A widespread dieback and decline of flowering dogwood (Cornus florida) was reported in southern New England in 1978 and 1979 (Pirone, 1980). Symptoms included leaf lesions, twig dieback and the formation of cankers on branches and trunk. Since the first symptom observed was sunken, ulcer-like leaf lesions, the disease was called anthracnose. The damage was attributed to infection by a fungus, later identified as Discula destructiva. It was commonly feared that the much-loved dogwood would disappear from the landscape and woodlands, as the native American elm and the chestnut had in previous years due to new diseases.

By 1983, the lower branch dieback was reported to occur in New York, Connecticut, New Jersey and Pennsylvania (Hibben & Daughtrey, 1988). Data collected in 1984 indicated that the disease was present in Maryland (Mielke & Langdon, 1986) and others reported that severe outbreaks of dogwood anthracnose were present as far south as Georgia (Britton, 1989) and as far west as Tennessee (Windham, 1989). More recently, anthracnose has been reported from most of the Midwestern states, as far west as the Mississippi River (Daughtrey et al., 1996). In short, dogwood anthracnose has been reported to occur throughout most of the natural range of the dogwood.

Infection occurs on the leaves and bracts, and proceeds through the petiole and into the twigs and branches of the tree. It is not known whether the fungus over-winters in the twigs or in leaf litter. Infection of flowering dogwood by the anthracnose fungus has been associated with acid rain (Britton, 1989), environmental stress (Daughtrey eta al., 1988), cool wet weather (Hibben & Daughtrey, 1988) and insect injury (Walton, 1986).

I began studies of dogwood anthracnose in 1990, with the following objectives:

• To identify the fungus causing dogwood anthracnose in Connecticut.
• To describe how dogwood anthracnose spreads in the landscape.
• To determine how the fungus over-winters in Connecticut.
• To determine the relative resistance to infection of the hybrids of C. florida and C. kousa.

What causes dogwood anthracnose?

The fungus that causes dogwood anthracnose, Discula destructiva, was named and described in 1991 (Redlin, 1991). The fungus grows slowly in culture, and exudes spores from fruiting bodies both in culture and on leaves. I sent samples from Connecticut leaves and twigs to various mycologists for comparison, and they confirmed that the fungus present on our dogwoods was indeed D. destructiva.

Even though the Connecticut fungus was identical to those in other locales in the U.S. where dogwood anthracnose was rampant, I did not observe the severe dieback and decline symptoms on dogwood that researchers in other states had attributed to anthracnose infection. In addition, I experienced difficulty in reproducing symptoms by artificially inoculating healthy dogwood trees. However, due to the consistent association of the fungus with diseased dogwoods, I accepted that this was the fungus causing leaf blight and twig dieback on our dogwoods.

How does dogwood anthracnose spread in the landscape?

To answer this questions, I sampled a grove of 15-year-old dogwood trees, beginning in 1991. Once each week, beginning at leaf emergence in the spring and continuing until leaf senescence in autumn, I collected 50 leaves from the trees, taking no more than 5 leaves per tree at any one sample. In the lab, I measured the leaf area, counted the number of anthracnose lesions, and measured the area of each lesion. I then calculated the lesion density (how much of the leaf area was occupied by lesions) and disease incidence (how many leaves had at least one lesion). I used these calculations as a measure of disease severity and disease incidence, respectively.

I found that both severity and incidence of dogwood anthracnose reached a peak at about six weeks after leaf emergence, about mid-to-late June (see table). Neither incidence nor severity increased significantly after that time, regardless of weather conditions. The increase in disease coincides with the rapid increase in leaf area in the early summer (Smith, 1992a, Smith, 1994). The lack of increase in disease severity or incidence after mid-July indicates that the disease is monocyclic; there is only one cycle of leaf infection per year, and only one crop of lesions on the leaf.

Anthracnose on dogwood in Connecticut is not the devastating epidemic reported from other areas where dogwoods are native. Defoliation due to dogwood anthracnose had been reported in other areas of the U.S., but I have not observed defoliation in any of the eight years of study on this disease (Smith, 1992b).

Infection of dogwood fruit by Discula destructiva, with subsequent reduction in seed germination, had been reported as well. In autumn of 1990, 1991 and 1992, I collected 400 ripe dogwood fruit and examined them visually for evidence of infection by D. destructiva. These same fruit were then washed in dilute bleach, rinsed with sterile distilled water, and placed on agar in petri dishes, to allow growth of any internal fungi from the fruit.

The many fungi growing from the fruit included Botrytis, Alternaria, Penicillium and Colletotrichum, but no fruit were found to be infected by D. destructiva. Infection of the seeds or fruit by D. destructiva was virtually non-existent in Connecticut (Smith, 1993). The isolate of Colletotrichum, C. acutatum, caused extensive rotting of the fleshy part of the fruit and staining of the seed coat. I examined the cotyledons of seeds with discolored seed coats, and found no staining or other evidence of damage by the fungus. I concluded that infection by C. acutatum was probably not reducing the viability of dogwood seeds or affecting the subsequent seedling survival.

