Chapter Nine addresses the lack of transitional forms found in the geological record, another issue that creationists like to bring up today. Of course, things have changed significantly since then: we have an incredible amount of transitional fossils now (not that that, or indeed anything, will satisfy the creationists), but we still do not have, and probably will never have, transitional forms for everything. The reasons for that may be more subtle than Darwin supposes, but he has it essentially right.
One important point that he makes early on in this chapter: "I have found it difficult, when looking at any two species, to avoid picturing to myself, forms directly intermediate between them. But this is a wholly false view; we should always look for forms intermediate between each species and a common but unknown progenitor; and the progenitor will generally have differed in some respects from all its modified descendants." In other words, identifying transitional forms is confounded by the fact that they are more properly transitional between their ancestors and their descendents than between any two of their descendents, and so we may have a faulty search image for the latter. As an exaggerated example, to say that crocodiles and ducks had a common ancestor does not mean that we must find a "crocoduck" in the geological record. Rather, we will find something with some ducklike features and some crocodilelike features, but it will also have features found in neither, while others will be found in all three. (In the mathematical jargon that inexplicably imposes itself upon me, the two sets are not necessarily mutually exclusive, nor will their union be isomorphic to the common ancestor.)
An important point that Darwin makes in passing is that our classification system for living things should properly be based on their patterns of descent from common ancestors. In perhaps most cases this can be taken backwards as a rule of thumb: common ancestors can be inferred for members of most of our heirarchical groups that are exclusive of all other groups at the same level. That this is the case was and is one of Darwin's more compelling but also more subtle arguments. He has made this point before, and will make it again.
One of the larger issues that many had in Darwin's day was that the amount of time required for the diversification of life into its myriad extant forms is huge. This is something that Darwin freely admits, and of course modern science says no different. The age of the earth was something of a hot topic at the time of the publication of the Origin. He mentions "Sir Charles Lyell's grand work on the Principles of Geology, which the future historian will recognise as having produced a revolution in natural science…" and in our day (Darwin's future) these words have borne out. Charles Lyell had made a compelling argument for an ancient Earth just before Darwin set out on his voyage on the Beagle in the 1830s, but many still argued against it. Lyell's view was ultimately to persevere, just as Darwin's, but it was sufficiently controversial that Darwin needed to defend it himself in his work. Darwin's arguments centre on rates of sedimentation and erosion, and while we now have many other forms of corroborating evidence, his arguments are sound at least in principle (I am not geologist enough to say how far off he may be on the particulars). One noteworthy point is his conclusion of an age of 300 million years for the earliest "secondary" (which is to say, Mesozoic) strata, which is actually not too far off.
Having shown that great spans of time will have passed between the deposition of successive strata, Darwin moves on to the fact that those ages have not resulted in abundant fossils. It is somewhat comical how easily some of these chapters' points can be collapsed into one pithy sentence. In this chapter's case, it is simply "Fossilisation is uncommon." This, again, is still quite accepted today, but Darwin takes the principle further. Specifically, he feels the optimal conditions for fossilisation and for speciation are opposite to one another, so that those organisms that do get fossilised will most likely be from long periods of morphological stasis. In his words, "Nature may almost be said to have guarded against the frequent discovery of her transitional or linking forms." Darwin's argument betrays a bias toward land-based life: when sea level falls, new habitat is exposed, and extant land-based life-forms have all sorts of opportunities to take advantage of otherwise virgin territory in adaptive radiations, but of course sea-based life-forms must retreat to formerly deeper regions. Meanwhile, Darwin supposes that only the sedimentation occurring at the floors of bodies of water is capable of initiating the process of fossilisation, rather than (say) mudslides or river runoff. Again, I do not know enough geology to say whether this is in fact accurate.
Another essential observation is that stratigraphic range is almost always smaller than actual historical range. In other words, the earliest fossil found of a given organism is unlikely to be from its point of speciation, and the latest fossil found is unlikely to be the point of its extinction (or diversification into other forms). Fossils can only establish minimum expectations, and unless compelling evidence exists otherwise, we should always assume conservatively. Meanwhile, linked to this argument (found at the end of page 298 in the original, and reprised on page 301) is a surprising (although unemphasised) precursor to the theory of punctuated equilibria! Specifically: evolution is likely slow and gradual everywhere, but slower in some places than others, and should a more-quickly-evolving form reinvade and conquer territory held by their slower-evolving cousins, the appearance in the fossil record will be of one form suddenly giving way to another. Should the rapidly-evolving form have done its evolving in a climate noncondusive to fossilisation, this sudden transition may represent as much of a record as we may be able to acquire, but it does not indicate that the transitional forms existed only for a very brief moment of time, or not at all.
In summary, Darwin dismantles fairly comprehensively the arguments against his theory that his critics were likely to make (and, in the case of modern creationists, continue to make) based on the fossil record. It is possible that he is being merely rhetorical when he says that "I do not pretend that I should ever have suspected how poor a record of the mutations of life, the best preserved geological section presented" until he started to examine the problem in detail, but it is certainly true that science is full of surprises, and in the course of finding a simple and elegant explanation for a phenomenon one first finds that the data to be explained are far fuzzier, fuller of exceptions and borderline cases, than previously imagined. Evolutionary history is full of these sorts of things. I will let Darwin have the last word here: "I look at the natural geological record, as a history of the world imperfectly kept, and written in a changing dialect…."