Esteemed Prof. Lunau:

Just a few ideas to fully expose my ignorance: I don't think "mimicry" is a good term for the situation you comment because -as far as I know- the concept of mimicry necessarily involves model/s. Or, if you prefer, an orchid with an empty spur should be considered as mimetic if similarly shaped nectariferous plants can be detected around. In South America, we have many Chloraea spp that display two paralell channels below the column (incorrectly called "nectariferous" since no nectar is secreted). A student of mine has recently demonstrated that the pollinating bumble-bees do in fact probe the cavities with their proboscices looking for nectar.

Sanguinetti, A., & Singer, R. B. (2014). Invasive bees promote high reproductive success in Andean orchids. Biological Conservation, 175, 10-20.

No obvious "models" were detected around. Well, all for now, I hope this is of some use. All the best.

http://www.jstor.org/stable/2096949?seq=1#page_scan_tab_contents

http://kbd.kew.org/kbd/detailedresult.do?id=74645

http://link.springer.com/article/10.1007/BF00349193

http://www.jstor.org/stable/2259728?seq=1#page_scan_tab_contents

Rodrigo:

Mimicry is a broad topic in orchids and does NOT always refer to Batesian or Muellerian model mimics. In orchids, we also recognize guild mimics and non-model (generalist) food mimics.  These "non-model mimics" attract  polylectic and/or polyphagic insect pollinators by exploiting the insect's optimal foraging patterns. This mode of mimicry often produces a flower larger and often smellier than other co-blooming species that offer rewards.  Ethologists once called this the "super normal stimulus.  If that's not mimicry, what is it?  Please see our chapters on Thelymitra and Cypripedium in our most recent book, "Darwin's Orchids; Then and Now" (U. of Chicago Press; 2014) or look up our recent papers on Cypripedium montanum, Thelymitra macrophylla and T. crinita. 

And by the way, Rodrigo, are you really, really, really certain that your Chloraea species fail to mimic anything else in bloom in the habit?  Have you compared color pattern and scent of all of those Chloraea species to other Andean, angiosperm species?  Have you checked pollen loads on your bees visiting Chloraea.  What  other pollen morphotypes are they carrying in their scope or corbiculae?

So yes, Klaus (esteemed or otherwise) an empty spur, councils or pouch may be seen as one of several character traits in mimicry by plants that make flowers without edible rewards.  An orchid does NOT have to have an empty spur or pouch to be either a Batesian or generalist food mimic.  Cypripedium and Thelymitra species offer no edible rewards and their flowers do not produce empty spurs for insects proboscides. Pollen load analyses tell you an awful lot about the foraging preferences of an insect BEFORE it flew to the orchid.

Dear Klaus,

I do not know of any papers that show bees are attracted to flowers because they see a floral spur.  I imagine it might be tested by cutting off spurs, if that does not affect other aspects of pollinator attraction.  Or perhaps create experimental flowers using a 3-d printer as was recently reported by Policha et al. (2016.  New Phytologist doi:10.1111/nph.13855).

Jim

Dear Jim,

I also prefer the interpretation that empty spurs represent relicts of previously nectariferous ancestors. What would convince me from the opposite is the evolution (or elongation) of an empty nectar spur in a deceptive species, or alternatively a behavioural tests as proposed by you.

Klaus

Dear Klaus,

For years that is what I had assumed (deception derived from reward flowers), but only in the last decade or so have there been phylogenies where we can map trait evolution.  Thus far, data seem to indicate the opposite: a more general trend for nectar producers to be derived, at least at the genus level.  Such results were somewhat a surprise to me, but the number of genera that have variation in nectar availability (deception vs reward) is not great, and the number of taxa where this has been tested is quite small (e.g., Orchis, Disa).  

There are probably other studies out there, but my memory is getting a little sketchy here since I have not worked on this problem for a number of years.

Jim

James:

And who says that a morphological character must function as a form of visible deception in a case of mimesis?  Obviously, it provides a place where the long-tongued insect can park its tongue long enough to complete the pollinium transfer.  Without a place to stick its proboscis the insect would leave before it contacted the rostellum  and/or the pollen receptive lobes of the stigma.  As spurs (empty or with nectar) are usually a little longer than an insect's tongue real contact between the insect and the orchid's column is a matter of violently ramming its head into the flower's sinus while forcing its tongue down a spur.  

