In reward-producing animal-pollinated plants, small populations are li
kely to be less attractive to pollinators than large populations. The
likelihood of pollinator limitation of seed production should therefor
e increase with decreasing population size. I documented the number of
flowering plants and plant fecundity in 18 populations of the self-in
compatible, tristylous herb Lythrum salicaria within an archipelago in
northern Sweden in two consecutive years. To test the hypothesis that
a positive correlation between population size and seed set is due to
a higher degree of pollinator limitation in small than in large popul
ations, I performed supplemental hand-pollinations in eight (1993) and
13 (1994) of the study populations. To test the hypothesis that commo
n mating types are more likely than rare types to experience inadequat
e pollination, I compared the natural level of seed production and the
effect of supplemental pollination in different style morphs in the f
ive populations in which greater than or equal to 10 plants per morph
were included in the experiment. There was no significant correlation
between population size and plant size in terms of number of floral sh
oots or number of flower-producing leaf nodes per shoot. However, ther
e was a positive relationship between population size and seed product
ion per flower and between population size and total seed number per p
lant. In contrast, there was no significant correlation between popula
tion size and seed production of flowers that had received supplementa
l pollination. In both years, the difference in mean seed production p
er flower between hand-pollinated flowers and controls decreased with
increasing population size. In two of five populations, the effect of
supplemental pollination differed significantly among morphs. Seed pro
duction was more likely to be pollinator limited in long-styled than i
n short-styled plants, but this difference could not be attributed to
a preponderance of the long-styled morph in the studied populations. R
esults of the supplemental hand-pollinations indicate that the positiv
e correlation between population size and seed production is a functio
n of insufficient pollen transfer in small populations. Additional dem
ographic studies are needed to determine to what extent the reduced le
vel of seed production in small populations limits the growth of young
populations, and to what extent it may threaten the local persistence
of L. salicaria.