Logo for the Journal American Rhododendron Society

Journal American Rhododendron Society

Current Editor:
Dr. Glen Jamieson ars.editor@gmail.com


Volume 46, Number 4
Fall 1992

DLA Ejournal Home | JARS Home | Table of Contents for this issue | Search JARS and other ejournals

The Color of Rhododendron Flowers
Benjamin Pecherer
Lafayette, California

        Among the great color spectacles of nature, the colors of flowers may be rivaled by the colors of sunrises, sunsets, the rainbow, and even by the color of gemstones. The last of these is probably reserved for the very wealthy, and the others are of a transient character and beyond our control. On the other hand, a floral display of many colors can be contrived by anyone who cares to make the effort, and with some thought it can be prolonged over a long interval of time. It's obviously of some interest to ask what is responsible for the color of flowers, and once we know, of what value is that knowledge.
        This is not a new question, and many cultures have investigated flower pigments and made use of them; saffron from the autumn crocus is a familiar example. In the latter part of the 1800s organic chemists began to investigate the matter to determine the structures and properties of these substances. Using the methods of classical organic chemistry, large quantities of flower petals were collected, the pigments isolated, and the constitutions established. Two of the best known investigators in this field were Richard Wilistaetter and Robert Robinson. The results of these investigations showed that the principal components responsible for the color belonged to a class of compounds known as benzpyranes, and that many different species of plants had similar, or closely related pigments in their flowers. The actual pigments are oxygenated derivatives of the benzpyranes, namely anthocyanins, flavanols, flavanones, and a few other closely related substances. The structures of these are shown in Figure 3.
        The pigments exhibit different colors in acid or alkaline solution, behavior similar to the litmus of high school chemistry. Additionally, the color is affected by traces of metals - usually becoming more deeply colored. The anthocyanins, when pure, are not too soluble in water, but they occur in the cells of the plant as a more or less loose combination with a variety of five and six carbon sugars, in which combination they are termed "anthocyanidins". The subtle variations in color that occur as the flower ages are due to slight changes in the acidity and the amount of metal ions that are present in the cells of the flower.
        Later, more careful investigations of plant coloring matters have shown that the color of many flowers is influenced by the presence of carotenoids - a class of substances related to beta-carotene, the precursor of vitamin A. These carotenoids are sensitive to the oxygen of the air and refined techniques are necessary for their isolation and characterization; hence the reason for their delayed detection. The formula of beta-carotene is shown in Figure 3.
        In addition to the above named substances, chlorophyll and its breakdown products also influence the overall color. Although the principal coloring matters of flowers have been firmly established by classical procedures, a modern investigation would make use of the wonderful new tools that have become available. A brief excursion into this domain follows.
        About 10 grams of flower petals are extracted with either weakly acidified methyl alcohol, or acetone if the carotenoids are sought. This is done at a low temperature - about 15-20 degrees below zero Centigrade to avoid loss due to oxygen. The insoluble petal substance is filtered, and the solvent removed at low temperature in a vacuum. A "goopy" residue remains. A small sample of the residue is subjected to thin layer chromatography, a procedure that separates the components of the mixture and reveals its complexity. The colored substances are apparent on inspection. They are recovered by scraping off the colored spots from the plate and washing the colored substance from the adsorbent. Removal of the solvent leaves a residue that is often crystalline. If not, the procedure is repeated once or twice more, which is usually sufficient. Using several plates, or plates with thicker layers of adsorbent it is not difficult to isolate 5-20 milligrams of substance. If larger amounts of material are required recourse is made to column chromatography. With the latter technique tens of grams of material can be obtained.
        The pure material is now ready for structure determination. The infra-red spectrum shows the presence of characteristic groups. The nuclear magnetic resonance spectrum reveals the types and locations of the hydrogen and carbon atoms; and finally the mass spectrogram in the low resolution mode yields the molecular weight. In the high resolution mode, the mass spectrum gives the molecular weight to four decimal places and hence the formula, or overall composition. This latter spectrum also gives the weights of the various fragments that appear, and these are integral parts of the whole molecule. From this information the complete structure can usually be assigned. In some cases it may be necessary to examine a derivative. With these modern instruments, a structure that once may have required years for its elucidation, can now be accomplished by a competent graduate student in a matter of hours.
        The schematic drawings of the rhododendron flower petals will aid in understanding how the various pigments contribute to the flower color. (See Figures 1 and 2.)

