[a different tim]

The nylon bug is gradualism, I think. I'd assume it's one mutation away from whatever it used to metabolise. It would count as a gradual change. The thing is, that when you get to the molecular level everything goes in short leaps because you can't get less than one base change in a DNA mutation. Gradualism argues that these small leaps are what add up to large evolutionary changes. A catastrophist or saltationist view would argue that some other mechanism is at work that can produce large changes all at once.

Hox genes kind of confuse the issue because a small molecular change (gradual) can produce a large change in form (saltation). Other small molecular changes might produce much smaller changes in form. From a DNA point of view there is no difference between these two but they would look very different in a line of animals. Bacteria confuse the issue as well because they can actually swap useful genes between species so one species might suddenly aquire useful genes from another.

Gradualism, I think, works OK for the cane toad legs but what the prof seems to be saying is that the sort of change that leads to speciation can happen in one leap. The cane toads haven't speciated and the longer legs may be part of normal variation. The standard position is that speciation is what happens when normal variation gets so great that parts of the species can't interbreed, and - I think- the prof is saying different.

I have to admit it's a bit hard to tell from the precis. The title of the paper suggests that he's arguing that the molecular clocks we use to construct animal "family trees" don't work in the way we think they do, but the precis is about something entirely different.