are necessarily associated with reproduction. If this is 
what he means, he seems to claim for sexual reproduction 
a remarkable power of modification. But apart from this 
difficulty, we must assume that in this passage he wishes 
to exclude the idea that they work with intelligence and 
acquired knowledge ; for he contrasts their instincts with 
man’s acquired knowledge. But this assumption is not in 
accordance with facts. Ants, bees and wasps are intel 
ligent creatures, adapting themselves to new conditions as 
they occur, and teaching others what they have themselves 
learned. It is obvious that the whole difficulty of the 
problem we are now discussing is not as to whether neuter 
insects are intelligent, but whether the habits suggested 
by their intelligence can be inherited. 
But even if these assumptions were not proved to be 
based in error, it would be difficult to understand how this 
phenomenon had been brought about by Natural Selection. 
We have to account for the origin of the neuters, not only 
as sterile females, but also possessed of special organisms 
with which special functions are correlated. We have to 
account for the development of two and even three different 
kinds of neuter insects, and we have to picture to ourselves 
the process by which this complex result has been brought 
about. The explanation given is that the selection which 
takes place is not the survival of the best individuals and 
the destruction of the rest, but the selection of those 
families which reap the greatest advantage through this 
subdivision of labour and the destruction of the rest; 
which results in the survival of those queens which possess 
or inherit the faculty for breeding the most workers. 
This suggests two points to be considered. How does 
one family gain this victory over another? and, secondly, 
what chance is there that the surviving queens of the 
surviving families will themselves be able to survive?