|| Checking for direct PDF access through Ovid
We review published records of laboratory experiments on peripatric and vicariance allopatric speciation to address the following three questions: (1) What was the true effect size of reproductive isolation? (2) Was the reproductive isolation persistent? (3) What influenced the development of isolation? Contrary to popular belief, laboratory evidence for allopatric speciation is quite weak. Assortative mating was only found among derived populations in vicariance experiments. Reproductive isolation against control populations was only intermittent, so there is reason to doubt if some cases showing significant reproductive isolation really should be attributed to speciation. The method of testing was at least as important as the speciation model. Experimental populations tested against each other were the most likely to demonstrate reproductive isolation. This study suggests that allopatric speciation experiments are more likely to yield conclusive results under divergent selection than under drift, and points to the benefits of large populations and many generations.