In this study we show that spatial scale of investigation affects the estimated strength of competition for pollinators between an invasive and a native plant species. The effect of the invasive herb Oxalis pes-caprae on pollinator visits to the native herb Diplotaxis erucoides was studied when the invader was (1) totally present, (2) present on a large scale (of hectares) but absent on a small scale (of square metres), and (3) totally absent. No difference in number of pollinator visits to D. erucoides was found between treatment 1 and 3, i.e., between total presence and total absence of the invader. However, when the invader was removed in the small scale while still remaining in the large scale, a higher number of visits to the native was recorded. Our study thus shows the importance of incorporating multiple spatial scales to allow for investigation of hierarchical effects on competition for pollinators, and it suggests that small-scale studies of effects of invasive plants on pollinator visitation might risk overstating negative effects of the invader.

Nomenclature: Fauna Europaea, 2004; Tutin et al., 2001.