The most major book which deals with this topic is Walter Remine's "Biotic Message".
Walter Remines Pages dealing with the Haldane Dilemma"
Robert Williams' lame "refutation"
Fred Williams' rebuttal of Robert Williams' lame "refutation":
FGU stands for Frequently Given-out Understanding, which is a better description of the material on the talk.origins website than "FAQ". Pronunciation of FGU is obvious...
The Haldane Dilemma itself is not complicated. More like higher arithematic than like higher math in fact; if you can count, you can understand it.
Remine's Simple Version:
Imagine a population of 100,000 apes or "proto-humans" ten million years ago which are all genetically alike other than for two with a "beneficial mutation". Imagine also that this population has the human or proto-human generation cycle time of roughly 20 years.
Imagine that the beneficial mutation in question is so good, that all 99,998 other die out immediately (from jealousy), and that the pair with the beneficial mutation has 100,000 kids and thus replenishes the herd.
Imagine that this process goes on like that for ten million years, which is more than anybody claims is involved in "human evolution". The max number of such "beneficial mutations" which could thus be substituted into the herd would be ten million divided by twenty, or 500,000 point mutations which, Remine notes, is about 1/100 of one percent of the human genome, and a miniscule fraction of the 2 to 3 percent that separates us from chimpanzees, or the half of that which separates us from neanderthals.
In a rational world, that should be as far as most people need to read.
That basically says that even given a rate of evolutionary development which is fabulously beyond anything which is possible in the real world, starting from apes, in ten million years the best you could possibly hope for would be an ape with a slightly shorter tail.
But nobody ever accused evolutionites of being rational. Surely, they will argue, the problem might be resolved by having many mutations being passed through the herd simultaneously.
Most of the answer involves the fact that the vast bulk of all mutations are harmful or fatal. ANY creature which starts mutating willy nilly will perish.
In other words, the evolution "process" here pretty much has to be one "beneficial mutation" at a time, even assuming that such a thing as a "beneficial mutation" exists which, in the real world, is very far from obvious.
In the real world, the common English term for "mutation" is "birth defect", and virtually all mutations have names. None of them are terribly beneficial:
The Wikipedia (very long) list of genetic disorders (another term for "mutations"):
Walter Remine notes that the 500,000 mutation figure he got from having a whole herd of animals drop dead other than for two and the two replace the herd every generation is very far from possible in reality.
Remine notes that:
Haldane's calculation goes something like this (when all the logic is filled in). Take the reproduction rate of a higher vertebrate (especially the problematic species with low reproduction rates, such as cows, humans, apes, whales, elephants). Next, subtract the reproduction rate that must account for random loss in the population (anything that affects the fit and unfit alike. That would include large components of the losses suffered in floods, fires, diseases, famines, and much more). Next subtract 1.0, which is the reproduction rate required merely to continue the population into the next generation. Next subtract the reproduction rate that must account for harmful mutations (because many of the fit progeny suffer harmful mutation too). Next subtract the reproduction rate that must account for the maintenance of polymorphisms in the population. And so forth down the list of costs. When you're all done, you have a reproduction rate of 0.1 leftover (according to Haldane) for paying the cost of substitution. Then, if the total cost is 30, and it is paid in installments of 0.1 per generation, then it takes 300 generations to complete one substitution.
J.B.S. Haldane was a brilliant mathematician and population geneticist and a devout evolutionist in an age when enough was still unknown that believing in evolution wasn't totally stupid like it is now. He nonetheless turned up results which pretty well damned evolution and he assumed he was probably missing something and that better info would turn up later, but he wasn't and it hasn't.
The numbers he got using assumptions which maximally favored evolution were that you could substitute something like 1700 genetic traits max into a population of "proto humans" in 10 million years, i.e. that human evolution would take longer than the entire universe has existed according to the big bang theory.
Evolutionism has to be the stupidest ideological doctrine ever devised by the mind of man. In a rational world it would have been abandoned decades ago. It persists because so many losers use it as a psychic security blanket of sorts.