Abstract
Native flowering dogwood (Cornus florida) in parts of northeastern United States have shown increasing dieback and mortality since the late 1970’s. Symptoms consist of necrotic spots, scorch, and blight of the leaves, nondehiscence of some blighted leaves, cankers, epicormic branches, and dieback beginning in the lower branches. The anthracnose fungus Discula sp. was identified as the primary cause. Post-infection stresses may have intensified tree decline. The fungicides chlorothalonil (Daconil 2787) or man-cozeb (Manzate 200) applied at 10-day intervals during leaf expansion reduced foliar symptoms. Additional recommendations are pruning, fertilization, and watering to promote tree vigor.
Pirone (6) first reported a rapid deterioration of Cornus florida in New York and Connecticut in 1978 and 1979. He attributed the decline to infection by Colletotrichum gloeosporioldes in wet springs. In 1983, we reported (2) a lower branch dieback disease with the same symptoms on C. florida in New York, Connecticut, New Jersey, and Pennsylvania. A species of the fungus Disculawas consistently associated with diseased trees.
Byther and Davidson (1) reported a similar disease of western flowering dogwood (C. nuttallii) and named it dogwood anthracnose. In 1983, Salogga and Ammirati (7) reported dogwood anthracnose on C. nuttallii in Washington State, Oregon, Idaho, and British Columbia and they associated the fungus Discula sp. with the disease. We have since concluded that the same anthracnose disease occurs in both geographic locations. Dogwood anthracnose is the appropriate name for the disease of C. florida caused by Discula sp. This new disease is different from spot anthracnose caused by Elsinoe corni, a disease of C. florida primarily in its more southerly range (3).
In 1982, we began a study of declining dogwoods at Planting Fields Arboretum, Oyster Bay, Long Island and at a woodland site in Ossining, New York. This paper summarizes our observations on the symptomatology, etiology, and control of dogwood anthracnose as it affects C. florida in southeastern New York State.
Symptoms
Dogwoods of all ages and sizes were susceptible. Even the woodland population of dogwood seedlings was drastically reduced. Diseased trees in open sites remained alive, whereas most infected understory dogwoods in the woodlands died in 2-5 years.
The most characteristic symptom of dogwood anthracnose was the yearly twig and branch death beginning in the lower part of the canopy (Fig. 1). The disease was first called “lower branch dieback” (2).
Dogwood anthracnose disease causes lower branch dieback in Cornus florlda.
Additional symptoms differentiated dogwood anthracnose from other diseases of dogwood. Leaves showed small purple-rimmed spots and larger brown blotches (Fig. 2). The dead tissue sometimes weathered away so that infected leaves showed “shot holes” and appeared ragged. The blotches would sometimes expand until the entire leaf became blighted. Some blighted leaves remained on the shoots through the winter. Reddish brown-purple spots and brown necrotic blotches formed on flower bracts after rainy periods. Leaf lesions were often similar to those caused by Septoria cornicola, but were generally larger than the tiny (< 1/25 inch diam.) spots typical of the spot anthracnose disease caused by Elsinoe corni.
Purple-rimmed leaf spots (A) and blotches (B) caused on leaves of Cornus florida by Disculasp. infection.
Shoots became infected through the petioles of blighted leaves or from tiny sunken lesions that formed in the bark of shoots. The tips of infected shoots turned grey-tan and a purple zone formed between the dead and healthy bark.
Epicormic branches (watersprouts) formed on the trunk or main branches of diseased dogwoods. When the epicormic branches became infected and died back, annual cankers developed on the trunk or branch at the base of the dead epicormic branches. The cankers were detectable by sunken, swollen, or cracked areas in the bark. Cutting into the wood revealed brown, elliptical discolored areas beneath the bark (Fig.3).
Bark cracking around the base of a dead epicormic branch (A, at arrow) is a clue to the presence of a Discula canker (B).