Once again, please read the chapters in "Darwin's Orchids; Then and Now" by Dafni, Johnson and others.  There's no evidence, for example, that any South African Disa species or European Orchis ever had functional nectar glands.  Darwin didn't get that one right, I'm afraid.  The consensus of authors in this book emphasizes the evolution of mimetic systems in orchids by asymmetric evolution, not via coevolution or mutualistic mechanisms.  Take a look at the Chapter by Weston.  If Australasian, diurid orchids have lost and then re-evolved nectar glands across their lineage why can't spurs evolve as components in mimesis more than once?  You can't generalize and insist that the presence of an empty spur on a flower must always mean that it once had an ancestor that secreted nectar.  

Dear Peter,

I did not say that the spur functions as a visible deception; I was offering a means of testing it because that is what Klaus was asking: whether it was part of the deception (other than a place to put a proboscis).  From what I remember, phylogenies show general trends with possible reversals.  And yes, evolution of deception, whether it is mimetic or not, would be largely asymmetrical.

Jim

Wow! This topic is becoming a really fruitful discussion! Esteemed Prof. Bernardht, in the specific case of Chloraea virescens, its native pollinator (Bombus dahlbomii) is almost extinct and pollination is mostly performed by introduced honeybees and Bumble-bees (B. terrestris). In either case, bees probe the cavities looking for nectar. My student (Agustin) filmed this behaviour and the short film can be seen as a supplementary material at journal's webpage. The fact that introduced and native bees behave in the same way makes me think that something "innate" (my apologies if this word is too subjective) in the bees is taking place. I still prefer using "mimicry" in the context of pre-existing models, but this is a question of preferences, I guess. And, no, we didn't detect any putative models. The plants flowering nearby where very different (Apiaceae and Proteaceae). All the best!

Dear Klaus

I agree with you - "I also prefer the interpretation that empty spurs represent relicts of previously nectariferous ancestors. "

and another illustration of nectarless evolutionary strategy - although not exactly "the spur" but related in a way

http://onlinelibrary.wiley.com/doi/10.1111/j.1442-1984.1997.tb00150.x/abstract

http://onlinelibrary.wiley.com/doi/10.1046/j.1442-1984.2001.00048.x/abstract?userIsAuthenticated=false&deniedAccessCustomisedMessage=

Thanks to all disputants,

the tongue-parking argument was really helpful and provided another straight-forward hypothesis why deceptive orchids should evolve a (empty) spur. Probing for nectar might represent the manipulation that makes a pollinators out of a flower-visitor.

Klaus

Klaus,

Merci de «ce sujet à propos de l’éperon, qui’ a fait une discussion» ...

je m’interesse depuis longtement de la microstructure de l’éperon das deux cas: quand est-il vide ou nectarifaire ?

J’ai étudie micromorphologie d’éperon d’Angraecum sesquipedalé et d'autre espèces des tropiques comme Acampe, Cleisostoma, Trichoglottis (Telepova à la présentation au congrès de la Flore du Cambodge, Laos et Vietnam) , etc.

Ces éperons étaient poiluà l’intérieur et donc, ce n'est pas possible que le proboscite de papillon rentre à l’interieur d'éperon.

Par contre, s’il y a le nectar dans l'éperon des insectes comme coléoptères vont faire deux trous … c’est plus facile

Using molecular phylogeny to map the evolution of nectar glands in orchid taxa is much more common than you think, James.  Once again, I must refer you to the Weston and Clements chapter in "Darwin's Orchid; Then and Now."  The diurids are a primarily deceptive lineage, or so we thought, but look at the genera we now know have at least one species with nectar glands. The glands can appear on some of the oddest organs.  I've seen the functional nectaries of spider caladenias secrete  sweet fluid on the margins (not the base) of the labellum.  Kingsley Dixon has a living collection and I've tasted the secretions.  I've just finished 3 months of research on Corunastylis in which nectar is secreted from the tips of the upper lobes on each column wing.  If it's so easy for a lineage to evolve a secretory gland (which must be linked to a phloem strand) what is so difficult about evolving a spur, pouch or empty tube from a pre-existing petal or a portion of the perianth where petals fuse during floral development (the probable case in Disa)?  If it forces the insect to spend a little extra time on the flower so pollinaria are transferred it's both mimetic AND selectively advantageous.

