xxii Preface to the English Edition. 



either to exclude or to infer an unknown de- 

 velopmental force, since we should have had to 

 deal with two unknowns which could in no way be 

 kept separate. 



We escape this dilemma in the markings of 

 caterpillars, because the latter do not propagate 

 in this state. If the phenomena are not here en- 

 tirely referable to Natural Selection and the direct 

 action of the environment if there remains an 

 inexplicable residue, this cannot be referred to 

 Sexual Selection, but to some as yet unknown 

 power. 



But it is not only in this respect that cater- 

 pillars offer especial advantages. If it is to be 

 attempted to trace transformations in form to the 

 action of the environment, an exact knowledge of 

 this environment is in the first place necessary, 

 i. e. a precise acquaintance with the conditions of 

 life under the influence of which the species con- 

 cerned exist. With respect to caterpillars, our 

 knowledge of the life conditions is certainly by no 

 means as complete as might be supposed, when 

 we consider that hundreds of Lepidopterists have 

 constantly bred and observed them during a most 

 extended period. Much may have been observed, 

 but it has not been thought worthy of publication; 

 much has also been published, but so scattered 

 and disconnected and at the same time of such 

 unequal credibility, that a lifetime would be required 

 to sift and collect it. A comprehensive biology of 



