Habit and Instinct. 435 



may therein develop, probably in the brood- sac, and may 

 even destroy the reproductive powers of the host for the 

 future good of her own offspring these and many others 

 would seem to have no basis in individual experience. 



In illustration of the second class of instincts, those 

 due to lapsed intelligence, Mr. Eomanes cites the case of 

 birds living on oceanic islands, which at first show no fear 

 of man, but which acquire in a few generations an instinc- 

 tive dread of him for the wildness or tameness may 

 become truly instinctive. "If," says Dr. Kae,* "the eggs 

 of a wild duck are placed with those of a tame one under a 

 hen to be hatched, the ducklings from the former, on the 

 very day they leave the egg, will immediately endeavour 

 to hide themselves, or take to the water if there is any 

 water, should any person approach, whilst the young from 

 the tame duck's eggs will show little or no alarm, indicating 

 in both cases a clear instance of instinct or 'inherited 

 memory.' " 



It must not be supposed that these two modes of origin 

 are mutually exclusive, and that any particular instinct 

 must belong either to the one class or the other. On the 

 contrary, many instincts have, as it were, a double root 

 the principle of selection combining with that of lapsing 

 intelligence in the formation of a joint result. Intelligence 

 may thus give a new direction to a primary instinct, and, 

 the intelligent modification being inherited, what is prac- 

 tically a new instinct may arise. Conversely, selection 

 may tend to preserve those individuals which perform 

 some intelligent action, and may, therefore, aid the lapsing 

 of intelligence in establishing and stereotyping an instinct. 

 Referring the reader to Mr. Romanes's work for the 

 examples and illustrations by which he enforces his views, 

 we may now proceed to consider the subject in the light of 

 recently developed theories of heredity. 



We have seen that a school of biologists has arisen 

 who deny the inheritance of acquired characters. But Mr. 



* Nature, vol. xsviii. p. 271 2 quoted in " Mental Evolution in Animals," 

 footnote, p. 196. 