Flower petal cross-section at tip
 
Flower petal cross-section near a spot

The Nature and Distribution of the Pigments
In the solid state the pigments form dark metallic-looking crystals that bear little resemblance to the flower color that they produce. In the liquid portion of the cell - the cytoplasm - the pigments are present in a very dilute solution; they have great tinctorial power. Floating within the cytoplasm are highly organized structures containing higher concentrations of the pigments. These are the chromoplasts and are probably the site of the synthesis of the pigments and from which they diffuse into the cytoplasm. The overall color of the flower is due to the relative concentrations of the soluble pigment and that in the chromoplasts, and the presence of varying amounts of chlorophyll. It is not uncommon for there to be more than a single pigment in the cell.
        Another phenomenon that takes place in certain varieties of rhododendrons is the color change of the newly opened flower as it ages. Aside from the changes that are referred to above, there may be a growth phase during which more pigment is being synthesized and a second phase when the pigments are being degraded. It is possible that the color change is highly dependent on weather conditions since it does not always occur to the same degree.

Plant pigment formulas

The Flower Colors
Pure White Flowers: These contain no anthocyanins; their color is due to the air-filled lacunae, i.e., empty cells. The two yellow flavanols, kampferol galactoside and myricetin-5-methyl ether are present in small amounts. Earlier reports that other pigments are present are due to the inclusion of the blotch and spot colors. For example: R. 'Sappho', R. 'Jacksonii', R. taronense, R. catawbiense.
Yellow Flowers: In this group the colors range from yellow green to the deep lemon yellow, and this is due to the varying ratio of the pigment gossypetin/chloroplast. A few other flavanols dihydroquercitin, myricetin, and quercitin as glycosides are present. In the pale greenish-yellow R. ambiguum and the yellow R. lutescens, the presence of gossypetin is difficult to establish. It is believed that the chloroplasts and the breakdown products are the main causes of the color of these two species.
Orange Flowers: Gossypetin as the 3-rhamnoside is the most frequently occurring pigment. Note that in the yellow group, gossypetin 3-galactoside was the principal component. Carotenoids are encountered for the first time. As the color deepens the ratio of carotenoid to chlorophyll increases; additionally the ratio of chromoplasts to chloroplasts increases, indicative of a change from green to yellow. It is not certain whether any anthocyanins are present to contribute to the flower color. In the range yellow to orange violet blotches or throat color are never observed, with the exception of R. 'Sonata', R. dichroanthum x R. 'Purple Splendour'.
Pink, Red, Lavender and Purple Flowers: Anthocyanins are responsible for the colors in this group. In but one case, have chloroplasts been found together with the anthocyanins. This combination gives the very dark red-brown or red-black of R. sanguineum. Usually it is the chromoplasts that are encountered with the anthocyanins, and depending on their concentration in the epidermal cells, the color will vary from red-orange to pure deep red. The carotenoids that were plentiful in the orange colors are less so here. The main contributions to the color come from three cyanidin glycosides. The bluish reds contain small amounts of malvidin and delphinidin and small amounts of flavanols that contribute little to the color. In the lavender-purple group, the color is due to cyanidin, malvidin, and paeonidin in the form of their glycosides: chromoplasts are not seen, but the methyl ethers of flavanols are present. The role of carotenoids in this color group has not been clarified, but they must certainly be present. A further demonstration that heavy metals contribute to these colors is shown by in vitro experiments wherein yellow solutions of the flavanols are turned blue on addition of metal ions. In summary, the above results show that the flower color is the result of combinations of anthocyanins, flavanols, carotenoids, chloroplasts and chromoplasts. The subtle effects of cell acidity and the presence of heavy metals are likely but not precisely demonstrated.
        The information obtained in this research is of value from two points of view. Firstly, in some cases it confirms the classification of some species, and in others it requires a revision. Secondly, recent research has been able to determine the genes that are responsible for the production of certain pigments. In theory this would be expected to be a basis for plant breeding for specific color objectives. Unfortunately, this type of research is beyond the capabilities of rhododendron hybridizers, and they will have to depend on the old hit and miss methods.

References
Much of the material in this paper is based on information found in The Chemistry of Natural Coloring Matters by Fritz Mayer, New York, Rheinhold, 1943. This is an ACS monograph. Also a doctoral thesis presented by Wolfgang Spethman to the University of Hamburg: Intergeneric Relationships in the Species of Rhododendron Based on the Flavanoid and Carotenoid Content of the Flower Colors. The two first figures are taken from the latter work.

Benjamin Pecherer studied chemistry at the University of Michigan (B.S., 1937; M.S., 1938) and the University of Hokkaido (Sc.D., 1973). After moving to New Jersey to work at a pharmaceutical firm he began growing and crossing rhododendrons and with the help of friends started a North Jersey chapter of the ARS. At his present home in Lafayette, Calif. he continues his hybridization work, trying for seedlings that can withstand heat and water rationing. Dr. Pecherer has written on the hybrids of Dietrich Hobbie for the Journal.


Volume 46, Number 4
Fall 1992

DLA Ejournal Home | JARS Home | Table of Contents for this issue | Search JARS and other ejournals