There has been some work by Dobson et al showing the importance of pollenkit volatiles in pollen detection. See

Dobson, H. E. M., 1994. Floral volatiles in insect biology. In: E. A. Bernays (ed.), Insect-Plant Interactions. CRC Press, Boca Raton, pp. 47-81. 

Dobson, H. E. M. 1988. Survey of pollen and pollenkitt lipids - chemical cues to flower visitors? American Journal of Botany 75: 170-182.

Dobson, H. E. M. and G. Bergström. 2000. Ecology and evolution of pollen odors. Pages 63-89 in A. Dafni, M. Hesse and E. Pacini, editors. Pollen and Pollination. Springer, Vienna.

In the chemical attraction of orchid flowers  the more important place presents Anter cap"s stomata and non pollens:

TELEPOVA-TEXIER M. 2011. Interaction entre les orchidées et leurs pollinisateurs. Bulletin trimestrielle Les Amis du Museum National d’Histoire Naturelle, 248: 61-66.

Bergström, G., H. E. M. Dobson and I. Groth (1995). "Spatial fragrance patterns within the flowers of Ranunculus acris (Ranunculaceae)." Plant Systematics and Evolution 195(3-4): 221-242.

Cook, S. M., J. C. Sandoz, A. P. Martin, D. A. Murray, G. M. Poppy and I. H. Williams (2005). "Could pollen odours play a role in the foraging behaviour of the honey bee (Apis mellifera)." Physiological Entomology 30: 164-174.

Dobson, H. E. M. (1987). "Role of flower and pollen aromas in host-plant recognition by solitary bees." Oecologia 72: 618-623.

Dobson, H. E. M., E. M. Danielson and I. D. van Wesep (1999). "Pollen odor chemicals as modulators of bumble bee foraging on Rosa rugosa Thunb. (Rosaceae)." Plant Species Biology 14(2): 153-166.

Dobson, H. E. M., I. Groth and G. Bergström (1996). "Pollen advertisement: Chemical contrasts between whole-flower and pollen odors." American Journal of Botany 83(7): 877-885.

All this is about scent, i.e. volatile compounds - is that what you are looking for, Klaus?

I instantaneously thought of (water-soluble!) compounds to be detected by gustation rather than olfaction, i.e. on a smaller spatial scale, as I would expect the scent of the whole flower to override the (as I would expect) rather delicate scent of pollen...

...these water-soluble cues might be anthocyanins (rendered water-soluble by esterified sugars) that at the same time color the pollen and protect the DNA from UV-damage?

Orchids pollinia are partially hidden  in the anther and pollenkitt is absent. Viscidia, instead, are exposed and could emitt some attractive molecules. Pollenkitt analysis was performed in only few species to generalize the few results.

Dear Klaus:

As you know now, the answer is a big yes based on the publications provided by Drs. Wackers and Inouye.  I saw the work of Dobson while she was a student of Herbert Baker at Berkeley in the early 1980's.  The solitary bees she raised recognized the smell of the pollen they ate as larvae based on the scent she removed from the pollenkitt and placed on filter paper.  It was very convincing.  Here is a recent paper from the Bernhardt/Meier lab that may be of interest as we analyzed the scents.  The previous paper discusses the mimicry of Thelymitra flowers that deceive pollen-collecting female bees.

Peter 

Dear debaters,

thanks for input in form of ideas and literature. I would like to define my question more precisely. I am looking for a general chemical cue of pollen by which pollen eating insects can identify pollen. Preliminary results have shown that Eristalis hoverflies can taste the free amino acid proline in the pollenkitt.  From honeybees it is known that can smell proline, but the detection threshold is rather high and olfactory detection of proline in pollen this very unlikely. So I am looking for other candidate substances that are common in pollen. Since I know from resins, that it possesses not only viscous and non-volatile compounds but also volatile monoterpenes and sesquiterpenes in order to avoid hardening I am not focused on gustatory but also on olfactory pollen detection via components of the pollenkitt.

