Why was Darwin's view of species rejected by twentieth century biologists?

Historians and philosophers of science agree that Darwin had an understanding of species which led to a workable theory of their origins. To Darwin species did not differ essentially from 'varieties' within species, but were distinguishable in that they had developed gaps in formerly continuous morphological variation. Similar ideas can be defended today after updating them with modern population genetics. Why then, in the 1930s and 1940s, did Dobzhansky, Mayr and others argue that Darwin failed to understand species and speciation? Mayr and Dobzhansky argued that reproductively isolated species were more distinct and 'real' than Darwin had proposed. Believing species to be inherently cohesive, Mayr inferred that speciation normally required geographic isolation, an argument that he believed, incorrectly, Darwin had failed to appreciate. Also, before the sociobiology revolution of the 1960s and 1970s, biologists often argued that traits beneficial to whole populations would spread. Reproductive isolation was thus seen as an adaptive trait to prevent disintegration of species. Finally, molecular genetic markers did not exist, and so a presumed biological function of species, reproductive isolation, seemed to delimit cryptic species better than character-based criteria like Darwin's. Today, abundant genetic markers are available and widely used to delimit species, for example using assignment tests: genetics has replaced a Darwinian reliance on morphology for detecting gaps between species. In the 150th anniversary of The Origin of Species, we appear to be returning to more Darwinian views on species, and to a fuller appreciation of what Darwin meant.