Studies On Rhododendron Dieback Caused By Botryosphaeria dothidea
W.H. Wills and R.C. Lambe
Virginia Polytechnic Institute and State University
W.H. Wills is Professor of Plant Pathology and R.C. Lambe is Associate Professor of Plant Pathology at Virginia Polytechnic Institute and State University. This study was made possible by funding from the American Rhododendron Society Research Foundation.
Among nurserymen, plant pathologists and rhododendron growers, there is a widely held view in the eastern United States that twig die-back and blight caused by the fungus Botryosphaeria dothidea is one of the most serious disease problems encountered in growing rhododendrons. Symptoms of this die-back include sudden wilting of the leaves of one or more branches, followed by death of the foliage which usually remain attached. Sometimes there is defoliation of the younger portions of the branches. Brown discoloration of the wood of affected branches usually occurs in a streak on one side of the branch. Jones and Benson in North Carolina (3) compared the symptoms of this dieback with the symptoms of die-back caused by several species of Phytophthora. There are superficial similarities between the two diseases, and culturing of the fungus is often necessary for a certain diagnosis. Botryosphaeria dieback may occur at any time of the year.
Some general assumptions are made about dieback caused by Botryosphaeria on rhododendron and on other susceptible woody plant species. These include (1) that it is a wound parasite requiring a wound or other opening on the host surface for entry; (2) that disease caused by the fungus is related to environmental stress, i.e. desiccation, freezing; and (3) that control is difficult to obtain. These assumptions should be addressed critically.
Diebacks and cankers caused by fungi described as B. dothidea or other species of Botryosphaeria occur on a wide range of woody plants (1, 2, 4, 5, 6, 7, 10). Although the pathogen has been described for some of these diseases and host-pathogen relations have been studied, there seems to be no extensive study of the host-pathogen relations in rhododendron dieback or of variation in the pathogen. This article summarizes some of the results of a study, funded by the American Rhododendron Society Research Foundation in 1982-84, of inoculation of rhododendron 'Nova Zembla' with certain isolates of B. dothidea and the potential for fungicidal control of dieback.
A collection of Botryosphaeria isolates was made by culturing the fungi from diseased twigs of 23 different cultivars of rhododendron from various locations in Virginia. In addition, a collection of 12 isolates was made from 12 different cultivars in a single home planting in Blacksburg, Virginia. The isolation techniques used yielded mass isolates which might be expected to contain genetically diverse material. Therefore, a collection of single-spore isolates was made from each of two of the latter group of isolates. These cultures provided the material for cultural and pathogenicity studies and the inoculum for the control studies.
Twenty-eight isolates selected from both groups were grown in an incubator without light and a smaller group of 18 isolates was grown under artificial illumination in the laboratory. A few isolates showed noticeably slower growth rates than that of most of the isolates and some few were intermediate in growth rate. The colony color of individual isolates varied from light gray to almost black. A few colonies produced pycnidia (asexual fruiting bodies). Some produced structures resembling pycnidia but in which no spores were formed. Most produced no reproductive bodies in culture and none formed sexual fruiting structures. The variation in growth rate and colony color was as great among single spore isolates from a single mass isolate as among isolates from various sources. All this indicates a great natural variation within an asexually reproducing population which probably gives the species great adaptability, especially if physiological characteristics related to growth and pathogenicity are as variable as the characters observed.
In one experiment in which an attempt was made to relate pathogenicity to asexual spore production, the relationship found was inverse. Contrary to expected results, the isolate which more aggressively colonized and destroyed host tissue did not produce spores and an isolate which sporulated was the least aggressive in colonizing and causing necrosis of host tissue. Wounding as a precondition to entry and colonization of host tissue was examined by several techniques. In one method excised branch ends about six inches long were maintained with cut ends immersed in water and inoculated with blocks of fungus vegetative tissue where buds were removed or stems were notched with a blade. In another type experiment excised sections of stems were maintained in deep-dish covered glass moisture chambers and inoculated in various ways. Using yet another technique intact potted plants had stems injured by bud removal or notching and were inoculated with blocks of vegetative mycelium of the fungus. In all three types of experiments entry of the pathogen could be traced to a wound in the stem or to a natural opening such as a leaf scar. In the absence of such portal of entry infection was not accomplished. Inoculum consisting of spores washed from fruiting bodies of the fungus were used in a few instances. In those cases also an opening in the surface of the host appeared to be necessary for colonization to occur. All our experiments confirmed the role of the fungus as a wound pathogen, but additional experiments should be performed using spores rather than vegetative tissue as inoculum.
