152 EXPERIMENTAL EVOLUTION LECT. 



repeated fact that in many cases while two forms of 

 life, fossil and extant for instance, seem to be directly 

 or genetically related to each other, some link seems 

 to be missing through which the transition from the 

 one to the other could be effected. In Mr. Ball's 

 opinion no links are really missing in such cases, 

 because none have existed ; the much asked for 

 transitional forms have never lived. He supposes 

 that sudden and considerable variation may take 

 place under the cumulative influence of a number of 

 small causes which remain a long time inoperative, 

 but become efficient when at last one more slight in- 

 fluence is added, which acts like the straw that breaks 

 the camel's back. A good deal may be said in favour 

 of this view, and many facts from zoology and 

 palaeontology go to support it, and it may be that 

 there is more truth in this manner of explaining 

 some facts than in the former opinion. At all 

 events abrupt or sudden variation is not an un- 

 known fact. Some years ago, in France, a variety 

 which stands between Begonia Schmidtii and B. 

 semperflorens made its appearance quite suddenly in 

 different points at the same time ; at Paris, at 

 Poitiers, at Lyons, at Marseilles, etc. At Lyons the 

 fact was very striking, as 500 specimens of the plant 

 suddenly exhibited the new characters, and these 

 characters were found in a plant which had been a 