An alternative might be cracking the pollen grains and tasting chemicals from the interior of the pollen grains.

Klaus

Dear Klaus:

Please go to the following link and read the paper...  http://aob.oxfordjournals.org/content/92/3/445.full

2-phenylethanol may be considered for the following reasons.

1) It is found in a syrphid fly-pollinated flower offering pollen as its only reward.

2) The same molecule evolved independently in the scents of many angiosperms including the most basal (ANITA group and eumagnoliids).

3)  It is found in an orchid (Thelymitra-see paper sent yesterday) that mimics other flowers offering pollen as their only reward.  

Review the literature in the Trimenia paper and find the flowers studied.  Remove pollen from the flowers and see if 2-phenylethanol is present.  Also, the rare molecule 8-heptadecene may also be considered.

Peter

I would advise to read the book "Pollen and Pollination" by Amots Dafni et al. (2000) Springer, New York

May also be interesting to contrast the chemistry between pollen from insect-pollinated and wind-pollinated plants. Furthermore, the fact that different pollen-feeding arthropods exploit and/or prefer different pollen types, indicates that pollen identification cues likely differ between arthropod species  

I like Felix idea, will add to compare species exclusively pollinated by wind or animal and also some ambophilous species. I think one can find the three pollination systems within monophyletic clades... Thank you for the discussion topic, great comments about a very interesting subject.

hell kalus launu

You know The most widely recognized plant-pollinating insect is the bee. Some species of bees, like the honeybee, pollinate flowers. Every day a single honeybee might visit up to several thousand flowers, carrying pollen between them. They are social bees that nest in hives. They do most of their pollination within 100 yards of their hive. Other bee species pollinate other plants. The squash bee specializes in squashes such as pumpkins. Squash bees are not social insects and interact little, though they will nest together underground.

Read more : http://www.ehow.com/list_7392403_insects-pollen-one-plant-another.html

Dear Klaus:

Dr Wackers is correct.  Dr. H. Dobson did the basic study on how bees identify pollen grains based on the odor in the lipid droplets on the pollen wall.  What is very interesting in her study is that bees fed on the pollen of one species as larvae "remember" the odor as adults and go to the odor preferentially.  I saw this when I visited Berkeley in the 1980's and Heidi Dobson was s student of Herbet and Irene Baker.  Heidi raised native, solitary bees (colletids?).  She could remove the odor from the pollen wall and put it on filter paper.  The bees would visit the filter paper.

Peter  

Dear Klaus,

probably adult bees recognize edible pollen grains according a sequence of different types of physical and chemical messages, simlilar to pollen stigma recognition.  Pollenkitt is typical of primary animal pollinated species having monad pollen, nevertheless some secundary anemophilous species, as Olea europaea (olive), have pollenkitt but are not visites by  domestic bees.

Dear Peter,

you are right: Heidi Dobson tested the feeding preferences of CoIletes fulgidus longiplumosus. And she said that inexperienced bees show a preference for the pollen on which they were reared as larvae; but all bees were reared on pollen of their sole foodplant; ther is thus no evidence for conditioning to pollen odors experienced as larvae. This is far from my initial question, because here specific components of pollen of a distinct species play a role.

However, in another paper Heidi Dobson found: The addition of 2-tridecanone and geranyl acetate reduced landing responses, but the addition of tetradecyl acetate and especially eugenol significantly increased landing frequency as well as the display of vibratile pollen-collecting behavior, suggesting that they are key chemicals in a bee’s assessment of pollen availability in individual flowers.This finding is very close to what I am asking, namely evidence for a widespread chemical component of pollen that triggres behavioural responses.

Klaus

Dear Ettore,  as you might know I have tested the collection of pollen surrogates with bumblebees in the laboratory. Bumblebees collect pure (chemically totally inert) glass powder, deposit it in the nest and come again to collect more (see attachment). The larvae even eat the glass powder, which then has been already mixed with honey.

Klaus