302 PROBLEM OF THE ORIGIN OF SPECIES. 



reach a point beyond which there will not be innumerable causative stages 

 which our perceptions can not discover. In this sense, then, the above law 

 of variation seeks only to express concisely and briefly the stages in their 

 general sequence as I perceive them. 



Variation, then, in organisms resolves itself into trial and error. The trial 

 we can not conceive of as arising dc novo in the organism, but there must 

 have been an antecedent stage, which we call a stimulus, and it is shown 

 that when a stimulus is applied in experiment to organisms there is a trial 

 in that there follows a response (movement, modification) dependent upon 

 pre-existing stages in the constitution of the organism, and more remotely, 

 perhaps, upon the nature of the stimulus. Thus we see that even though we 

 can express the sequence of stages briefly by this law, and find it of use in 

 systematizing our perceptions of variation, in any single case, it simply states 

 what we may expect to happen when an organism meets the condition of 

 its existence. 



This discussion brings us to the point where we see that variation is the 

 response or sequence of stages following the meeting of two preexisting 

 series of stages, one in the material world environment, the other within the 

 organism. Obviously, if we hope to state more clearly the sequence of stages 

 in variation phenomena, such as orthogenetic variations, fluctuating varia- 

 tions, skew variations (or "mutations"), we must understand more fully the 

 developmental principles underlying the antecedent stages within the organ- 

 ism before they meet the causes of the variations. It has been demonstrated 

 in my experiments that there is no specific response (variation) following any 

 given stimulus; hence I conclude (Chapter III) that the environment acts 

 simply as a stimulus, and, therefore, its sequence of stages is of no impor- 

 tance. It is shown (Chapters III and V) that the response (variation) follow- 

 ing stimulus is entirely determined within the organism. In the organism, 

 therefore, there are series of steps (stages) whose sequence and relation we 

 must understand before we can begin to get at the questions of orthogenetic 

 variation, "mutation," etc. This all-important series of states can be investi- 

 gated only through the study of development. Any other plan of attack 

 seems utterly hopeless. 



The first and one of the most fundamental questions which we have to 

 consider is whether development is an epigenetic or a predetermined process ; 

 upon how we answer this question will depend the progress we shall be able 

 to make in the analysis of the phenomena of variation and evolution. 



That development in Lcptinotarsa is an epigenetic process has been demon- 

 strated in the study of their embryonic development and coloration. In Chap- 

 ter III it was shown that there is no predetermination of the coloration, but a 

 development, unfolding, or specialization based upon a general phyletic system 



