28 THE EVOLUTION OF THE METAZOA 



Something About the Methods That Can Be Used 



in Connection with the Construction of the Natural 



Animal System 



It should first be mentioned that we are discussing here 

 a phylogenetic construction which should form a basis of an 

 attempted reform of the system. This touches the very roots 

 of the concepts that have so far prevailed with regard to 

 classification, and the consequences can be of considerable 

 importance. Furthermore, it should be mentioned, even if it 

 seems superfluous, that I base my work constantly on the 

 neo-Darwinian principles of evolution. I find any idealistic 

 morphology, or any system based on it, and any purely idea- 

 Hstic taxonomy, unacceptable ; they are both out of date as 

 well as unscientific. Modern zoologists are endeavouring to 

 construct a natural animal classification which, in agreement 

 with this, must be based on the supposed phylogenetic develop- 

 ment. The results, for better or w^orse as they are achieved, 

 depend above all on our present positive knowledge— which 

 is an objective element — and, naturally, to a certain extent 

 on the scientific w^orker— which is an unavoidable subjective 

 element. We can make no progress either with agnosticism or 

 with an ultraconservatism, if it is Unked with a complete 

 absence of the creative imagination. It is therefore necessary to 

 form a reaUstic concept even in those cases where the obstacles 

 we have to overcome are due to our present insufficient 

 knowledge. In this connection we must not be afraid to dare 

 to conjecture or to use our imagination. Yet at the same time 

 we have to be aware of the fact that these are hypothetical 

 constructions only. They have not been made for all eternity. 



It would be detrimental to our progress is we reject on 

 principle and without discussion, any more or less daring 

 construction; if we declare any work of this kind to be a pure 

 waste of time; and if we look upon it disparagingly as is 

 frequently the case. The numerous and mutually different 

 attempts to construct the animal system that we have met even 



