The Pathogen and Environmental Effects on its Development

D. destructiva, the pathogen causing dogwood anthracnose on leaves, asexual fruiting bodies (conidiomata) of form most often on the undersurface, beneath a trichome (Redlin 1992 ). Twigs with conidiomata remain on trees over winter and provide primary inoculum for new infection cycles in the spring. Under wet conditions, the conidia ooze from these fruiting bodies in a slimy white to beige or pinkish cirrhus. Most short-distance dispersal of conidia probably occurs via splashing rain, although dispersal by convergent lady beetles (Hippodamia convergens) (Colbyand others1995) and birds is possible (Brittonand others1993). Conidial germ tubes penetrate leaves directly (Grahamothers 1991). Necrosis precedes hyphal proliferation in palisade and spongy parenchyma cells (Walkinshaw and Anderson 1991), indicating toxin activity. Four phytotoxic phenols have been identified in culture filtrates of the pathogen (Venkatasubbaiah and Chilton 1991, Daughtrey and Hibben 1994, Daughtreyand others1996).

Light exposure, water relations, acid rain, and topographic factors have all been correlated with the incidence and severity of dogwood anthracnose. Research on the effects of these factors is summarized below.

There is an inverse correlation between leaf exposure to sunlight and disease severity. Infection is greater in shaded understory dogwoods (Gould and Peterson 1994, Hibben and Daughtrey 1988), in the interior and on the north side of canopies (Chellemi and Britton 1992, Parham and Windham 1992), and in trees with north to east aspects (Chellemiand others1992, Windhamand others1992). Light is important for both the vigor of the host and for production of anthocyanin, a light-activated disease-limiting compound. Also, reduced light is often correlated with reduced evaporative potential, whichpromotes conidial germination, infection, and lesion development. For example, Chellimi and Britton (1992) measured photosynthetically active radiation (PAR) values within three types of dogwood canopy, and found evaporative potential was more correlated with differences in disease severity than PAR. The heating effects of sunlight may also affect lesion type and sporulation (BROKEN-LINK Parham and Windham 1992, Daughtrey and Hibben 1994).

The amount of rainfall also affects the intensity of anthracnose infection. Periods of drought have preceded severe dogwood anthracnose epidemics (BROKEN-LINK Gould and Peterson 1994, Hartmanand others1992). (BROKEN-LINK Erbaugh and others 1994) hypothesized that drought mayincrease tree vulnerability to the disease. However, moisture also plays a key role in facilitating epidemics by fostering infection. In the Northeast, leaf symptoms begin to appear following the first extended rainy period after leaf expansion (Hibben and Daughtry 1988, Smith 1992). A 3-year study in North Carolina (Britton 1993) confirmed that disease severity was correlated with rainfall, particularly in the summer when temperatures averaged 64-70°F. Although rainfall clearly is conducive to infection, it also promotes host vigor and lessens disease impact (Gould and Peterson 1994, Williamsand others1987, Daughtrey and Hibben 1994).

Acidic rainfall may also predispose dogwoods to infection. Andersonand others(1993) reported that disease incidence and severity increased with decreasing pH (2.5 to 5.5) of caused bydestruction of leaf cuticles by the acid treatments (Anderson 1991, Brownand others1994, Thornhamand others1992), orby changes in soil nutrient status. The mechanism is complicated by the neutralizing effect dogwood leaves can have on acid droplets by releasing calcium and magnesium ions (Wiley and Hackney 1991, Daughtrey and Hibben 1994).

Topography of dogwood habitat may also affect disease incidence. At several sites in the Southern Appalachians, dogwoods growing on north-facing slopes had higherfrequencies of disease incidencethan did trees growing on east-, south-, or west-facing slopes (BROKEN-LINK Windham and others 1992, 1993). Aspect also influenced disease development; lesions developed more rapidly on the north side of trees than the east, west, or south sides of the same trees (Chellemiand others1992). Evaporative demand (an index of drying potential) was found to be lowest at sites with north-facing slopes. Early-morning fogs may be a disease-promoting factor. Disease-prediction models based on these geographic factors have been generally accurate when tested against field survey data (Chellemiand others1992, Langdonand others1992). Notably, Wilds (1997) found that although the degree of infection is influenced by elevation, slope curvature, slope position, and potential soil moisture, stem density alone explains 25percent of the variation in disease severity.

Proximity of trees to streams is also associated with greater dogwood anthracnose severity (Anderson 1991, Chellemiand others1992, Knighten and Anderson 1992). For example, dogwoods located within60 feetof streams declined more rapidly from anthracnose than did trees located further from streams (Knighten and Anderson 1993). These relationships mean that the highest mortality rates are restricted to dense stands in damp, sheltered sites at low slope positions. As a result, Wilds (1997) suggests, surviving populations of flowering dogwood may represent a biased genetic subset of the original population.