All isolates of the fungus, both mass isolates and single-spore isolates, were able to colonize rhododendron stems and cause necrosis. The principal difference among isolates appeared to be in the rate of colonization of host tissue. Those rated as more pathogenic colonized faster than those rated as less pathogenic once entry had been effected. Relative susceptibility of cultivars of rhododendron has not been studied but should be. The popular cultivar 'Nova Zembla' appears to be quite susceptible.
Experiments were designed to test the potential for control of this disease with fungicides. No attempt was made to survey a range of materials for relative efficacy; rather, eight fungicides currently available on the market were tested in vitro at three different concentrations for inhibition of a single isolate of the pathogen. Four of these fungicides with the greatest activity at the lowest concentration were then tested against eight isolates of the pathogen in vitro. These four were then tested for reduction of lesion formation in the host when the pathogen was artificially inoculated into stems and the fungicides taken up from solutions in which the cut ends of excised stems were immersed. All of the four caused reduction of lesion development by a weakly pathogenic isolate and the use of three of them resulted in reduced lesions in stems inoculated with a highly pathogenic isolate. In all variations of these tests benomyl (Benlate 50W«) was the most effective fungicide tested and was used in all subsequent experiments.
Additional experiments were conducted on wound-inoculated greenhouse-grown plants treated by spray or drench application with benomyl. A high level of control was obtained when benomyl was sprayed on the plants immediately following wounding, and artificial inoculation with the pathogen was less effective when inoculation immediately followed spraying. These experiments were designed to simulate what might happen while pruning plants in a nursery operation. The results suggest that benomyl sprays could be effective in preventing natural infection of stems wounded by pruning. However, infection could still occur at other times through other portals of entry. Other fungicides and methods of application should be studied.
The results of these studies have been presented in talks before the Potomac Division of the American Phytopathological Society (8, 9) and are in preparation as technical papers. Additional studies which should be made include screening of rhododendron cultivars for resistance to this disease. In order to effectively screen for resistance, inoculation methods should be improved; specifically, a method of inoculation should be developed using spores as inoculum. This, in turn, will require development of a reliable method to produce spores in quantity. Finally, studies should be made of the effectiveness of combinations of pruning dead tissue and fungicide application for control.
1. Brown, E.H. and F.F. Hendrix. 1981. "Pathogenicity and histopathology of Botryosphaeria dothidea on apple stems." Phytopathology 71:375-379.
2. Gerlach, W.W.P., H.A.J. Hoitinik and C.W. Ellett. 1974. "Shoot blight and stem dieback of Pieris japonýca caused by Phytophthora citricola, P. citrophthora and Botryosphaeria dothidea." Phytopathology 64:1368-1370.
3. Jones, R.K. and D.M. Benson. 1984. "Rhododendron diseases." J. Am. Rhododendron Soc. 38:154-156.
4. Milholland, R.D. 1972. "Histopathology and pathogenicity of Botryosphaeria dothidea in blueberry stems." Phytopathology 62:645-659.
5. Schreiber, L.R. 1964. "Stem canker and dieback of rhododendron caused by Botryosphaeria ribis." Gross & Dug. Plant Dis. Rep. 48:207-210.
6. Spiers, A.G. 1977. "Botryosphaeria dothidea of Salix species in New Zealand." Plant Dis. Rep. 61:664-667.
7. Weaver, D.J. 1979. "Role of conidia of Botryosphaeria dothidea in the natural spread of peach tree gummosis." Phytopathology 69:330-334.
8. Wills, W.H. and R.C. Lambe. 1985. "Pathogenicity of Botryosphaeria dothidea to Rhododendron cv. Nova Zembla." Phytopathology 75:630 (Abstr.).
9. Wills, W.H. and R.C. Lambe. 1985. "Potential for fungicidal control of dieback caused by Botryosphaeria dothidea on Rhododendron." Phytopathology 75:630 (Abstr.).
10. Wolf, F.T. and F.A. Wolf. 1939. "A study of Botryosphaeria ribis on willow." Mycologia 31:217-227.