious 

 development of all characters which we regard generally, and 

 in particular with regard to man, as improved (edel). 



But let us return to our example. If it should be possible 

 to increase and improve, for instance, in the gooseberry, simul- 

 taneously with the size of the fruit, the strength of the branches, 

 the formation of the sieve vessels, tracheae, roots, circulatory 

 system, etc., selection would in all probability be able to obtain 

 still further success, and would thus remain for ever dissatisfied. 



Many accurate investigations into the variability and the 

 power of increasing it have been made, in particular by botanists. 

 But, as is the case in the experiments with sugar-beet and 

 gooseberries, they have not hitherto led to incontrovertible 

 results; the conclusions, therefore, which have been drawn 

 from these experiments by the different investigators are widely 

 divergent. 



If we examine a number of peas or beans, or the leaves of a 

 shrub, with reference to size we observe a remarkable law : 

 all the sizes vary around a mean. Most numerous are the 

 leaves of a medium size, and the variations become rarer in 

 proportion as they are larger or smaller than the medium size. 



