The genetics of homostyly in populations of Primula vulgaris

Abstract

Natural populations of primroses containing homostyles were found by Crosby (1940) near Sparkford, Somerset, and in the Chilterns. Homostyle frequencies up to 80% have often been recorded, but few populations containing higher frequencies have been found. In explaining observed configurations of homostyle frequencies Crosby (1949) assumed that they always selffertilized. Bodmer (1958), however, obtained evidence for the occurrence of crossing between the homostyles under natural conditions. The estimated amount of crossing increases with lower viability for the homostyles and the data indicate a frequency of crossing of at least 80 %. Analysis of counts of populations containing homostyles in Somerset made by Professor Sir Ronald Fisher, F.R.S., over the years 1941-54 shows heterogeneity within populations, but no evidence of an upward trend in homostyle frequency. The only significant change detected is a downward trend in the homostyle frequency. There is also a marked deficiency of thrums relative to pins for high homostyle frequencies. Investigation of theoretical models which take into account crossing between homostyles shows that the homostyles will increase in frequency so long as there are no viability disturbances. However, a method for calculating an effective initial selective value for the homostyle gene shows that when the frequency of crossing is high only moderate viability disturbances are needed to prevent the homostyles from increasing in frequency. Numerical investigation of the models indicates that only a slight decrease in viability is needed to decrease the proportion of homostyles in populations where they are already abundant. Moreover, if the decreased viability affects particularly the mature plants it is possible to account for the thrum deficiency and also a possible very rapid decrease in homostyle frequency. With high initial selective values for the homostyle gene, such as are obtained when the homostyles only self-fertilize, even when they have considerably reduced viability, only a few occurrences are needed to make them almost certain to gain a hold and increase in frequency. It does not seem possible, therefore, to explain why the homostyles have occurred only in a few restricted areas, unless the effective initial selection for them is very small or negative, which can only occur when there is a considerable amount of crossing between the homostyles. It thus seems that the tendency of the homostyles, and possibly also of pins and thrums, to cross-fertilize is an essential buffering mechanism against the possible reversal of an outcrossing system to inbreeding.