Meyers Conservatory - troymeyers.com
Site Navigation
Flasking Home Page
Your Picks List
Comprehensive List
Articles in the Site
Search
Ready-To-Go Flasks
Fast-Turn Flasks
Greenhouse Plants
Current Reservations
Testimonials
 
If you would like to direct someone to this web page, please copy and paste this URL into your email:
http://troymeyers.com/about?Coryanthes-Pollination

 
 
Pollination of Orchids in the Genus Coryanthes
 
 

The pollination method discussed here was described to me by Dr. Gernot Bergold of Venezuela, an orchid expert for whom two species of Coryanthes have been named.

This article deals specifically with the mechanical details involved in pollinating
Coryanthes, which have a bizarre flower structure, but the principles are the same for other orchids. All orchids have a column, ovary, pollinia, and stigma--those reproductive structures are what make a flower an orchid.
   -Troy

Pollination of Orchids in the Genus Coryanthes
Text by Troy Meyers, Photos by Dan and Marla Nikirk


 Coryanthes Flower Structure

The flower of the Coryanthes has the usual parts of the orchid flower, but they are unusually shaped and highly specialized. Please refer to the three photographs below for a better understanding.

The ovary is the stem-like connection between the flower and the flower spike.

The sepals, which initially serve as the outer layer of the unopened flower bud, are thin and membrane-like. The lateral sepals resemble "wings" when the flower is freshly opened and fully extended, but with a little age they may curl or roll up. The dorsal sepal forms a "cape" draped over and toward the end of the column.

The column itself isn't too oddly formed, but it fits tightly into the complex lip almost as if it were connected along its whole length, which it is not. The column carries both the male anther and the female stigma, and is the focus of the pollination attempt.

The relatively small lateral petals have the same thin substance as the sepals. They are arranged more or less parallel to the column and sometimes drape over the sides of the column.

The third petal, which has evolved into the lip or labellum in orchids, is especially complex in the genus Coryanthes. The lip is attached near the apex of the ovary by a slender connecting structure called a "claw." Beyond the claw, the lip is divided into three primary structures called the hypochile, mesochile, and epichile.

The hypochile is usually cap-like and usually the scent of the flower comes from osmophores underneath the cap.

The mesochile descends from under the cap toward the bucket-like epichile. The mesochile sometimes has shelf-like or scale-like structures protruding from it, known as lamellae. The pictured Coryanthes mastersiana does not have lamellae.

The epichile is the last segment of the lip, and it is bucket-shaped and fits tightly against the column at its terminus.

A pair of water glands known as pleurids are located usually on the column near where the claw connects to the ovary. These drip water into the bucket of the epichile.

Profile and Flower Parts
Profile of Flower with Major Parts Indicated

Dorsal View and Flower Parts
Dorsal View of Flower with Major Parts Indicated

Dissected Flower
Dissected Flower with Major Parts Indicated

 


Coryanthes Pollination in Nature

In the native habitat, most Coryanthes are pollinated by euglossine bees of the genus Eulaema or Euglossa. The male bees (sometimes of only one species) are attracted to the flower by their fragrance, which is emitted by osmophores under the cap of the hypochile. They attempt to collect this fragrance (an aid to attracting mates) but the waxy mesochile is difficult to get a firm foothold on, so a bee might slip and fall into the bucket-shaped, waterfilled epichile. The bee's wings get wet when it drops into the fluid filled bucket, which makes flying out of the bucket impossible, and because the walls of the bucket are too slippery, climbing up the steep bucket walls is not an option.

The only possible exit is for the bee to crawl up the escape ramp and out the small space between the end of the epichile and the tip of the column. This escape route is visible on the preceding photograph of the dissected flower. The following photo shows that the space allowed for the bee to escape is a very tight fit, since the right bee for the job is one that barely fits through the hole. The bee can be caught in the outlet for quite awhile before it finally works its way through. The bee's struggle to free itself loosens the anther cap which frees the pollen so it can be stuck to the back of the bee.

