252 
to mean really useful habits, the usefulness of which 
the animals did not observe. But we cannot accept this 
interpretation of the phrase, because from the illustrations 
given of the non-intelligent habits of a non-adaptive 
character, we see that they are for the most part assumed 
to be useless, foolish, unintelligent. 
It will conduce to a clearer understanding of the 
question if we regard these two passages as suggesting two 
entirely different methods by which instincts might be 
developed by Natural Selection. Now, with respect to 
the first statement, we have to deal with variations which 
are from the first useful to the race ; and our contention is 
that the intelligent animal will observe this utility, and so 
anticipate the action of Natural Selection, especially when 
such utility is a matter of life and death, as it must be the 
very moment that Natural Selection comes into action. 
The second mode of stating the case assumes that 
certain habits, which display no intelligence on the part of 
the animal, and are not adapted in any way to the con 
ditions in which it lives, may have arisen in connection with 
intelligent animals, so that when a change of conditions 
makes them of use, they may come under the influence of 
Natural Selection. 
Now, in order to justify this argument, we must first be 
sure that such habits are unintelligent. We must next be 
well assured that these actions could have arisen de novo 
and are not the result of some habit which was once more 
adaptive than it is now; we must be sure that such actions, 
supposing that they could have been developed in the way 
supposed, would become subject to the influence of Natural 
Selection when, through change of circumstances, they 
become useful. Now the first point to be observed is that 
some of the instances cited are not non-intelligent and 
non-adaptive habits. The Abyssinian pigeon when fired