To compile our insect list, we started with a list of non-indigenous insects native to Europe that feed on woody plants in North America (Mattson WJ et al. 1994, Mattson W. et al. 2007). Species that did not meet our criteria (e.g. citrus feeders) were deleted. We added additional insect species that met our criteria, including insects native to regions other than Europe and recently established insect species, based on published reports and communication with experts (e.g. Haack 2006, Langor et al. 2009, R. Rabaglia, US Forest Service, pers. comm.).
For each insect species on our list, we recorded taxonomy (order, family, genus, species), year or approximate year that the species was detected or identified in the US (when known), feeding guild, and primary host(s) species (when available). For the family Curculionidae, we distinguished between the subfamily Scolytinae, which includes bark and ambrosia beetles, and other curculionids (weevils) for some analyses because of differences between these groups in behavior, survey efforts and potential impacts. Insects in the order Hemiptera were grouped by suborder for some analyses when we wished to distinguish among Prosorryncha (true bugs), Clypeorryncha (leafhoppers, froghoppers), and Sternorrhyncha (aphids, adelgids, scales, whiteflies).
When the date of detection for a given species was approximate, we used the following guidelines for analysis: approximately 1950 = 1950; < 1950 = 1949; 1950s = 1955; 1800s = 1850. When more than one date was reported, we used the earliest date for analysis (e.g. 1950; 1956 = 1950). In some cases, the detection date was noted only as occurring in the 'early', 'mid', or 'late' portions of the 19th or 20th century. For these species, we divided the century into thirds and used the mid-point for analysis. For example, 1816, 1850, or 1883 were used for species first identified in the 'early', 'mid', or 'late' 1800s, respectively. Similarly, we assigned dates to species first identified in the 'early', mid', or 'late' portion of a given decade. For example, 1922, 1925, or 1928 were used for pests first identified in the 'early', mid', or 'late' 1920s, respectively. Although we used the earliest date of detection for our analyses, we recognize that lag times are inherent in these records.
Insect species were assigned to feeding guilds based on their dominant or most damaging feeding mode. Defoliators included insects that feed externally or internally on foliage, including leaf miners and shoot-feeders. Borers included species that feed aboveground on phloem, cambium and/or wood. Sap-feeders included gall-forming adelgids, as well as other insects such as scales, aphids, psyllids and whiteflies. Species that feed primarily on roots, seeds, cones or fruit were assigned to the “Other” guild. Insects that feed on different plant tissues as immatures and adults were assigned to the feeding guild most associated with injury or damage. For example, emerald ash borer was assigned to the borer guild because phloem-feeding by larvae causes tree mortality, while foliage-feeding by adults causes negligible injury.
Host breadth of each species was also included in our list, following the classifications of Niemela and Mattson (1996). Species with a host range confined to a single genus were recorded as monophagous, while species that feed on multiple genera of a single plant family were recorded as oligophagous. Species that feed on hosts representing multiple plant families were considered to be polyphagous.