Homo habilis - there is a growing consensus amongst most paleoanthropologists that this category actually includes bits and pieces of various other types - such as Australopithecus and Homo erectus. It is therefore an 'invalid taxon'. That is, it never existed as such.
When you think about, this particular case is actually rather good evidence for evolution.
There is by no means yet a consensus that
H. habilis should be thrown out, though it is a viewpoint increasing in popularity. Does this mean, as the OP implies, that habilines do not belong in the human evolutionary story. Not at all, there is still a consensus that some of these fossils are either on, or very close to, our ancestral line.
What's happened is that our fossil collections have grown. When you have very few fossils to work with; it's relatively easy to divide them typologically. This is an australopithecine, this is
H. habilis; this is
H. erectus. As we uncover and study more and more fossils, however, we fill in the gaps and our picture becomes more fine-grained. This has the obvious result of blurring the lines between taxonomic units. We realise that early
H. erectus specimens have considerable overlap with habilines; just as habilines have considerable overlap with australopithecines. Some researchers start to notice that there are no features diagnostic of Homo habilis[/i]; and thus question whether it's a meaningful taxon.
This is an issue for taxonomists, but not for human evolution. Doing away with the clear lines of distinction between fossil hominins is a sign that we're filling in the missing links and completing the picture. It's not a sign that there's something wrong with the story.