SEARCH

SEARCH BY CITATION

Summary

The distribution of pollen upon the insect-visitors to three species of Viola is described and correlated with their feeding and cleaning behaviour. Visitors adopt one of two basic positions while feeding. In the prone position the insect alights on the anterior petal and probes for nectar so that its dorsal surfaces contact the floral genitalia. The supine position may be adopted after alighting on the anterior petal when the insect rotates so that it is virtually standing on its head. While probing for nectar from this position it is the ventral surfaces which contact the floral genitalia. Bumblebees adopt either feeding position; many large solitary bee species almost invariably adopt the supine one while large hoverflies always feed from the prone position. The behaviour of short-tongued visitors tends to be highly individual with respect to contact with the floral genitalia and many body surfaces may be involved.

The varying activities of insect-visitors while feeding result in the deposition of pollen at different locations on the integument. Consequently insect-borne pollen is presented to the stigma in various ways and there is a spectrum of pollination effects: medium and long-tongued nectar-seeking insects tend to be systematic cross-pollinators while short-tongued or pollengathering species tend to be chance cross- or self-pollinators.

The cleaning behaviour of insect-visitors strongly influences pollination as it removes much integumental pollen. Two requirements of the pollen deposition mechanism are postulated: that pollen should be placed (a) in a position not occupied by foreign pollen and (b) in a position irregularly or inefficiently cleaned. There were no positions which were free of foreign pollen throughout the visitor-species. On the other hand, violet pollen was found to be concentrated at locations inaccessible to the cleaning mechanism.

Systematic cross-pollination results from the visits of a variety of insects with proboscides over 5 mm long including bumblebees, honeybees, solitary bees, hoverflies and beeflies, therefore, the violet flower is not simply a ‘bee flower’ as previously supposed. The widespread practice of generalizing on the role of insect-visitors in pollination is questioned and the error in labelling flowers according to the taxonomic position of their pollinators is pointed out; the greater relevance of the morphological and behavioural characteristics of the insect-visitors is emphasized.

Differences in the floral genitalia of these Viola spp. may reflect a shift from allogamy to facultative autogamy which, in turn, may be a response to a paucity of pollinators at certain seasons. However, the diverse range of pollinators may maintain some out-crossing by the least efficient but more abundant short-tongued pollinators.

The diversity of pollinators, therefore, provides a mechanism whereby a sexual system can simultaneously produce variation and invariance. The evolutionary versatility of a mating system incorporating this mechanism may have been partially responsible for the success of the genus Viola throughout the temperate regions of the world.