Agromyza canadensis Malloch

(Figs. 93, 94)

Material examined. ARKANSAS: Pope Co., Hector, Bayou Bluff RecreationArea, 4.v.2022,em. 13–15.iii.2023, C.S. Eiseman & J.A. Blyth, ex Andersonglossum virginianum, # CSE7997, CNC2104680 (1♀); MASSACHUSETTS: Berkshire Co., Egremont, Fenton Brook, 42.141158, −73.447775, 13.ix.2024, em. 15 – 23.x.2024, C.S. Eiseman, ex Myosotis scorpioides, # CSE9128, CNC2130722–23 (2♀); MICHIGAN: Ingham Co., Okemos, Ted Black Woods, 19.vi.2019, em. 13–17.vii.2019, C.S. Eiseman & J.A. Blyth, ex Hackelia virginiana, # CSE5585; same collection, em. 24–29.iii.2020, # CSE6087, CNC1904874–8 (3♂ 1♀ 1 puparium); WISCONSIN: Lafayette Co., Blanchardville, 42.786, −89.929, 27.vi.2019, em. 26.iii.2020, C.S. Eiseman & J.A. Blyth, ex Hackelia virginiana, # CSE6094, CNC1905409 (1♂).

Photographed mines. MINNESOTA: Isanti Co., 45.421296, −93.197401, 30.x.2022, M. Zappa, empty mines on Hackelia virginiana (iNat 140607742); NEW JERSEY: Passaic Co., Wayne, 40.924622, −74.2863, 24.vi.2020, S. Rall, empty mine on Hackelia virginiana (iNat 51242772); OHIO: Muskingum Co., 39.90117, −82.118148, 31.v.2020, M. Spring, occupied mine on Hackelia (iNat48058567); PENNSYLVANIA: Cumberland Co., 40.186162, −76.967763, 28.v.2023, tuesday_, empty mine on Hackelia virginiana (iNat 165627181); VERMONT: Chittenden Co., South Burlington, Winooski Gorge, 27.vi.2013, C.S. Eiseman, occupied mines on Hackelia virginiana (iNat 188557484); Washington Co., Calais, 44.32672, −72.472301, 28.vii.2021, E. Mitchell, occupied mine on Hackelia (iNat 96430544); WISCONSIN: Dane Co., Crystal Lake Wildlife Area, 19.x.2019, T. Klein, occupied mine on Hackelia virginiana (BG 1741425); MANITOBA: Division No. 14, Warren, 17.vi.2023, C. Friesen, occupied mines on Hackelia deflexa (iNat 167928508).

Hosts. Boraginaceae: * Andersonglossum virginianum (L.) J.I.Cohen, Cynoglossum officinale L. (Priest et al. 2020), [Hackelia deflexa (Wahlenb.) Opiz], * H. virginiana (L.) I.M.Johnst., * Myosotis scorpioides L.; also see Comments.

Leaf mine. (Figs. 93, 94) An upper-surface blotch, often elongate, with either no initial linear portion or a short, irregular one; brown with clean, whitish margins; frass in irregular black particles that are scattered in a broad central trail; distinct concentric feeding lines often present. Mines of two or more larvae may coalesce, and larvae are able to exit their mines and establish secondary mines in the same or another leaf (Fig. 94). Larvae exit and enter through the upper epidermis.

Puparium. Reddish-brown; formed outside the mine.

Phenology and voltinism. The only reared specimens have been collected as larvae in early May (Arkansas) and June (farther north), with adults emerging either in 3.5–4 weeks or not until the following spring. Larvae have been found in the northern USA and southern Canada in late May, mid- to late June, late July, and mid-October.

Distribution. USA: *AR, CA, *MA, MI, [MN, NJ, NY (Eiseman & Lonsdale 2018), OH, PA], VA, [VT], *WI; Canada: [MB], ON, QC (Priest et al. 2020; Lonsdale 2021); also see Comments.

Comments. This is the unidentified species on Hackelia that Eiseman & Lonsdale (2018, p. 13) reported has the rare ability among agromyzids to exit its mines and form new ones elsewhere. Spencer (1969 a, 1981, 1990) noted that the genitalia of Agromyza canadensis are very similar to those of the European Boraginaceae feeder A. pseudorufipes Nowakowski, which differs externally in having much darker body coloration. Sehgal (1971) reported A. canadensis from YT based on a single male reared from Mertensia paniculata (Aiton) G.Don by G.C.D. Griffiths; he also recorded this species from AB based on numerous leaf mines on the same host. Sehgal noted that Griffiths’ specimen was dark, casting doubt on the separation of A. canadensis and A. pseudorufipes based on color, and he suspected that the two are synonymous since the genitalia “appear indistinguishable”. As the two species are still considered distinct, it is perhaps more appropriate to treat Griffiths’ specimen as A. pseudorufipes. Sehgal’s records were not mentioned by Spencer (1981, 1990), but seem to be the only basis for Mertensia being listed as a host for A. canadensis by Spencer (1990) and Benavent-Corai et al. (2005).