140 ELEMENTS OF BIOLOGY. 



m 



the species, would, to begin with, be so minute as to be use- 

 less, if not injurious, in which case their preservation and 

 ultimate intensification must have been caused by something 

 else than Natural Selection alone. 



2. Whilst Natural Selection cannot 'initiate even the 

 smallest variations, the behef in its being a constant and 

 universal agent in modifying all living beings, requires that 

 variations should be continually occurring, and that they 

 should not be extensive in amount. The probability, how- 

 ever, that all variations depend upon some internal law far 

 below the surface, and unconnected with outside conditions, 

 is greatly increased by the undoubted occurrence of sudden 

 and striking variations, for which no cause can be shown, 

 and for which Natural Selection is unable to account. 



3. It has been shown that it is not sufficient for the pro- 

 duction of a new breed or variety simply that a favourable 

 variation should occur, unless the change should occur 

 simultaneously in a greater or less number of individuals 

 of the species. However favourable a variation might be, 

 there would be little or no chance of its being perpetuated, 

 unless it presented itself in more than one individual at the 

 same time. But the probabilities are enormously against 

 the simultaneous appearance of the same variation in numer- 

 ous individuals of a species. We are thus led to doubt if 

 even highly favourable variations would necessarily, or even 

 probably, end in the establishment through natural selection 

 of a permanent new breed or variety. 



4. The same parents may give rise to several groups of 

 individuals which differ very widely from one another, and 

 from their parents in their characters, but which are sexless, 

 and are therefore unable to transmit their peculiarities to 

 future generations. Thus the workers and 'the soldiers 

 amongst the Termites differ greatly both from one another 

 and from the fertile individuals, both in their actual struc- 

 ture and in their instincts ; and yet both are neuter and 

 have no power of transmitting their peculiarities by the 



