Hello Klaus:

I have looked for answers for the dispesral functionality of pollen exine ornamentation for many years, in fact for about 400 million years.  The spores of some Devonian plants have amazing ornamentations, including spinres and retrorse hooks that have been interpreted as dispersal aids (see paper Kevan, Chaloner & Savile in Palaeontology). There are oligolectic bees that specialize on collecting pollen from genera/families that produce spiny pollen (e.g. Cucurbit bees, Cactus bees, Sunflower bees, etc.).  I have watched various Bombus spp. collecting spiny pollen from Althea (Malvaceae), Convolvulus, Opuntia, Cucurbita and Cucumis, and many Asteraceae.  

The results of your experiments do not coincide with my observations in the field, so I suggest that something else is going on, perhaps more subtle.  It may be true that some species of Bombus are unable to compact the spiny pollen produced by some species of plants in their corbiculae.  Why might that be?   Perhaps the pollen form allows the bees to transport pollen between flowers (enhancing male plant function), but, for some species of bees, inhibits collection destined for larval food.  Collection for food takes the pollen out of sexual circulation.  

Dini Eisikowitch  (Tel Aviv University) was interested pollen packing by pollinating bees working at the flowers of plants with heteromorphic pollen (e.g. distylous and tristylous species).  John McNeil (Royal Botanic Gardens, Edinburgh) noted some Caryophyllaceae with strikingly dimorphic pollen, but I don't think he offered a functional explanation in dispersal.

The review by Roulston and Cane indicates that pollen form and content seems not so co-evolved with pollinators as are nectars and floral forms, but reflects the plants' needs in male reproductive success.  Palynological literature reviewing pollen form and ornamentation tends to emphasize, quite strongly, the importance of development and formation of the intine and exine.  There was a review on pollen forms from plants pollinated by hummingbirds, but, if I remember, the author was rather inconclusive about any patterns.

Your question is fascinating.  So, there are some ideas.  

With best wishes for 2016!

Peter

Hi Klaus, 

Yes indeed, it is a question I paid attention for many years without finding a satisfictionary answer. My first idea was that all these exine structures or ornaments are good for adhesion on the stigma. We had a look for fine structures of the stigma surfaces which may correspond with exine structures.  Another idea was that these spines evolved for pollen packing so that during collecting it, much more pollen can be transfered for pollination. The case of  Malvaceae we had a look for pollen collecting structures of Malva specialist bees like Tetraloniella malvae in comparison with these of Tetraloniella salicariae (specialist on Lythrum) or Tetraloniella dentata (specialist on Centaurea). But i could not find substantial differences in the hind legs of females. Ok, we have to look for further ideas.

Best wishes

Hannes Paulus

Dear Klaus:

I found very interesting the anterior answers.

I bring my little help. My hypothesis is the echinate pollen have the function to be selective with their pollinators. But I don't know if you have tested all their floral visitors  to catalog as pollinators.

Klaus:

Why not place insects foraging on plants with spiny pollen under the SEM?  Do the spines make it more likely that they will catch between the different branched hairs and bristles on an insect's body?  After all, the morphology of hairs on a bee's body changes in morphology and density according to the bee taxon and where grains are most likely deposited.  Here's another possibility.  Regard the spines as templates were the endothelium of the anther deposits the adhesive pollen oils (pollenkitt).  Are spines wearing globules of pigmented and fragrant oils more likely to adhere to bee bodies? The spiny pollen walls of a number of Asteraceae are very greasy with oil droplets that are propotionately large compared to the exile surface. Will larger globules contain more pigments and volatile odors making them more attractive to certain bees?  If you add spines you increase the surface area of the grain making it possible to add more oil globules between the spines.

Look at the literature from the 1970's - 190's on bird-pollinated flowers in Australia.  The grains are often triangular, trilobed and/or spiny and it's largely assumed they are micro adaptations permitting the pollen wall to become attached in between the father-barbules of passerine plumage on the head or crown.  That's what I found when I placed crown feathers of honeyeaters (Meliphagidae) under the SEM.  The birds were foraging for nectar on Amyema species (Loranthaceae), a genus with trilled grains.

The late Bruce Knox was interested in static electricity of pollen but never published.  Are spiny grains more highly charged then smooth grains?  Once again, this may help them adhere to hairy or feather surfaces.