Cause
Tiny conidiomata (fruiting bodies) proliferated on the dead leaf tissues and dead bark of twigs and epicormic branches. From characteristics of the conidiomata, the fungus Discula sp. was identified. The conidiomata appeared as reddish brown to black bumps on twigs and leaves when viewed at high magnification. When moistened, the conidiomata exuded single-celled conidia (Fig. 4) in gelatinous masses or tendrils. These spores initiated infection on the new leaves in the spring and on leaves that formed throughout the growing season. By the use of spore traps, conidia of Discula sp. were shown to be wind-spread during rainy periods.
A closup of a dead dogwood twig covered with Discula conidiomata oozing spores under moist conditions (A) and a microscopic view of Discula sp. conidia (x = 8.0 × 2Oum).
The Discula sp. was isolated repeatedly from portions of leaves, buds, and shoots of dogwoods showing disease symptoms. Successful inoculations of leaves and stems of healthy dogwoods with cultures of Discula sp. provided convincing evidence that this fungus is causing the anthracnose disease of C. florida.
Contributing Factors
Other diseases and insect injuries, most notably Armillaria root rot, Septoria leafspot, Botrytis blight of flower bracts and young leaves, dogwood borer (Synanthedon scitula), and dogwood club gall midge (Mycodiplosis clavula), were occasionally observed on dogwoods. We have interpreted these as “normal“ stresses on dogwood, not responsible for the marked decline of C. florida since the late 1970’s. Each of these agents, however, might have contributed to the death of dogwoods already stressed by dogwood anthracnose disease. Borer infestation has been associated with the recent incidence of decline in some dogwoods (8). Dogwood borer, however, is not considered a significant pest in woodland situations (4). We have observed only rare borer injury, but commonly severe anthracnose symptoms, in woodland sites. Native C. florida populations in the woodland understory in Catoctin Mountain Park, MD have also been severely injured by dogwood anthracnose (5).
There have been suggestions that stresses induced by moisture deficits, minimal snow cover, or severe winters have predisposed dogwoods to attack by an otherwise weak pathogen. The anthracnose fungi, including those attacking oak, maple, and sycamore, are strong pathogens and do not require a weakened host to become established. To reverse the predisposition scenario, it is more likely that infection by Discula sp. has predisposed dogwoods to injury from recent climatic stresses. For example, infected C. florida in our woodland site showed far greater winterkill than comparable healthy dogwoods.
Why dogwood anthracnose now?
Why is anthracnose suddenly affecting dogwoods? What about the coincidence of dogwood anthracnose on C. nuttallii in the Pacific Northwest and on C. florida in the Northeast? We have only educated guesses for now. The early reports of dogwood anthracnose near ports of entry on both coasts suggest the importation of the pathogen, similar to the entry of the Dutch elm disease fungus into the U.S. in the 1930’s. Although this possibility cannot be discounted, the sudden and widespread distribution of anthracnose on both hosts renders this explanation less tenable. Since its initial East Coast appearance in the vicinity of New York City, the disease has been reported on C. florida in parts of Massachusetts, Connecticut, New Jersey, Pennsylvania, Delaware, Maryland, West Virginia, Virginia, North Carolina and Georgia.
We suspect that dogwood anthracnose was present in the Northeast prior to the recent outbreak, but at a low level of infection. Some coincidental sequence of abiotic factors, most likely including temperature and moisture, dramatically intensified the impact of this disease on dogwood. The increased leaf infection and stem dieback then allowed the rapid build-up of abundant fungus spores, which provided inoculum for repeated cycles of new infection on dogwood.
Additional abiotic influences like drought and winterkill undoubtedly have intensified the decline of dogwoods already infected. Of the biotic agents, Armillaria mellea is the one most often seen attacking woodland dogwoods infected with dogwood anthracnose, hastening their death.
We anticipate the continuation of dogwood anthracnose, with disease severity varying yearly depending on weather conditions, especially during the spring months. Extended rainy periods during leaf expansion will particularly favor disease development.
Fungicide Trials
We conducted fungicide trials in 1985 and 1986. In 1985, spray applications at 14 day intervals were tested. No registered fungicides performed particularly well with such a long time between treatments. During the 1986 growing season, we attempted to identify fungicides which would be effective for leaf spot control when applied on a 10-14 day schedule. Two trials with the same fungicides but slightly different methodologies were conducted at the Long Island Horticultural Research Lab, Riverhead, New York and the Brooklyn Botanic Research Center, Ossining, New York (Fig. 5).
