Seems that I recall that Morris mentioned this theory in the book (I don’t think he was the first to propose it, but I could be wrong), but as I read it a long time ago and am too lazy to go look it up (today, anyway), I can’t tell you the specifics. However, I don’t give it much credence. IIRC, the hypothesis relied on arguments about the possible (former) adaptive values of some rather unconnected features (e.g. hair distribution, supposed “webbing” between fingers, etc.); but I think it is amply contradicted by the fossil record, human developmental patterns, etc. I think it was proposed when very little was known about human evolutionary history.
What we really haven’t discussed yet is the basis for the apparent difference between apes and humans in swimming ability. Is human swimming ability entirely learned, or is there evidence for some innate ability?
At least for adults and older children, humans show little more instinctive swimming ability than apes; an adult or toddler that has never learned to swim will usually quickly drown in deep water (although as I remarked earlier possibly able to save him/herself in circumstances where an ape would drown, such as in a narrow moat; but this could be due more to intelligence than instinct).
On the other hand, I have heard that very young human infants, pre-walking, will swim, apparently instinctively, when placed in water. Can anybody confirm/contradict this? If so, it would be very interesting. If it exists, at what age is this ability lost?
While I stated earlier that even young apes appear to be unable to swim, I doubt that the experiment has ever been made on a very young ape. It would simply be too risky to place a valuable infant gorilla or chimp in the water.