Very good ideas and answers about the spiny pollen question! Thank you very much. However, I know already a little more than I told you.

1. Spiny pollen grains adhere very well to the bristles of the bees' body; following a few visits to Althea flowers bees are completely covered with pollen grains. Another effect may be due to the colour and size of the pollen grains, which are visible very good.

2. Bumblebees and honeybees fail to store the pollen into the corbiculae. One idea was that the spines produce a kind of lotus effect preventing the admixing of nectar which honeybees and bumblebees use to store the pollen in the pollen baskets. Another observation is that the pollenkitt obviously does not cover the spines, which means that pollen grains do not adhere together be ymeans of their tied pollenkitt.

3. In a flight cage we have tested bumblebee workers, that previosly collected pollen fronm flowers of a non-target plants, at flowers with spiny pollen: Most of them failed completely to store pollen grains in the pollen baskets. They used thus only two options which were switching to nectar collection or stop visiting the flowers at all. Noteworthy, some bumblebees were able to collect (small amounts of) cucurbit pollen, whereas all failed to collect malvaceous pollen.

4. The oligolectic specialists like Tetralonia nana seem to possess long bristles on the scopa of the hindlegs and collect pollen grains without compacting them like corbiculate bees do.

Our conclusion is that spiny pollen grains train the bees to restrict to nectar collection and help the flowers to save more pollen grains for pollination.

I have enjoyed reading everyones thoughts here concerning pollen exine ornamentation, it has been enlightening and thought provoking.

Dr. Lunau, Is the work you did with bumblebees and cucurbit/malvaceous pollen written up?  I would be extremely interested in reading more about this work.

Dear Jacob,

some results have been published and are available in Research Gate:

Lunau K, Piorek V, Krohn O, Pacini E (2015) Just spines  ̶  Mechanical defence of malvaceous pollen against collection by corbiculate bees. Apidologie. 46: 144-149.

Klaus

Dear Klaus,

Thank you for the citation, I look forward to reading this paper!

Best,

Jacob

Dear Klaus and Peter,

Interesting topic!

Peter and I collected some bumblebees and other bees from a species of commercial Althea this year in Haba Snow Mountain, NW Yunnan. I need to check our specimen and their pollen load slides to see whether bumblebees carrying large pollen loads of Althea on their baskets or not.

Dear Zong-Xin,

interesting idea to check pollen collection on photographs. I just checked my photos and found a nectar-drinking bumblebee with a good pollen load in the pollen baskets (see attachment)  on an Althea flower. However, in this case the pollen in the pollen basket is obviously not from the flower the bumblebee visits due to its deviant colour.

Klaus

I am enjoying this fascinating discussion!  Klaus has asked an interesting question and we have discussed a number of possible answers.  

The triangular/trilobate pollens of Australian bird visited flowers may be a reflection of phylogenetic constraint, but even that still begs the question.  

The paper we published some years ago on Peponapis pruinosa presents an SEM of the spiny pollen of Cucurbita pepo trapped in the long scopal setae.  

There are other SEMs of pollen trapped on the proboscides of various Lepidoptera, and, as Peter B. points out, from bird feathers. The triangular, rather dry pollen of oil palm (Elaeis guineenis) adhers well to the bodies of the pollinating weevils (Elaeidobius) (SEMs have been published). 

The idea that the pollenkitt may act as a sort of water-proofing against nectar/honey being used by the bees to clump spiny pollen (vs. other pollens) is interesting.

The emerging view, as expressed by Klaus and in a different context by Rouston & Cane, indicates that the male functionality of pollen in plant fertility and fecundity is probably as, or more important, than bee-collectability.   

As far as I remember the false pollen produced by various flowers, presumably as a nutritional resources (cryptic dioecy, Melastoma, Cassia etc.), is not spiny.  

With best wishes to all!  Peter 

Dear Peter and all discussants,

this was a good last word for this discussion, but probably not the final word of the general discussion about the pollen dilemma of bee-pollinated plants. I expect evidence that flowers do not only (try to) exclude bees as nectar thieves and illegitimate visitors, but also that flowers are selective among bees and exclude corbiculate bees that are not only collecting pollen, but also are recruiting very effectively other pollen collecting worker bees and at the same time are unable to pollinate flowers by the pollen grains stored in the pollen baskets in contrast to bees which do not use nectar to clump pollen grains for transport.

