Publication from research funded in part by a grant from the Australian Flora Foundation
Oldfield, A.C.†, Moncur, M.*, Leach, D.N.#, Wyllie, S.G.# & Tidemann, C.R. †
(† Department of Forestry, School of Resource and Environmental Management, Australian National University, Canberra, ACT 0200; * CSIRO Division of Forestry, Yarralumla, Canberra, ACT 2600; # Department of Science, University of Western Sydney, Hawkesbury, NSW 2753).
Proc. 6th International Theriological Congress. UNSW 1993
Many flowers produce strong odours which are thought to attract a variety of pollinator types (Armstrong 1979; Williams 1983). Flying-foxes have only recently been shown to be able to discriminate between plant odours (Oldfield 1993). Studies on bat pollination have previously concentrated on the bat-flower interaction from a descriptive ecological perspective (Howell 1977) or from analysis of nectar sugar composition (Baker and Baker 1983). Analysis of floral odours during pollinator foraging has previously been largely ignored.
This initial study used headspace-GC-MS techniques to detect changes in the composition of odours emanating from flowers of E.gummifera over a single 24-hour cycle. Differences in patterns of visitation by the Flying-fox, Pteropus poliocephalus and other, non-mammal, pollinators including bees and honeyeaters were then investigated in relation to the compositional changes in the floral odour “fingerprints”.
Analysis of 13 volatiles comprising 96% of the total floral headspace revealed a dominance of monoterpenes, particularly a- and ß-pinene, both peaked at 0346 and 1504 hours. Bat activity was not correlated with either nectar volume, nectar sugar concentration, nor ambient temperature, and only approached significance with floral volatiles concentrations over the time period during which the bats were active (r = 0.781, p>0.05). Only bee activity showed significant correlation with floral volatiles concentrations (r = 0.999, p<0.001). Both nectar and non-nectar producing trees were visited equally by all pollinator types. More importantly, flowers from visited trees had significantly lower concentrations of volatiles than control trees (X2 = 4.88, p = 0.027), however it was not clear whether this was the result of pollinator activity or that the trees were switching off volatiles production in order to attract pollinators. In visited trees, flowers producing nectar had lower volatiles concentrations than non-nectar flowers. In addition, those floral developmental stages associated with nectar flow and pollen shed were significantly lower in volatiles than other stages (X2 = 24.107, p <0.0001), the opposite of what might have been expected. This raises the possibility that nectar is not an important source of attractive volatiles for nectivorous bats and other proposed pollinators. Further study in this regard is warranted.
Armstrong, J.A. (1979), New Zealand J. Bot. U: 467.
Baker, H.G. and Baker, I. (1983), p.117 in ‘Handbook of Experimental Pollination Biology’, C.E. Jones and R.J. Little (eds), Van Nostrand, Rheinhold, New York.
Howell, D.J. (1977), Nature 270: 509.
Oldfield, A.C. (1993). pp.45-51 in: Update ’93: Proceedings of the National Low-Chill Stonefruit Conference, J. Slack (ed), NSW Department of Agriculture, Ballina.
Williams, N.H. (1983), p.50 in ‘Handbook of Experimental Pollination Biology’, C.E. Jones and R.J. Little (eds), Van Nostrand, Rheinhold, New York.