Oh, Rodrigo, if you are going to call me esteemed try spelling my last name correctly.  I mimetic systems are asymmetrically evolved systems why are you surprised that Chloraea now fools naturalized Bombus and Apis species?  Your orchid din't evolve mechanisms that fooled specific Bombus species.  It evolved mechanisms that fooled large, foraging bees with long proboscides with bodies between 13 - 20 mms.  If I were you I'd try to find a grant (National Geographic?) that lets you work on a broad range of Chloraea species, over the SAME season, such as the ones that bloomed along the roadside at Mar de La Plata as described decades ago by Maevia Correa.  Do such populations still exist in Argentina?

And of course, the response of the alien Bombus to the native Chloraea is innate.  It's responding to the range of colors and odors it can perceive.  What's the big deal?  A hardy, Himalayan orchid (Bletilla hyacinthoides; name change) so in bulb catalogs  blooms in my Missouri garden every spring and is pollinated by the queens of at least two native Bombus species.  The flowers are lovely purple and mauve and lack nectar.  As P.T. Barnum once said, "There's a sucker born every minute."  And that's especially true in the case of anthophilous insects.  

Prof. Bernhardt (well spelled now?)

My sincere apologies for: 1) having ideas of my own and 2) not having adopted "Darwin's Orchid's...." as some sort of Holly Bible. You know, old-fashioned people like me think that scientific knowledge deals with ideas/questions to be tested/verified; preferably through unbiased points of views and rigorous methods, avoiding dogmas. I dare to say Darwin himself should have supported such idea, but you know, he lived at the IXX century...

By the way, Nicolas Vereecken, one of the authors of your book seemingly also supports the idea that "mimicry needs models", see the article below. All the best!

https://www.researchgate.net/publication/236210908_Cheaters_and_liars_chemical_mimicry_at_its_best_Can_J_Zool

Article Cheaters and liars: chemical mimicry at its finestThe presen...

Rodrigo:

Our book is not a bible.  Chapters are recommended to catch you up with more modern treatments and interpretations of hypotheses.  They represent exhaustive reviews on orchid mimicry and I suspect you have missed a number of them.  Nick Vereecken  writes all sorts of things so why not read his comprehensive review in our book? Nick is also a protege of Dr Dafni, one of the great pioneers on orchid flower mimicry, and Amots also has a chapter. If you agree with them now reading another two chapters won't kill you. 

AND I never said that mimics DON"T need models.  However, there are all sorts of model systems.  Some are very specific (Batesian) while others are far more diffuse.  Blue Thelymitra flowers do not mimic blue Orthrosanthus (Iridaceae) exclusively.  They mimic a wide range of co-blooming taxa with blue-purple, usually nectar-less (dry) flowers with porose-poricidal others.  Likewise, Dipodium is a big purple-spotted flower that does not resemble much of anything in the summer flora of Easter Australia.  It's a generalist mimic of purple-nectar secreting flowers (so attractive to larger Australian Hymenoptera) and rather indicative of the super-normal stimulus that exploits optimal foraging behavior.  The only other explanation is to invoke a false history in which this orchid's model became extinct but the mimic species somehow survived.  

One of the big questions is how do sex mimics evolve?  Amots Dafni and I looked at that in 1989 and suggested they are derived from food mimics.  They represent a more specialized state canalizing fewer but more efficient pollinators as they are hard-wired to go to pseudo-phromone flowers again and again.  Food foragers learn by their mistakes and they learn fast.  Remember, male wasps, bees and flies feed on nectar as well as their females so the directional selection from food to sex flowers can be very fast since receptive female insects employ scent molecules also found in flowers.  Surprisingly, Dr Weston and Clements concurred with us following an exhaustive molecular study on almost ever diurid genus in Australia.  Amos and I had only to wait 25 years for vindication.  Unfortunately, the Weston/Clements  study is in the same book you prefer to ignore on what,... quasi-athiestic reasons?

You are a naughty, naughty boy, Rodrigo.  It's time to return to the pinned specimens you caught on Chloraea and have a look at the pollen they carry of other species.  Scrape pollen off the hind legs and/or give the specimen a bath with ethyl acetate on a glass slide.  When the solute evaporates stain the pollen residue with Calberla's fluid.  You can't have a pollination lab without Calberla's fluid.  I strongly suspect you are dealing with a non-model mimic similar to the Dipodium and Cypripedium species I've worked on in the past.

PS Your spelling has improved so why not also consider expanding on broader reading and incorporating additional lab and field techniques?

PPS Don't worry Klaus - Fieldwork in Australia always brings out the mischief in me.