The study of 101 phylogenies reveals a new interpretation of speciation. Nature Vol 463 21 Jan 10 p349
The hypothesis that speciation follows the accumulation of many small events that multiply or simply add together (gradualism as Dawkins promotes) is supported in only 8% and 0% of cases.
78% of phylogenies fit the simple model where new species emerge from single rare stochastic events that produce reproductive isolation sufficient to cause speciation.
Species simply wait for the next sufficient cause of speciation to occur. Speciation is freed from the gradual tug of natural selection. There need not be an arms race between species. Gradual genetic and other changes may often be consequent to the event that promotes the reproductive isolation and hence speciation, rather than causal themselves. Factors that can cause speciation include polyploidy, altered sex determination mechanisms, chromosomal rearrangements, accumulation of genetic incompatibilities, sensory drive, hybridization and physical isolation.
It seems that the clever prose of Richard Dawkins demonstrating how easy it is to gradually scale Mt Improbable are simply not supported by the data.