We examined changes in understory vegetation under an intact forest canopy during the first decade following the deposition of tephra (aerially transported volcanic ejecta) during the 1980 eruption of Mount St. Helens, Washington State, USA. Objectives were (1) to document vegetation response to a major disturbance that has received little attention but is widespread and relatively frequent in the northwestern United States, and (2) to analyze vegetation responses in terms of characteristics of the disturbance, responses of growth forms as well as those of species, components of vegetation change, and species autecology. We used permanent plots at four study sites, representing two tephra depths (≈ 4.5 and 15 cm), to examine understory vegetation change in old-growth, subalpine conifer forests. The two sites at each tephra depth differed in understory vegetation and amount of snowpack at the time of disturbance. At each site, plant cover and density were measured in 100 1-m2plots with undisturbed tephra covering the soil surface, in 50 1-m2plots from which the tephra was removed in 1980 (cleared plots), and, at 1 site, in 50 1-m2plots from which the tephra was removed in 1982. Values in cleared plots were used to estimate pre-eruption vegetation composition and to calculate inertia and components of resilience. After a decade, the impact of the tephra was still pronounced: cover was reduced for bryophytes at all sites, for herbs in deep tephra, and for shrubs where deep tephra had fallen on snow. Most cover was contributed by plants that survived the eruption. Vascular species initially absent from the site contributed little to the vegetation. Seedlings of most herb and shrub species did not survive, and many surviving species produced no seedlings. In contrast, at least two species of weedy, widespread mosses (Ceratodon purpureus and Pohlia annotina) extensively colonized the tephra surface. Although cover of surviving small trees changed little after the eruption, many seedlings of the shade-tolerant conifer species that dominated the canopy established and survived well, with the most germinants and highest survival on undisturbed deep tephra. Tsuga spp. constituted a higher proportion of the new seedlings than of the pre-eruption seedling population. A dense layer of conifer saplings appeared to be developing, unlike any present before the eruption. We calculated inertia (the percentage of pre-eruption importance remaining after the eruption) and four measures of resilience for each understory growth form. For shrubs, herbs, and bryophytes combined, inertia and one measure of resilience, the importance of the growth form at the end of the decade as a percentage of pre-eruption importance (called "c/a"), were highly correlated; both components increased as tephra depth decreased, as plant size increased, and where the snowpack had melted before the eruption, and both were higher for species richness and shoot density than for cover. Inertia for species density decreased with tephra depth and increased with plant height. Regressions predicting inertia for single growth forms included only a single, and different, environmental factor for each growth form: for shrubs, cover of snowpack; for herbs, tephra depth; and for bryophytes, light intensity. Other measures of resilience were less variable than c/a. Although this disturbance did not generally favor establishment of new species or totally eliminate species from the system, it did modify greatly the structure, species composition, and overall abundance of the forest understory. Recovery to the original state will not occur soon. Even for this example of vegetation recovery in a single stratum, generalities are difficult to state, because the relationships of inertia and resilience to growth form and variation in the disturbance were so complex. Because the initial damage played the dominant role in determining understory patterns a decade later, it is critical to focus attention on details of the disturbance and mechanisms of survival in order to understand early vegetation recovery following tephra deposition.