Klaus

Klaus:

How spiny must a grain bee to discourage bees with corbiculae.  Think of the sunflower (Helianthus) industry.  Honeybees are the dominant pollinators in this commercial industry,  I've never seen a pollen grain in the Asteraceae that was not spiny (regardless of pollinator).  What about all the bee taxa that specialize on the Asteraceae (genus Svastra?).  How do they cope with the spines?  Is it only because they have scopae instead of corbiculae.

You say that the spines prevent the grains from adhering to each other with their pollenkitt in the corbiculus.  Perhaps but don't all eusocial apids bind the grains with regurgitated nectar?  Isn't that why the pollen in the cobirculus looks like a smooth, shiny tablet or pill?  My hosts here in Sydney have a colony of trigonids.  We watch them return to the nest with little smooth pills of pollen on their corbiculi.  The "pill" breaks apart quickly and the grains separate on a glass slide when you add Calberla's solution as that solution contains the solvents of water and ethanol.

I was with Ren when he took that photo of the bees visiting Alcea in Shangri-la.  Both bumblebees and anthophorids (probably Anthophora or Amegilla) were actively collecting pollen.

Dear Klaus,

sorry I can not add more information or discuss the topic, but  I agree with you and Peter.  However  I suggest a colleague Dra. Maria Cristina Gaglianone she studied bees that collect pollen on several Malvaceae species in Brazil. Perhaps she has more ideas about your question.

  Warm regards and wish you and family a nice new year

Marlies

Dear Peter and Marlies,

you are right with your comments: we have tested pollen collection of bumblebees for some spiny pollen of Malvaceae, Asteraceae, Cucurbitaceae, and some control plants. Our results suggest that presence or absence of spines is not the only factor, but spine length and pollen grains diameter play a role. And obviously some individuals learn better than others to handle the spiny pollen.

It is interesting to read that pollination in some crop plants with spiny pollen seems problematic: Vaissière, B.E., Vinson, B. (1994) Pollen morphology and its effect on pollen collection by honey bees, Apis mellifera L. (Hymenoptera: Apidae), with special reference to upland cotton, Gossypium hirsutum L. (Malvaceae). Grana 33, 128–138

... and that specialist oligolectic bees seem to benefit from loose scopal bristles pollen collection: Gaglianone, M.C. (2000) Behavior on flowers, structures associated to pollen transport and nesting biology of Perditomorpha brunerii and Cephalurgus anomalus (Hymenoptera: Colletidae, Andrenidae). Rev. Biol. Trop. 48, 89–99; and Schlindwein, C., Pick, R.A., Martins, C.F. (2009) Evaluation of oligolecty in the Brazilian bee Ptilothrix plumata (Hymenoptera, Apidae, Emphorini). Apidologie 40, 106–116

As always also the world of pollen collection is not black and white but smudgy and grey. Attached is a series of photos demonstrating a bumblebee's fail to collect spiny pollen grains of Alcea rosea.

with many thanks for your contributions and the best wishes for a happy new year

Klaus

Some of the discussion confuses pollen dusting on pollinators' bodies (and so available for male function of the plants) vs. pollen loading into scopae (and so removed from male functionality).  

Dear Peter, you are right with this important comment. Most pollen that has been groomed into the pollen transport structures, scopae or corbiculae, might be lost for pollination, at least if regurgitated nactar is added as in honeybees, bumblebees, stingless bess and many other bees. A notable exception might megachilid bees and other bees with ventral scopa which have different pollen storing techniques which are grooming the pollen with the legs into the ventral scopa or taking up pollen directly from the anthers with the ventral scopa; the latter could easily result in pollen transfer to a stigma.

Hi Klaus,

A very interesting topic. I’m glad you’re looking at the subject in detail.  The adaptive significance of pollen grain characteristics merits far greater investigation.  Unacceptability of Malvaceae pollen to Bombus is certainly quite striking.  We found hollyhock, Alcea rosea, pollen adhered well but was groomed off and rejected by bumblebees (‘Bumblebees’ Prŷs-Jones and Corbet 1987; 2011).  Some of these bumblebees arrived carrying pollen loads, but still groomed off Alcea pollen. Along with spininess, absolute size of Alcea pollen grains could also be a factor making collection difficult for Bombus.  Spines may certainly be important, along with a variety of other potential factors including differential tolerability to contained alkaloids - as Jane Stout has found for Rhododendron nectar. A multivariate analysis of foraging preferences and pollen characteristics would be interesting.

