Laboratory experiments designed to elucidate the mechanisms of sympatric and parapatric speciation may have been handicapped by too small population sizes, although this possibility has seldom been discussed. In this paper we review the published records of sympatric and parapatric speciation experiments to test the relative importance of selection intensity applied, duration of experiment and effective population size. Our results show that among these factors only effective population size has had a general effect on the generation of assortative mating. Reduced interbreeding is less likely to develop in small populations where the selection process often seems to have been opposed by inbreeding depression or loss of genetic variation. This study demonstrates that the experimental evidence frequently used as an argument against sympatric and parapatric speciation models is not as strong as previously believed.