AbstractTree gaps may be important habitats for the establishment and spread of understory herbs in eastern (US) deciduous forests. Data from past studies, however, have not shown any clear patterns for either species diversity or increased percent cover. The lack of clear patterns of response to gaps by herbs may be due to evolutionary differences among species or, in part, due to the fact that most studies have been short‐term and long‐term patterns have, therefore, not been observed. In addition, few studies have considered the responses of individual genets and/or ramets. We have been conducting a long‐term study ofCynoglossum virginianum, now in its 16th year, to determine subpopulation and individuals' responses in gap and non‐gap habitats. A larger percentage of individuals flowered in the gap subpopulation shortly after it was created. Seed production rates were also higher in the gap subpopulation. The long‐term benefits gained by the short‐term flowering response have, however, been few as all but two of the seedlings produced within a few years of gap formation have died. In addition, all of individuals that flowered and produced seeds decreased in size and most have remained in the smaller size classes. In addition, only one of the individuals in the gap subpopulation has flowered since the initial flowering response while plants in non‐gap subpopulations have flowered and produced seeds throughout the study period. Gaps may only be beneficial toC. virginianumif seeds are dispersed into non‐gap areas or if seeds can remain dormant in the soil until conditions are more suitable for se