A transitional fossil is one that looks like it's from an organism intermediate between two lineages, meaning it has some characteristics of lineage A, some characteristics of lineage B, and probably some characteristics part way between the two. Transitional fossils can occur between groups of any taxonomic level, such as between species, between orders, etc. Ideally, the transitional fossil should be found stratigraphically between the first occurrence of the ancestral lineage and the first occurrence of the descendent lineage, but evolution also predicts the occurrence of some fossils with transitional morphology that occur after both lineages. There's nothing in the theory of evolution which says an intermediate form (or any organism, for that matter) can have only one line of descendents, or that the intermediate form itself has to go extinct when a line of descendents evolves.
To say there are no transitional fossils is simply false. Paleontology has progressed a bit since Origin of Species was published, uncovering thousands of transitional fossils, by both the temporally restrictive and the less restrictive definitions. The fossil record is still spotty and always will be; erosion and the rarity of conditions favorable to fossilization make that inevitable. Also, transitions may occur in a small population, in a small area, and/or in a relatively short amount of time; when any of these conditions hold, the chances of finding the transitional fossils goes down. Still, there are still many instances where excellent sequences of transitional fossils exist. Some notable examples are the transitions from reptile to mammal, from land animal to early whale, and from early ape to human. For many more examples, see the transitional fossils FAQ in the talk.origins archive, and see http://www.geo.ucalgary.ca/~macrae/talk_origins.html for sample images for some invertebrate groups.
The misconception about the lack of transitional fossils is perpetuated in part by a common way of thinking about categories. When people think about a category like "dog" or "ant," they often subconsciously believe that there is a well-defined boundary around the category, or that there is some eternal ideal form (for philosophers, the Platonic idea) which defines the category. This kind of thinking leads people to declare that Archaeopteryx is "100% bird," when it is clearly a mix of bird and reptile features (with more reptile than bird features, in fact). In truth, categories are man-made and artificial. Nature is not constrained to follow them, and it doesn't.
Some Creationists claim that the hypothesis of punctuated equilibrium was proposed (by Eldredge and Gould) to explain gaps in the fossil record. Actually, it was proposed to explain the relative rarity of transitional forms, not their total absence, and to explain why speciation appears to happen relatively quickly in some cases, gradually in others, and not at all during some periods for some species. In no way does it deny that transitional sequences exist. In fact, both Gould and Eldredge are outspoken opponents of Creationism.
"But paleontologists have discovered several superb examples of intermediary forms and sequences, more than enough to convince any fair-minded skeptic about the reality of life's physical genealogy." - Stephen Jay Gould, Natural History, May 1994