Kousa dogwood and its hybrids with the flowering dogwood.

In 1993, I acquired one each of the then-new series of hybrid dogwoods, C. florida x C. kousa, produced by the Rutgers University plant breeding program, from a breeder in Tennessee, along with healthy kousa dogwoods. (Editor's note: Rutgers hybrids include 'Rutban' Aurora,^tm 'Rutcan' Constellation,^tm 'Rutdan' Celestial,^tm 'Rutlan' Ruth Ellen^tm and 'Rutgan' Stellar Pink^tm.) The hybrids were touted as being resistant to anthracnose, powdery mildew and dogwood borer, and combining all the good qualities of C. kousa (anthracnose resistance, winter hardiness, interesting plant architecture, striking flowers, and attractive fruit) with the beauty of C. florida. I planted these trees at the Experiment Station's Lockwood Farm, located in Hamden, CT. Trees were maintained on trickle irrigation and mulched with wood chips.

I have found that the hybrid dogwood trees indeed are very resistant to dogwood anthracnose; I have observed only one or two very small lesions in the four seasons since planting. These lesions were inconsequential in the health and survival of the leaves. The trees also are very resistant to powdery mildew, a disease that causes severe leaf deformation and defoliation on C. florida. Winter hardiness is satisfactory for this area; the trees sustained very little damage even after winters with severe ice and wind. However, the trees grow rapidly and tend toward legginess, and crotches are weak and prone to breaking. Bloom time of the hybrid trees is between that of C. florida (mid-May) and C. kousa (mid-June). Though resistance in hybrid dogwoods to anthracnose and powdery mildew is high, the trees have drawbacks of their own that may make them problematic to use in some landscapes.

In summary, I have found that dogwood anthracnose is not the devastating disease that will erase dogwoods from the Connecticut landscape, but rather it is a leaf-spotting disease of comparatively little consequence. Of more concern to the health of landscape dogwoods is powdery mildew, another fungal disease that I have observed to cause severe leaf deformation, discoloration, premature senescense and defoliation. Cornus florida, especially C. florida rubra, is very susceptible to this disease. The hybrid dogwoods offer hope for reducing the consequences of anthracnose and powdery mildew in the landscape.

Written by Dr. Victoria L. Smith, Department of Plant Pathology and Ecology, Connecticut Agricultural Experiment Station, New Haven, CT 06504.

References.

Britton, K.O. 1989. Temerature, pH, and free water on in vitro germination of conidia of a Discula sp. isolated from dogwood anthracnose lesions. Phytapatholagy. 79:1203.

Daughtrey, M.L.; C.R. Hibben and G.W. Hudler, 1988. Cause and control of dogwood anthracnose in Northeastern United States. Journal of Arboriculture. 14:159-164.

Daughtrey, J. L.; C.R. Hibben, K.O. Britton, M. T. Windham, and S.C. Redlin. 1996. Dogwood anthracnose: understanding a disease new to North America. Plant Disease. 80:349-357.

Hibben, C.R. and M.L. Daughtrey. 1988. Dogwood anthracnose in Northeastern United States. Plant Disease. 72:199-203.

Meilke, M. and K. Langdon. 1986. Dogwood anthracnose fungus threatens Catoctin Mountain Park. Pages 6-8 in National Park Service, Park Science, Winter.

Pirone, P.P. 1980. Parasitic fungus affects region's dogwood. New York Times. Feb. 24, pp. 34, 37.

Redlin, S.C. 1991. Discula destructiva sp. nov., cause of dogwood anthracnose. Mycologia. 83:633-642.

Smith, V.L. 1992a. Quantitative estimation of anthracnose of flowering dogwood in Connecticut. Phytopathology. 82:247-248.

Smith, V.L. 1992b. Lack of defoliation due to anthracnose in landscape and woodland dogwood trees. Phytopathology. 83:247.

Smith, V.L. 1993. Infection of dogwood fruit in Connecticut by Colletotrichum acutatum. Plant Disease. 77:536.

Smith, V.L. 1994. Dogwood anthracnose severity is influenced by timing of spring rains. Frontiers of Plant Science, Vol. 46, Number 2, Spring 1994.

Walton, G. S. 1986. Association of dogwood borer with the recent decline of dogwood. Journal of Arboriculture. 12:196-198.

Windham, M.T. 1989. Epidemiology of dogwood anthracnose in Cherokee National Forest in southeastern Tennessee. Phytopathology. 79:1149.

reprinted from Yankee Nursery Quarterly. Winter 1998. Vol. 7, No. 4. pp 3-6.