Linaria vulgaris

Zouhar (2003) reviewed the subject as follows: ‘Several insect species that feed on L. vulgaris have been purposely or accidentally released in the USA and Canada. The flower feeding beetle Brachypterolus pulicarius and seed-feeding Gymnetron antirrhini (= Rhinusa antirrhini) appear to be the most important insects for reducing seed production. B. pulicarius larvae develop inside floral ovaries, and adults feed on buds and young stems. B. pulicarius can reduce seed production by 80 to 90%. G. antirrhini can reduce seed production by 85-90%. McClay (1992) recorded a 74% reduction in seed viability due to infestation by Brachypterolus pulicarius;however, Mason and Gillespie (2013) stated that ‘ B. pulicarius and G. antirrhini are already widespread in Canada but the impact is unknown. Egan and Irwin (2008) also obtained very variable results. There were reductions of seed and shoot production of L. vulgaris, but these were modest and minimal at high densities of the weed. Harris (1981), however, commented on a decline of the weed (in Canada) due to Brachypterolus pulicarius, Gymnetron antirrhinii and Calophasia lunula, and Quartes (2007) stated that seed weevils (such as G. antirrhini, have already led to the decline of L. vulgaris throughout Canada. The moth Calophasia lunula, a long-term accidental introduction to Canada, and two other insects, the moth Eteobalea serratella (approved for release in Canada in 1991) and the root-galling weevil Gymnetron linariae [ Rhinusa linariae ](approved for release in Canada in 1995) are referred to by McClay and De Clerck-Floate (2001), but were apparently not effective enough to deserve comment in later publications. Also, the seed capsule weevil R. neta appears to occur only in scattered populations in the eastern and western USA and Canada (Wilson et al., 2005).<br>Among other agents reviewed by Mason and Gillespie (2013), Mecinus janthinus has been released at numerous sites in Canada since 1996, but with mixed results, possibly because the closely related M. janthiniformis has sometimes been involved, which is more specific to L. dalmatica. Toševski et al. (2013) have now described a fast and accurate way of distinguishing the two Mecinus species haplotypes using PCR-RFLP diagnostic assay of the mitochondrial cytochrome oxidase subunit II (COII) gene. At a site in Alberta where M. janthinus was released in 1996, observations in 2012 found that the population of L. vulgaris had declined to a very low density, with the few remaining stems heavily infested with M. janthinus (Mason and Gillespie, 2013). More recently studies have started on another stem-boring weevil, M. heydenii (Tosevski et al 2016).<br>More recently, the stem-galling weevil weevil Rhinusa pilosa has been approved for release in USA (Gassmann et al., 2014). Although not quite specific to L. vulgaris, it is considered to be a suitable biocontrol species due to its high host specificity, minimal risk to related species in North America, robustness during rearing, impact on host growth and reproduction through galling and expected population release from the effects of a European gall intruder, the inquiline weevil Rhinusa eversmanni (Mason and Gillespie, 2013).<br>The considerable genetic diversity in L. vulgaris (Ward et al., 2008) may help to explain its variable response to biocontrol organisms.