251 
intelligent and also successful, it would become the object 
of imitation by others. 
Nor is it possible that Natural Selection should be 
required to complete what intelligence had begun. For 
Natural Selection works by life and death, and it is 
difficult to understand how the intelligent animal which 
had faced previous crises with success should yield itself, 
the passive victim of fate, at any subsequent period of 
its history. 
(4) The fourth way of explaining how Natural Selection 
can act in the development of instincts so as to escape the 
interference of intelligence, is to assume that the intelligent 
animal may be unobservant of certain actions which are 
not intelligently performed, and are not of an adaptive 
character, but which are nevertheless inherited and fixed in 
the race. Mr. Romanes says that one mode of the origin 
of instincts 
“consists in Natural Selection, or survival of the fittest, con 
tinuously preserving actions which, although never intelligent, yet 
happen to have been of benefit to the animals which first chanced 
to perform them.”—(.Mental Evolution in Animals, p. /77.) 
The proof which he gives of this assertion is:— 
“ 1. That non-intelligent habits of a non-adaptive character occur 
in individuals. 2. That such habits may be inherited. 3. That such 
habits may vary. 4. That when they vary the variations may be 
inherited. 5. That if such variations are inherited we are justified in 
assuming, in view of all that we know concerning the analogous case 
of structures, that they may be fixed and intensified in beneficial 
lines by Natural Selection.”—{Ibid. p. /So.) 
It should be observed that the selection of non-intelli 
gent habits of a non-adaptive character is cited as a proof 
of the selection of actions which, although never intelligent, 
yet happen to have been of benefit to the animals that 
first used them ! The two expressions are not identical, 
unless we interpret habits of a non-adaptive character