The bee is so attracted to the odor that it never learns its lesson, so it continues to search for the source of the fragrance on the same flower or another blossom. Again it falls into the liquid filled bucket of the Coryanthes lip, but this time, as it struggles to leave through the tiny opening, its back is forced against the column and the pollen is deposited into the sticky stigma thereby fertilizing the flower. The stigma of the flower usually does not open or produce sticky fluid until after the anther cap is removed, a sequence which discourages self-pollination.

Escape Route
Escape Route between Column and Epichile

 


Coryanthes Pollination Technique for Humans

The method described to me by Gernot Bergold retains some of the features of the chain of events that occur with natural pollination.

On the day the flower opens, or the next day, remove the pollinia as follows. This is done by removing the anther cap and then grasping the stipe ("stem") of the pollinarium with forceps, or by simulating the passage of a bee with a toothpick or something similar, to which the pollinarium may adhere. Sometimes the anther cap may come off with the pollinia embedded in it, and you should tease the pollinia out.

Store the pollen at room temperature, and not in a sealed container unless it is large. This will allow it to dry somewhat, making it firmer and easier to manipulate the next day.

Anther Cap Removed
End of Column with Anther Cap Removed
Note that sometimes with human pollination, the pollinia will come off in the anther cap.

Pollen Removal
Removal of Pollinarium with a Toothpick

On the day following removal of the pollen, it should be inserted into the stigmatic cavity. Assuming that everything goes well, removing the pollen and anther cap the day before should have caused the stigmatic slit to open.

If the stigmatic slit isn't open, you can wait another day, but this is risky because the flower may simply become too old to be pollinated successfully. Consider dissecting a slit into the column where it should be (best), or progressively slicing the end of the column off until the chamber is just revealed. If this is necessary, use a very sharp, sterile razor blade or scalpel. (See the notes below about the bucket being in the way for all this.)

In the case of the flower being pollinated for this series of photographs, the stigmatic slit appeared to be open enough to accept the pollen even on the day the pollen was removed, but by waiting a day, the flower has time to produce the hormones needed to encourage growth of the pollen tubes.

Stigmatic Slit
End of Column, as view from below, showing location of Stigmatic Slit
The Epichile is being held away from the Column to allow access to the Slit for pollination.

The pollination should be attempted around noon (within a couple of hours), and in sun if possible. Dip the pollinia in the bucket-water of the epichile. Accessing the stigmatic slit is difficult because it is on the underside of the column and against the end of the epichile. Have a helper hold the epichile bucket to one side, or away from the column so that you can access the stigmatic slit.

If you don't have a helper, and can't manage to hold the bucket aside and handle the pollen at the same time, you can remove the entire labellum by cutting through the claw with a sterile blade.

Work the pollen into the stigmatic slit, trying to get it to move in and toward the direction of the ovary. Try to put both pollinia all the way into the slit. In the photograph below you can see the the viscidium (sticky foot) and stipe (pollen stem) hanging out a bit... that is fine.

Dipped Pollen
Somewhat dried Pollinia, Dipped

Pollinia in the Stigmatic Slit
End of Column after insertion of Dipped Pollen

Expect a thick stigmatic fluid to fill the stigmatic chamber in response to the pollination in the next few days. This might tend to push the pollinia out... if some remains inside the chamber, don't disturb anything, but if all comes out, try to work it back in. Usually the slit will swell shut, sealing in the pollen, which will germinate and grow upward inside the tube of the column and ovary, fertilizing the ovules. These tubes can be seen in the dissection photo above.

Stigmatic Slit Closed
Swelling Column and closed Stigmatic Slit

The flower sepals, petals, and labellum will wilt in the next few days, but if the pollination was successful, the column and ovary will remain healthy looking and will stay attached to the stem of the inflorescence. The ovary will swell.

Capsule at 45 days
Developing seed capsule at 45 days after pollination

The typical amount of time from pollination to having a ripe capsule (ovary) is 65 days, varying a bit with different species and cultural conditions. The ovary will get surprisingly large, and when ripe will begin splitting where the capsule joins the column. Harvest the capsule as soon as the smallest split occurs.

Capsule Splitting at 59 days
Capsule Splitting at 59 days

 
 

Go to Flasking Home Page -- Contact Us
Contents of all pages Copyright © 1999-2024. All rights reserved.