^1^- BIOLOGICAL THEORIES. 503 



individuals have, and their jaws are more powerful. The soldiers of 

 one generation are indistinguishable from those of any other ; their 

 specific characters are constant. This constancy of specific charac- 

 ters, this reproduction, generation after generation, of individuals 

 indistinguishable from one another, this likeness among individuals 

 of different generations, is a phenomenon belonging to the class I 

 have referred to. Whether it is " heredity " or not all depends upon 

 the definition of the term. If " heredity " means the transmission to 

 the offspring of characters possessed by the parent, then this is not 

 heredity, for the soldier termite is very unlike its parents, and also 

 unlike all its remoter ancestors. All its ancestors, so far as we know, 

 for thousands of generations, have possessed wings and eyes, and 

 the power of reproduction, while none of them have possessed the 

 enormous head and powerful jaws which form a most striking 

 character of the soldier. The soldier is exactly like millions of other 

 individuals from which it is not descended, and which cannot, there- 

 fore, have transmitted their characters to it ; it is unlike all those 

 which might have been supposed to transmit these characters, but 

 for the fact that they have never possessed them. 



This example is not given to disprove any other theory, and I 

 do not think that it would be altogether impossible to explain it on the 

 assumption of pangenesis or of continuous germ-plasm, as satisfactorily 

 as those assumptions explain more commonplace forms of heredity. 

 I give the example merely to aid in the exposition of my own view. 



If we examine into the factors which by their co-operation have 

 determined the form and structure and consequent functions of the 

 soldier termite and of its parts, we may recognise among them several 

 classes. Whether the soldier develops from a fertilised eg^ or not I 

 do not know ; we must, therefore, to be on the safe side, assume that 

 it at least may do so. The factors may then be classed as follows : — 



1. Those determining the constitution of the ovum at the time 



of fertilisation. 



2. Those determining the constitution of the spermatozoon at 



the time of fertilisation. 



3. Those acting upon the fertilised ovum and the developing 



soldier from without. 

 Among the factors determining the constitution of the ovum are 

 to be included the constitution of the female parent. Its structure 

 and consequent functionality — for I include chemical as well as 

 physical constitution in " structure " — have been determined by 

 factors which might be classed in the same way as we have classed 

 those determining the structure of the soldier. The origin of these 

 factors I shall have to discuss in a future paper ; for the present, we may 

 take the structure of the parent as our starting point, and similarly also 

 with the male parent. One result of that structure and consequent 

 functionality of the mother is the production of ova. Though all 

 were formed in one ovary, it would not follow that all should be alike, 



