This article was published in the CSA Journal, 4(4): 186-189 & 191, Jul-Aug 2004.

Sexual Self-incompatibility in the genus Cymbidium.

One of the major thrusts to be consider in biology is the concept of "Adapt or Die". As environments change, life forms need to be constantly changing in order to continue their lineage. It is as well to remember that all life forms that exist on earth today derive in a straight line from those first drops of life that formed on earth, perhaps 4 billion years ago. It is such an advantage to learn to appreciate the long and tortuous road that life has taken. It gives one the opportunity to regard the splendor of the world and its inhabitants with some of the awe deserved by something so wonderful. Many forms of life have existed at various times which lost the struggle to continue, and now leave nothing except ghosts. In order for life to continue, an inherant ability to vary according to conditions needs to have come about.

With simple single-cell organisms, which can multiply exponentially, mutation is usually enough. A mutation can be looked apon as mistake or misprint in the DNA message contained in the cell nucleus. Usually such a mistake is fatal, but occasionally, the changes which results are of benefit to the organism. In a population of billions of single-cell organisms, many thousands of mutations must occur, and it is possible that but a single individual could be created which could survive a particular disaster, and in a short period this single individual could multiply to give rise to a huge population of an updated form; ensuring the continuation of that that particular line of DNA.

Large multicellular organisms only really appeared when sexual fusion arose. As a mutation in a single cell of a multi-cellular organism could no longer be depended upon to allow an organism to survive a catastrophe, a new way of ensuring that variations would exist needed to develop. Through the process of sexual fusion, new individuals were produced, arising from half of the genetic material of each of two other individuals who had survived to the point of reproduction. To maximise the variation derived from this process, it is obvious that the donors of the two half-doses of DNA should be different individuals, and as a result, sexual differences came into existence. This is seen in most animals, and also some plants, for example the dioecious papaya and willow (male and female flowers on different plants). The majority of plants are, however, hermaphroditic, having both male and female organs present in the flower. And this is also the case with plants of the genus Cymbidium.

In some species of cymbidiums, aspects of their pollination biology operate against self pollination. Plants with low flower-counts, such as Cymbidium eburneum, are outbreeders by virtue of the fact that they have low flower counts. In these cases the pollinator collects the pollinarium whilst departing from a visited flower and the next flower visited is most likely to be on a different plant.

In the case of Cymbidium insigne, while it does have a multiflowered spike, it attracts its pollinators by deception. In Thailand, this species grows under bushes of Rhododendron lyi and is visited accidently by bumble-bees feeding on the nectar of the Rhododendron, the flowers of the Cymbidium mimicking the flowers of this bush. (See Dupuy & Cribb, 1988 The Genus Cymbidium, (Christopher Helm and Timber Press), p 24.) The bumble-bees, having been caught out, avoid the flowers of the Cymbidium, until at a later time, having forgotten the last fruitless experience, repeat the accident. As there is a waiting period governed by the memory of the bumble-bee, it is most likely that carried pollen will be deposited on a flower on a spike far removed from the original flower, and therefore on a plant of a different clone.

In a species like Cymbidium tracyanum, where large multispiking plants occur in nature, there is a higher chance of a pollinator visiting more than one flower of a single plant, and where pollination results, inbreeding would take place. It is not surprising therefore, that some species of cymbidiums exhibit the phenomenon known as sexual self-incompatibility, which itself is widespread in the plant kingdom.

I have found very little written on this subject with regard to the genus Cymbidium, so most of the following is based on personal observation.

In general, in the subgenus Cyperorchis (the parent species of "standard" cymbidiums), it seems that the green/brown species exhibit varying degrees of sexual self-incompatibility but the light-coloured species do not. The one species that seems least likely to produce viable seed following self-pollination is Cymbidium tracyanum. I have attempted selfings of this species on many occasions, and although a pod is frequently formed and this grows and persists for about 6 months or more, there is never any trace of seed present when the pod yellows and opens, even the placental tissue being undeveloped. Such empty pods, although appearing large in comparison to those of many other cymbidiums, are quite small when compared to a cross-pollinated Cym. tracyanum pod of the same age. The only successful selfing in this species of which I am aware, is that of Cym. tracyanum FCC/RHS by Andy Easton. Failure of selfing in this species in cultivation is not that important, as a number of different clones can be found, all likely to be cross-compatible. Self-incompatibility could be a useful tool to determine whether two plants of this species are separate clones, or actually represent two divisions of the same clone.

