Intelligent Selection. 147 
it. The fact is that while such dogs as preserve their natural 
instincts and conditions with little impairment might regain their 
original feral condition, those whose variation is extreme would 
simply die out. They could not survive in the struggle for exist- 
ence if immediately removed from the artificial conditions to which 
they have been accustomed. Yet if two widely different varieties of 
dogs were slowly restored to nature, being protected and fed until 
they had learned the art of self-preservation without man’s care, it 
is by no means improbable that they might retain their peculiar 
characters of form, habit, and adaptation to particular food, and if 
interbred for a considerable period might continue to interbreed. 
Though there is no proof of this, there is no disproof. It is an 
open problem, which can be settled only by experiment. The state- 
ment that all variational differences would disappear if any of the 
domesticated species were restored to feral conditions, is an unproved . 
assertion, which cannot be verified without a much wider series of 
scientifically-directed experiments than have yet been made. Doz- 
ens of problems of this kind are settled in men’s minds. Very few 
of them have been settled in fact. 
It will be of interest, in this connection, to consider what has 
actually been done by Intelligent Selection, and the influences which 
have controlled its results. A mere glance at the subject shows us 
that industrial and pecuniary considerations have almost solely been 
at work. Among trees, for instance, the effort has been to select 
fruits of large size, agreeable taste, and early or late maturity. 
Among flower bushes, bright colors and odd shapes of petals, with 
variations in the size of the flower and its number of petals, have 
been the ruling considerations. Commercial value has been the 
sole thought, and superficial variations only have been preserved. 
A scientist would have watched for changes in the character of the 
pistils and stamens of the flowers, and if such appeared, by their 
- careful preservation might in time have produced undoubted new 
species. Yet no extended series of experiments from this scientific 
point of view has been made, so far as the present writer is aware. 
Such changes may be of- comparatively rare occurrence and incon- 
spicuous, yet there can be little question that they occasionally arise, 
and they may be as susceptible to selective processes as any other 
variations. 
Among animals the purposes aimed at by trainers vary in the case 
