That's precisely the problem with neo-Darwinism. It doesn't allow for the fact that nature apparently makes huge jumps. Their position is "Natura non salta" (nature does not jump).
a. If all organisms, and all structures in organisms evolved v ia small incremental changes, each change adding to the survivability of the organism, where are all the intermediate forms? They do not exist. Honest paleontologists admit this and say the honest thing: "we don't know." Dishonest ones say the dishonest thing: "Next year, on another dig, I feel confident that we'll discover all the intermediate forms. The theory is solid, so the facts will eventually found to support it."
b. Most important biochemical processes -- the blood-clotting cascade, the vision cascade, and some others -- cannot be reduced to a series of incremental steps. It's the whole cascade or nothing, because none of the intermediate steps add up to anything biochemically significant. I.e., when you put the beads and string in the bag, even granting incremental improvements (one bead on a string, then two beads on the string during the next shake, then three, etc.), the particular process (e.g., vision) needs all the beads on the string IN THE RIGHT ORDER at once. One or two or three -- even in the right order -- don't lead to "partial vision"; it leads to nothing. Magazines with titles like "Molecular Evolution" claim that these partial strings were useful for some other process (unknown, unknowable, unstated, unproved), and then transferred over to the process of vision once the string was complete. Behe ("Darwin's Black Box") is very good at debunking these claims.
The information theory guys (especially Hubert Yockey, in his "Information Theory & Molecular Biology") show that even if you allow for one atom sticking to another atom successively during each trial, there's not enough time in the universe (as calculated by Big Bang assumptions) to permit enough trials to form a single molecule of something essential for life: cytochrome c.
c. During each shake of the bag, it's just as likely that two beads on a string could come off as come one; so "productive evolutionary change" accomplished during one trial might very well be undone during the next. That's exacly the problem with pseudo-explanations like random point mutations: most of them are destructive, even in species that have high populations and rapid reproduction (like fruit flies) allowing us to see most of the genetic changes "in vivo" in real time.
Neo-Darwinism is probably useful as a way of explaining micro changes in an already established species. e.g., how does one variety of rose morph into another? How does one kind of dog become another? etc. This is useful, but it's the kind of modest, limited knowledge that was already known in a practical "rule of thumb" sort of way by professional livestock breeders in the 19th century. Darwin was actually immensely influenced by them in his own thinking. The difference is that he took the idea of selective breeding and carried it to absurd extremes and absurd conclusions. He excised a passage from the first edition of "Origin of Species" claiming that bears probably waded in shallow waters to catch fish, venturing out farther and farther as they "adapted" to a water environment, and therefore eventually morphed into whales! Most of his peers said "You've gotta be kidding, Chuck." Breeders knew very well that if you keep altering the basic stock, trying to tease out desired traits, one of two things happen: the new species is sterile (evolutionary dead-end); or it snaps back to the original stock you started out with.
The concept of a jump here is relative. Major changes can occur over "small" period of time in the context of a geological time scale of billions of years. Keep in mind "small" here can refer to thousands of generations of a population.
where are all the intermediate forms?
Everything is technically an "intermediate" form. A necklace with 2 beads is "intermediate" to one with 3 beads and so on. They are all necklaces in their own right.
b. Most important biochemical processes -- the blood-clotting cascade, the vision cascade, and some others -- cannot be reduced to a series of incremental steps.
These (and other examples) of irreducible complexity have been torn to shreds.:
c. During each shake of the bag, it's just as likely that two beads on a string could come off as come one; so "productive evolutionary change" accomplished during one trial might very well be undone during the next.
To make this analogy closer to the truth, you would have multiple strings shaking in multiple bags occuring simultaneously. The string that lose their beads "die" to be sure. The ones that gain beads not only proceed to the next trial, but "reproduce" to be greatly represented in subsequent trials.
he difference is that he took the idea of selective breeding and carried it to absurd extremes and absurd conclusions.
How do you explain the fact that humans, apes, mice, fungi, plants, paramecium, and dictyostelium share many of the same genes and have virtually identical codes (keeping in mind much of the codon designations are for the most part arbitrary) ?
I know, I know "godwanteditthatway".