In an attempt to discover something about the nature and genetics of self-incompatibility in this species, I took eight flowering seedlings of a cross between two of my clones of Cym. tracyanum and selfed them and crossed each of them with the other in both directions (a total of 64 pollinations). Half of the flowers fell off within 4 weeks after pollination, and all but six fell off within seven months. The six survivors produced full pods; unfortunately there were too few pods produced to draw any conclusions. Only in one case were both pods of a reciprocal cross produced. I still know nothing of the nature of the self-incompatibility of this species.

What I had wanted to explore with this experiment was whether Cym. tracyanum was a plant exhibiting multiallelic, single-locus, gametophytic self-incompatibility. If you think that sounds difficult, getting you mind around the theory of this little area of botany is a lot harder, and should anyone be looking for a real intellectual challenge, I can guarantee that they will find one here. It is not something that sticks in the brain with any ease and everytime I wish to ponder over it, I need a quick refresher course. In gametophytic self-incompatibility, the nature of the pollen-surface proteins involved in the incompatibility reaction is determined by the DNA of the haploid nucleus of the pollen grain itself. In the other major form of self-incompatibility, sporophytic (occurring mainly in the cabbage and daisy families), these surface proteins are determined by the genes of the diploid nuclei of the mother plant. This apparently small difference, gives a significantly different nature to these two forms.

With the gametophytic self-incompatibility, one would expect to find that the progeny of the Cym. tracyanum 'McNeil' x Cym. tracyanum 'Cook's' that I used, when crossed with one another, would sort into four groups, the members of each being compatible with members of other groups, but incompatible with members of their own group; the so-called "four-group family".

Recent research shows that some forms of gametophytic self-incompatibility operate by causing an enzyme that affects RNA (an RNase) to enter the pollen tubes of incompatible pollen and thereby abort the growth of such pollen tubes. The complete story has yet to be worked out, but one can be assured that this system will show all the elegance that is the hallmark of all of nature's processes. A recently-published paper in this field, which I found on the Internet is "S-RNase complexes and pollen rejection" by Felipe Cruz-Garcia, C. Nathan Hancock and Bruce McClure, and can be found at: Although not an easy read, after a few readings one should get some idea of the nature and amazingness of self-incompatibility. There are many other relevant pages to be found on the Web.

When it comes to plants of the species Cymbidium i'ansonii (Cym. lowianum var. iansonii), we are confronted with the unfortunate situation that there appears to be only one single clone in cultivation. Selfing in this species is almost as unlikely to reward as it is in the case of Cym. tracyanum. I experienced a number of unsuccessful attempts, where pods persisted for about 12 months, but when opened, contained no apparent seed, although having some placental development. I finally secured a minute quantity of seed after using one of the published tricks for overcoming compatibility barriers, namely bud-pollination. Here, I forced open the terminal bud, perhaps one week before it was due to open, and used mature pollen from an open flower from near the base of the spike. From this seed, I raised about seven seedlings. When the first of these flowered, I put its pollen back onto the parent plant, but the pollinated flowers fell off in quick succession.

Other techniques described for attempting to overcome self-incompatibility include:

Cymbidium lowianum and its variety 'Concolor' are more generous than Cym. i'ansonii and although they do not produce very full pods on selfing, one can usually raise sufficient seedlings from a pod. It should be borne in mind that in cymbidiums, self-incompatibility is a relative and not absolute process. Only in cases where one only gets a full pod of seed following a selfing, can one deduce that no self-incompatibility exists. Cymbidiums, being capable of producing many hundreds of thousands of seed, require a more complicated definition of self-incompatibility that do most non-orchidaceous plants.

