Pollination 'principles' revisited: specialization, pollination syndromes, and the evolution of flowers

We call attention to the continuing value of the contributions of K. Faegri and L. van der Pijl in their classic book on pollination ecology (1966, 1971, 1979). We try to show that: 1) Some of the recent disagreement about the importance of specialization in the evolution of pollination systems is the result of confusing evolutionary specialization with ecological specialization. 2) Empirical and theoretical arguments about the importance of multiple pollinator species have sometimes ignored the observation that not all floral visitors are pollinators, only some of these pollinators are quantitatively important, and only a further subset generates selection. 3) Several lines of empirical evidence support the commonness of evolutionary specialization, although tight specialization in pollination ecology is extremely rare. 4) There are limits to specialization in pollination ecology and it is important to understand them. 5) If defined broadly, pollination syndromes have utility, and may reflect stable correlations among functionally related traits. The maintenance and stability of these traits may be explained by contemporary stabilizing selection, genetic/developmental constraints or homeostasis, or some combination of these processes.