Oliver

I want to add the static electric aspect to this interesting discussion. A flying bee collects electric charge, which concentrates on its body's hairs. The oily pollen grains on (many plant species) anthers attract to the charged bee hairs and almost jump and adhere to them. While the adhered grains contact a wet stigma they discharge through the stigma-style-ground and adhere to the wet surface by hydrogen bonds. For grooming the pollen grains from the body hairs and building pollen load the bee has to discharge the grains, which may be achieved by wetting them. A spiny pollen grain surface may act as an electric charge collector and enhances the electric effect. Below see SEM photos of the spiny avocado pollen grains: a. Loose adhered grains on an anther. b. Electrically strong grains' adhering to honeybee hairs. c. Pollen grains on a wet stigma. d. Wetting grains organizing into pollen load.

Dear Gad,

very intering information. I have some questions: Is it important that the pollen gains are oily for being attracted to the charged bee hairs as suggested by your comment? Is there any experimental data showing that grooming pollen into the pollen baskets need electric discharge? How does wetting discharge the pollen grains; because there is a concucting bridge to the flower or because the regurgitated nectar is loaded differently? Why are the pollen grains in your photo "Electrically strong grains' adhering to honeybee hairs" so deformed?

Sorry I forgot to mention that the photo in my previous answer shows bristles of the scopa of Tetralonia nana with adhered malveceous pollen grain glued together by means of the pollenkitt.

Dear Klaus, your wise questions are significant.

For being activated electrically pollen grains should not be electrically conductive. Thus, if a bee hairs are negatively charged the grain surface facing towards the hairs charged positively and attracted to the hairs. Live cells are usually wet and conductive. They achieve electrical isolation by means of oily layer (like in nerve cells). When an electrically adhered pollen grain is moisturized it loses its charge towards the bee hairs. I assume the deformation of the pollen grains adhering to the bee hairs in my photo is due to the electric attraction, induced by the high hairs' charge. However, the pollen grains attached to the anther valve (see my previous photo) are not deformed, and may be adhered to the valve (and to each other) by weak oily bonds. This is interesting that the old world honeybee meets with difficulties while building an avocado pollen load. It loses much pollen, uses lot of nectar and ends with a small pollen load (5-7 mg, see the photo). The new world stingless bees succeed much better with avocado pollen, and our honeybee easily builds 15 mg load of the local Sinapis alba pollen. See the attached photos of a honeybee's Sinapis pollen load, which is well organized, with minimum aid of nectar.

Spiny may somehow related to the enzymes that required for recognition of precise pistil to proceed the germination of pollen tubes, an idea 

Dear Caser, most pollen grain species are not spiny, and still activate the stigma-pollen recognition mechanism. I assume the specific spiny forms of some pollen grains have a function in pollen adhere mechanism to the insect heirs (maybe using electric charge), and maybe also in pollen load arrangement.

I am pretty sure that the spines are involved in pollen load arragement. In our Apidologie paper from 2015 we explain that bumblebees were unable to collect spiny pollen grains of Alcea rosea due to malfunction of pollen load arrangement. However, if the pollenkitt has been removed or if the spines have been bent by vortexing, the bumblebees were able to store the pollen in the pollen baskets. In all experiments we offered hand-collected pollen in pipette tips combined to artificial flowers that were buzzed by the bumble bees.

Lunau K, Piorek V, Krohn O, Pacini E (2015) Just spines ̶ Mechanical defence of malvaceous pollen against collection by corbiculate bees. Apidologie. 46: 144-149.

Spines are species -specific pattern of the sculpturing on the pollen ektexine. These spines will increase the capturing potential of these pollen grains against the receptive portion of the stigma. Since these spines are very small compared to the baskets, I think, it is a matter of compatibility facts rather than the defensive hypothesis.

Thank you

spines facilitate pollen trans location and sustain landing on stigma besides improving the communications between pollen and stigma  

I also found bumble bees not succeeding to collect Alcea pollen in their corbiculae (see picture). However, I've seen beetles eating lot of Alcea pollen (picture). Also see a honeybee starting to build its pollen load (picture) and succeeding well (picture).