I consider it to be fortunate that my two favourite species, Cym. erythrostylum and Cym. sanderae can both be selfed successfully. In the former case, a full pod of good seed is formed, which however does have one unfortunate characteristic; it is rather short-lived when stored dry under what I assume to be ideal conditions. In Cym. sanderae, a plant also represented in cultivation by a single clone, the harvest tends to be on the low side, but one can usually expect to raise at least four or five flasks from a pod, It appears that some degree of self-incompatibility must exist in this species.

I have had little experience with species from the subgenus Jensoa (terrestrial Oriental cymbidiums), as I do not have the green thumb needed to grow them successfully, and I have had little success in raising them from seed in the few cases where I have tried. I have obtained full pods in each case, on selfing three species of this group.

The thick-leaved species of the subgenus Cymbidium are interesting in that I have found Cym. aloifolium to self easily and Cym. finlaysonianum to be impossible. It was interesting to discover this year that the hybrid between these two species, Cym. Hanalei would also not carry a pod following selfing. This is in contrast to hybrids between self-incompatible and self-compatible species of the subgenus Cyperorchis, where experience with Cym. Ceres, Cym. Eburneo-lowianum and Cym. Pauwelsii showed that these could all be selfed with ease, yielding large pods filled with seed, despite having one parent showing some signs of self-incompatibility.

I offer the following table of selfing results that I have gathered over the last 20-odd years of toothpick madness:

SpeciesNo of ClonesResults
finlaysonianum1 -
floribundum (pumilum)6+++
suavissimum1+++ / - (reported)
devonianum2+++ / - (reported)
tracyanum4 - / + (reported)
iridioides (giganteum)2+++ and -
hookerianum (grandiflorum)1+++
lowianum3 +
i'ansonii (low. var. iansonii)1+- and -
sanderae (parishii s.)1 +

Key: +++ full pod + some seed +- very little seed - no seed

Self-incompatibility is a condition under the control of genes and therefore can be passed on to hybrids. While the majority of hybrids in cultivation have large percentages of self-compatible species in their background, I think that it is likely that red cymbidiums may have some self-incompatibility genes in operation. While there are many reasons why a particular cross may fail or only produce a tiny amount of seed, the only explanation I can find for the failure of the classical "blue" cross Cym. Flavian 'Latour' x Cym. Fascination 'Amethyst', that I attempted quite a few years ago, is self-incompatibility. Both parents will make a reasonable amount of seed when crossed with other plants, but when crossed with each other, out of the scrapings from three separate pods, I was only able to raise one seedling, a plant that has, in addition, never flowered (Murphy's Law),

Interspecific unilateral incompatibility is another interesting, and probably related study. In this case, an interspecific hybrid can be made in one direction but fails in the other.
This is something that appears to take place when using Cymbidium devonianum and C. dayanum; the pollen of these species can be used successfully, whereas they fail to work as pod-parents. It has been shown that in Nicotiana, the RNases involved in self-incompatibility also cause interspecific unilateral incompatibility.

Back in 1899 (Journal of the RHS, 24: 87), it was reported that Veitch had found that Cymbidium lowianum, C. giganteum (iridioides) and C. tracyanum were pod-sterile and only useful as pollen parents. In interspecific unilateral incompatibility, it has often been found that one parent is self-incompatible and the other not, and that valid pollination occurs where the self-incompatible parent is the pollen-donor. This supports Veitch's findings for these three Cymbidium species.
There are obviously huge fields of study available here for interested students. I look forward to seeing their results.

Bibliograpy: Botany online by Peter v. Sengbusch Self-Incompatibility: How Plants Avoid Inbreeding

Lewis, D. 1979. Sexual Incompatibility in Plants. London: Edward Arnold,

Allard, R.W. 1960. Principles of Plant Breeding. New York: John Wiley & Sons

Richards, A.J. 1986. Plant Breeding Systems. London: George Allen & Unwin

© 2003-2011 Greig Russell