Land snails of the genusPartulaFérussac offer several advantages for the study of micro‐evolutionary change. They are highly variable both within and between populations. Their mobility is low, so that there are genetic differences between natural populations only short distances apart (20 m or less). They are ovoviviparous, and easy to rear in the laboratory, so that these genetic differences can be investigated experimentally. Finally, they show, at least in some places, a very curious pattern of speciation.Since 1962 we have been studying the population genetics ofPartulaspecies on the island of Moorea in French Polynesia.The late Professor H. E. Crampton, in a classic monograph, recorded ten species from the island. Later, he added an eleventh. One of the species (P.dendroicaCrampton) is an allopatric replacement ofP. suturalisPfeiffer and probably deserves only the rank of geographical race, since the two forms cross freely in the laboratory. We have evidence thatP. tohiveanaCrampton,P. olympiaCrampton, andP. mooreanaHartman will all, in some localities, hybridize withP. suturalis.We also have evidence of natural hydridization betweenP. aurantiaCrampton andP. suturalis, and betweenP. exiguaCrampton andP. taeniataMörch.There are thus two major‐species groups on Moorea (thesuturaliscomplex and thetaeniatacomplex). The status of the other ‘species’ (P.mirabilisCrampton, P.solitariaCrampton and P.diaphanaCrampton and Cooke) remains uncertain, although they are probably members of thetaeniatacomplex.Two members of a species‐group may behave as distinct species at one locality, but hybridize freely or intergrade at another. Such changes can take place over distances of 200 m or less. Some of the changes seem to be related to geographical barriers, There is, for instance, an apparent ring‐species with a diameter of about 5 km. Other changes are more difficult to interpret, since they occur without obvious relation to geographical features.Ecological and genetic studies on P.taeniatahave led to the suggestion that striking divergence between adjacent populations can take place even in the absence of geographical barriers. Whether this can continue to the point of speciation is still uncertain, but an explanation in these terms would clarify many of the puzzling phenomena we have observed in populati