Dear Gad,

thank you for pictures and comments. Did you observe the honeybee collecting pollen grains? I ask because the pollen grains of the flower look yellow, whereas the pollen in the corbicula looks orange: Are you sure that the collected pollen is from the flower on the picture?

Besides this question, we know from not yet published results that some bumblebee individuals were able to collect Cucurbita pepo pollen, but others not. Our impression was that using a higher amount of regurgitated nectar enabled the bumblebees to collect the spiny pollen grains; but we did not yet confirm this impression by experimental tests.

Klaus

The Hibiscus pollen looks yellow and turn orange when wet with the bee theregurgitated nectar. Moreover, honeybees rarly collect and mix different pollen species into the corbiculae. The avocado pollen is also spiny. Honeybees hardly succeed to put it in the corbiculae and use lot of regurgitated nectar for that. However, I found 9 stingless bee species in Mexico who easly perform it using dry avocado pollen.

Klaus Lunau , we just published a paper on this topic, check out Lynn et al. 2020, Sexual and Natural Selection on Pollen Morphology in Taraxacum. The work was heavily inspired by your papers on Malvaceae pollen.

Dear Austin,

nice to see more implications of pollen spines due to your study; I downloaded your paper already february 15. We are still struggling to unravel the role of pollenkitt in the system of echinaceous pollen grains, since we have found meanwhile spiny pollen collected by honeybees and bumblebees. It seem that these bees might also increase collectability by adding more nectar to the pollen during packing.

Klaus

Avocado also has spiny pollen grains. While on the anther's valve the grains are loosely attached to the valve and to each other (pic. 1).

However, when moving to the honeybee's hairs the pollen grains are strongly attached to each other and to the hair, apparently by electrostatic forces (pic. 2).

Honeybees meet with difficulties building avocado pollen loads. They use lot of nectar, losing large amount of pollen to the ground, and usually collecting not more than 5 mg loads (pic. 3).

This may be the reason why honeybees mostly avoid avocado pollen collecting, and stay with the nectar. Nevertheless, they act as an effective pollinator, due to the avocado pollen carried on their body.

A reasonable function of the pollen spines may be to enable efficient adhering to the bee body, then preventing building the pollen loads, and by this keeping a larger pollen amounts available for pollination.

I can add an anecdotal observation. We tried to force honey bees to collect pollen exclusive from single plants, in a tunnel setup. To my surprise, after 3 weeks, honey bees had not collected any pollen from Papaver rhoeas. Assuming the bees were the problem we exchanged two colony collecting pollen, between Papaver rhoeas in one tunnel and Reseda odorata in another. The new colony stopped collecting pollen also in the Papaver tunnel, whereas the old colony went on to collect pollen from Reseda. In the field, honey bees readily collect the pollen from Papaver rhoeas. Now I wonder if the honey bees need nectar from another plant to help bind the pollen grains of Papaver rhoeas, that does have nectar. Are the spines of Papaver rhoeas the problem?

Dear Per, I don't know the answer to your question, but only have some suggestions. In my garden bumblebees, but not honeybees collect papver rhoeas pollen. Bumblebees have the capability to buzz the stamens, whereas honey cannot buzz. This could explain the inability of honeybees to effectively collect poppy pollen.

Another thought about Per's interesting observation: Honeybees, for providing all their nutrition requirements, need more than one pollen source (as well as for the nectar). May be after a while, under a single pollen source provision they would stop collecting it.

The honeybees' main nutrition problem in the big monoculture fields is that single food source.

However, Per said honeybees did not begin to collect poppy pollen.

Right, I haven't notice it. So the need for nectar for binding the pollen grains could be an explanation.

yes, but also honeybees leave the hive with nectar in the honey stomach and improve this behaviour when only poppy pollen is available. I guess that the inability to buzz the stamens is responsible.

Inability to buzz sounds reasonable. Interestingly, while honeybees visit Lycopersicum esculentum (Solanaceae), which carries buzz flowers with no nectar that are pollinated efficiently by Bumble bees, they succeed to obtain some pollen (not that efficiently) and to pack it into the corbicula.

There are some other examples like this.

Even in red poppy flowers most bumblebees buzz, but few are just running between the many stamens and get some pollen which might have been released by the buzzing ones.

Buzzing demands expensive energy. Bumble bees may save it, if possible.