The results on proboscis length and flower-tube length indicate that the amount of force that a bee has to exert to reach the nectar at the flower base of S. sclarea largely depends on its proboscis length. Bees with a longer proboscis have to exert a much smaller force than bees with a short proboscis, because the latter have to press their heads deeper in the corolla tube to reach the nectar. The forces that a bee with a short proboscis (6 mm) needs to reach the nectar of S. sclarea (260 mN to 407 mN, for a short or a big flower, respectively) are much higher than the forces honeybees can exert (maximum measured force 29 mN), and even if a honeybee could exert forces comparable to the ones measured in the strongest bumblebee queen (90 mN), they would not be able to reach the nectar. This means that the length of the proboscis of a bee has to fit the flower-tube length of a visited Salvia species. Honeybees with their short proboscis (6.05 to 6.40 mm) are not able to reach the nectar in S. sclarea flowers and thus are excluded from the nectar by the length and narrowness of the flower tube. Therefore, flower-tube length seems to be an important factor in specialization of sages on a certain set of flower visitors because the forces required to reach the nectar drastically increase with decreasing proboscis length.
The important role of flower-tube length or spur length in specialization on flower visitors is known for many plants, e.g., ecotypes of Satyrium hallackii, the Disa draconis complex, and many other species [36-39]. Several authors have mentioned the possible role of flower-tube length for specialization in Salvia. Dieringer et al.  discuss that S. mexicana might be specialized on a single pollinator, the bee Deltophila elephas because the very long corolla tube of this sage species corresponds to the long tongue of this bee species . Westerkamp  writes that S. glutinosa is a bumblebee flower, and that honeybees are not able to reach the nectar of this sage because of their short proboscis. ClaBen-Bockhoff et al.  emphasize that for a definition of pollination niches in Salvia species, a variety of flower characters including length and width of the corolla tube are essential (see also [24,41]). This was recently confirmed by Kuschewitz  who distinguished 11 flower types in 39 European sages that differ in floral proportions and thus in the spatial patterns of pollen transfer.
Our results, based on the first quantitative measurements of the forces involved in flower visits by bees, support these hypotheses, which were based on morphological characteristics, and allow a more detailed discussion of the function of various flower parts in the genus Salvia.
Was this article helpful?