Results of fungicide spray trials in (A) Riverhead and (B) Ossining, NY during 1986. Sprays were applied at a 10-14 day interval with the materials indicated.
Second-year dogwood seedlings were in 36 2’ square field plots for the Riverhead trial. Trees were inoculated by suspending epicormic branches covered with Discula sp. sporulation one foot above the plots. Inoculum was supplied and spray treatments were begun on May 5th. Treatments were made in 6 replications, using the following fungicides applied on a 10-day schedule: Award 10W (5 oz/100 gal), Benlate 50W (0.5 lb/100 gal), Daconil 2787 75 WP (1.5 lb/100 gal), Manzate 200 80WP (1.5 lb/100 gal) and Pratt Bordeaux (5T/gal). Leaf spot counts were made on June 16, following a natural infection period that occurred with protracted rains on June 6-12. Of the materials tested, Daconil 2787 gave the best control, providing a 64% reduction in leaf spotting compared to unsprayed control plants. Award, Manzate 200 and Benlate also showed significant disease control, not significantly different from Daconil at the 5% level (Duncan's Multiple Range Test).
In Ossining, second-year dogwood seedlings were transplanted into 38 4’ square plots in a woodland clearing where natural inoculum was provided by the diseased native dogwood population. The same fungicides were used as in Riverhead, again with 6 replications. The first spray application was made May 5 when dogwood leaves were beginning to expand, the second was made 9 days later, and the last three sprays were made at approximately 14-day intervals. Plots were rated on July 10 for percent of infected leaves per tree for 4 trees/plot. Both Daconil 2787 and Manzate 200 gave a significant reduction in percent of infected leaves per tree (Duncan's Multiple Range Test). Other fungicides also reduced leaf spotting, but the results were not significant at the 5% level.
In both test locations, under different environmental conditions, Daconil 2787 and Manzate 200 provided significant reduction of leaf spotting of dogwood caused by Discula sp. Both of these fungicides are registered for use on Cornus florida, so we feel that sprays with these materials may be of use as one aspect of a multifaceted approach to control of dogwood anthracnose.
Control recommendations
None of the cultivars of C. florida that we examined in arboretum and nursery collections showed notable resistance to the anthracnose. The Kousa or Japanese dogwood (C. Kousa) has shown excellent resistance and can be recommended as a replacement for C. florida.
In our fungicide trials, chlorothalonil (Daconil 2787) or mancozeb (Manzate 200) applied at approximately 10-day intervals during leaf expansion provided the best protection of dogwood foliage. As with other tree anthracnoses, protective fungicides applied in the spring, when conditions are ideal for infection, are the most effective. For valuable landscape specimens, additional applications during the summer are advisable. Secondary infection of dogwoods can occur during wet periods throughout the growing season.
Cultural practices will also help to control dogwood anthracnose in the landscape. Dead twigs and branches of salvageable trees should be pruned to reduce the sources of spores for new infections. Epicormic branches should be clipped off to prevent their infection and the subsequent formation of branch and trunk cankers.
Improve tree vigor by applying a balanced fertilizer late in the fall or in early spring. Protect dogwoods in sunny exposures from borer attack. Dogwoods are particularly vulnerable to water stress because of their shallow root systems.
Supplemental watering should be done during drought periods in the summer and fall, but wetting the foliage should be avoided. Mulching trees may be of benefit in reducing root stress and discouraging lawn mower injuries to trunks. Dogwoods receiving good cultural care will be better able to withstand anthracnose during the years in which the disease is favored by environmental conditions.
Acknowledgments
This research was supported in part by Alpine Tree Care, Inc., White Plains, N.Y.; the International Society of Arboriculture Research Trust, the Long Island Arborists, and the New York State Arborists ISA Chapter, Inc.
- © 1988, International Society of Arboriculture. All rights reserved.
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.