A simple bleaching model

back / zurück

Abgeschickt von Bill Steele am 20 September, 2003 um 07:06:14

Orchid propagators sowing mature seed have long used chlorine bleach, sodium- or calcium-hypochlorite, to surface sterilize the seed to prevent contamination of cultures. Ben Lindén in 1980 and Allan Anderson in 1990 independently discovered that seeds of temperate terrestrial orchids left in bleaching solutions for time periods in excess of the minimum required for surface sterilization often germinated at higher percentage than those bleached only long enough for decontamination. The percentage germination increased with increasing bleaching time up to a point and then decreased with further increase in bleaching time.

Since this initial discovery, many of us working with mature seed of Cypripedium plants have used extended bleaching as a means of promoting germination of this seed. Indeed in my view, the degree of bleaching of mature seed is one of the most important cultural parameters for obtaining successful germination. The detailed nature of the mechanism by which bleaching enhances germination remains unknown, and the action of the bleach has been variously hypothesized to result from: 1) physical destruction of the carapace covering the embryo, 2) dissolution of germination inhibiting compounds within the embryo, 3) oxidation of such compounds, or 4) chemical action resulting from the high pH of the bleach. From an empirical standpoint, bleaching time for optimal germination is found to vary from one Cypripedium species to another, from one population of a species to another, from one clone to another within a population, and even from year to year for seed resulting from selfing of a given clone.

While the chemical mechanism by which bleaching promotes germination is unknown, my own thoughts about the bleaching process are guided by a simple model that I first proposed in 1996. In this model, I suppose that the population of bleaching times necessary to cause the seeds from a given seed capsule to germinate has a certain frequency distribution. Because many biological properties such as the length or the weight of an organism follow a Gaussian or normal distribution, I have assumed that the distribution of bleaching times necessary for germination is also normal. Although the actual distribution may not be normal, the mechanism by which bleaching overcomes germination, or the essence of the resistance to bleaching, is likely to be the result of an aggregate of individual random variables, and in such case the Central Limit Theorem of probability theory requires that the resulting distribution approach normality. The figure below is a graphical representation of the distribution of seed bleaching times required for germinating the seeds from a capsule. The graph can alternately be thought of as representing the number of seeds from a capsule germinating after bleaching for time t1, the percentage of seeds in the capsule germinating after time t1, or the probability that a randomly selected seed will germinate after bleaching for time t1.

Following similar reasoning, the distribution of the bleaching times sufficient to kill the seeds in a capsule may also be conjectured to be normal:

Since the independent variable is the same for these curves, they may be plotted on the same graph. In doing so, we note that the tails of the curves will always intersect, and there may indeed be significant overlap of the two curves as shown:

In the figure immediately above, the yellow area represents seeds that will germinate after being bleached for time t1; the red area indicates seeds that would be killed by bleaching for time t1; the orange area indicates seeds that would germinate if bleached for a time greater than t1, and the pale blue area represents seeds that are impossible to germinate because the bleaching time required for them to germinate is greater than the bleaching time sufficient to kill them.
We can see that in this model, the possibility of successful germination of the seed from a given capsule depends on the degree of overlap of the two curves. Fortunately, for many Cypripedium species, capsules often do produce germination approaching 100%. There remain some species, that at least for me, repeatedly germinate poorly if at all. The model presented here shows that there is a theoretical possibility that seeds of some species never will be successfully stimulated to germinate to high percentage simply by use of a bleaching pretreatment.
I would like to state most emphatically, however, that the failure of seeds to germinate, even after employment of a wide range of bleaching times, may be caused by many other factors, perhaps most importantly by improper composition of the germination medium. I did not present this model so as to discourage propagators, and they should most definitely not assume that even repeated failure to germinate seed from a given capsule is necessarily due to an overlap of the germination and mortality curves. Rather, the right outlook - the productive attitude, is to assume that some factor besides bleaching is to blame and to carry on with many more experiments.


Steele, W.K. (1996): Large Scale Seedling Production of North American Cypripedium Species. Pages 11-26 in North American Native Terrestrial Orchids Prop. and Prod. Conference Proceedings March 16 & 17, 1996. Germantown, Maryland.


back / zurück