300 PROBLEM OF THE ORIGIN OE SPECIES. 



least, variation in organisms is to be explained from its start by the principle 

 of trial and error trial due to stimuli, response according to the nature 

 of the organism. The response, however, is entirely conditioned within the 

 organism, and is manifestly a success or error. If now we are able to dis- 

 cover whether or not variations in complicated organisms are the responses 

 to stimuli, we shall be better able to judge whether organic variations are 

 explicable in conformity with the general law of variability of mathematical 

 and physical science, or as peculiar phenomena sufficient unto themselves 

 and needing undemonstrable hypotheses to explain their origin and workings. 



In the second chapter there is brought together data concerning variations 

 in the genus Leptinotarsa, and the fact is demonstrated that variations in 

 these forms are distributed according to the law of the distribution of error, 

 as are the variations of all organisms thus far studied. In the third chapter, 

 where color characters are used as subjects, it is demonstrated that variation 

 is directly produced by stimuli that from relatively invariable parents, stim- 

 uli produce variable offspring ; and again in the fifth chapter it is shown that 

 variations arise in direct response to stimuli. Similar facts exist in abundance 

 in the literature of the action of temperature and moisture on insects, of 

 pressure or friction, causing callouses and exostoces in Vertebrata, and of 

 changes of plant form through the stimuli of soil, light, and moisture. So an 

 endless recital of facts might be given to show that there is variation in 

 response to stimuli, and I have shown this to be true for these beetles. That 

 all variations are responses to stimuli will be admitted by some [neo- 

 Lamarckians], denied by others. That some variations may be responses to 

 stimuli, others not, will be the contention of the Weismannians. 



In Leptinotarsa it is shown that there are variations which are inherit- 

 able, and others that are not ; and in Chapters III and V it is further shown 

 that the heritable variations as far as discovered arose in the germ plasm, the 

 non-inheritable during development in the soma. Further, the experiments of 

 Chapter V show that heritable variations arise as the response to stimuli 

 applied to the germ plasm. These responses, moreover, follow exactly the 

 law of error in their distribution about the normal mean. I maintain, there- 

 fore, that all organic variations are responses to stimuli, and are not due to 

 inherent tendencies or latencies, or the product of mystical elements. 



The correctness of this conclusion regarding the nature of variations 

 will be immediately granted by many, but the supporters of Weismannism 

 will, I think, be prone to say that while the variations did result from the 

 stimuli used in the experiments, the variations were due to the existence of 

 ids, or determinants, carrying the variations in question, and were by the 

 stimuli employed brought forward, and so on. These elements, and the 

 Weismannian view that an "organ is undoubtedly predetermined in the germ 

 plasm," together with all its activities and variations, "can only" result in one 



