This criticism is based on the fact that the theory of lapsed intelligence demands that the actions which the animals of one generation have acquired by their intelligence should be handed down through heredity to the next generation, and so on.

It is evident that unless this be true it does no good to the species for one generation to do things intelligently, seeing that if the effects on the nervous system are not transmitted to their children, then the next and later generations will have to start exactly where their fathers did, and the actions in question will never become ingrained in the nervous system at all.
Now, the force of this criticism is overwhelming to those who believe--as the great majority of biologists now do[1]--that none of the modifications or so-called "characters" acquired by the parents, none of the effects of use or disuse of their limbs, none of the tendencies or habits of action, in short, none of the changes wrought in body or mind of the parents during their lifetime, are inherited by their children.

The only sorts of modification which show themselves in subsequent generations are the deep-seated effects of disease, poison, starvation, and other causes which concern the system as a whole, but which show no tendency to reproduce by heredity any of the special actions or functions which the fathers and mothers may have learned and practised.

If this difficulty could be met, the theory that intelligence has been at work in the origination of the complex instincts would be altogether the preferable one of the two; but if not, then the "lapsed intelligence" view must be thrown overboard.
[Footnote 1: The matter is still under discussion, however, and I do not mean in any way to deny the authority of those who still accept the "inheritance of acquired characters."] Recent discussion of evolution has brought out a point of view under the name of Organic Selection which has a very fruitful application to this controversy over the origin of instincts.

This point of view is one which in a measure reconciles the two theories.