EXOTIC BIRCH-LEAFMINING SAWFLIES (HYMENOPTERA: TENTHREDINIDAE) IN ALBERTA: DISTRIBUTIONS, SEASONAL ACTIVITIES, AND THE POTENTIAL FOR

COMPETITION

SCOTT C. DIGWEED and JOHN R. SPENCE~ Department of Biological Sciences, University of Alberta, Edmonton, Alberta, Canada T6G 2E9

and DAVID W. LANGOR Canadian Forest Service, Northwest Region, Northern Forestry Centre, 5320 - 122 Street, Edmonton, Alberta,

Canada T6H 3S5

Abstract The Canadian Entomologist 129: 319-333 (1997) The exotic birch-leafmining sawflies Fenusa pusilla (Lepeletier), Profenusa thomsoni (Konow), and Heterarthrus nemoratus (Fallen) occurred in Alberta during 1992-1995, but only the first two were abundant. Birch-leafmining sawflies occurred at all sites surveyed in central and southern Alberta, and appeared to be expanding their ranges northward. Adult F, pusilla began emerging in mid-May (approximately 220 DDo~), and there were one to three generations per year, depending on location and year. Female F.pusilla were relatively less abundant on young than on mature trees. Profenusa thomsoni began attacking trees in June (between 400 and 550 DDo~), and was invariably univoltine. Both species were more abundant and were active earlier on urban trees than at a nearby rural location. The highest catches and most complete representations of seasonal activity were obtained using yellow sticky traps. Larval F. pusilla and P. thomsoni are unlikely to compete directly for leaf resources because their leafmining activities are separated spatially and temporally, but they probably compete intraspecifically.

Digweed, S.C., J.R. Spence et D.W. Langor. 1997. Mineuses du bouleau (Hymenoptera: Tenthredinidae) exotiques en Alberta: Ripartition, activiti saisonnikre et possibiliti de compitition. The Canadian Entomologist 129: 319-333.

Resume Des mineuses du bouleau exotiques, Fenusa pusilla (Lepeletier), Profenusa thomsoni (Konow) et Heterarthrus nemoratus (Fallen), ont CtC trouvCes en Alberta entre 1992 et 1995, mais seules les deux premibres espbces Ctaient abondantes. Des mineuses ont CtC trouvCes B tous les sites inventories dans le centre et le sud de 1'Alberta et semblaient se rCpandre vers le nord. Les adultes de F, pusilla ont commencC i Cmerger i la mi-mai ( i environ 220 DJo5) et I'espbce produisait de une B trois gCnCrations par annCe, selon I'endroit et I'annCe. Les femelles de F.pusilla Ctaient relativement moins abondantes sur les jeunes arbres que sur les arbres B maturitk. La mineuse P. thomsoni s'est rCpandue dans les arbres en juin (entre 400 et 550 DJos), et I'espbce a toujours CtC univoltine. Les deux espbces Ctaient plus abondantes et elles Ctaient plus actives plus t6t dans les arbres des villes que dans les zones rurales avoisinantes. Ce sont les pibges collants jaunes qui ont donnC les tchantillons les plus abondants et les reprksentations les plus complbtes de l'activitk saisonnibre. Les lames de F.pusilla et de P. thomsoni ont peu de chance de se faire une compCtition directe pour les ressources foliaires, car leurs activitks alimentaires sont sCparCes dans I'espace et dans les temps, mais il y a sans doute compCtition intraspkcifique chez ces espbces.

[Traduit par la RCdaction]

'Author to whom reprint requests should be sent.

320 THE CANADIAN ENTOMOLOGIST MarchJApril 1997

Introduction At least five species of birch-leafmining sawflies from the tenthredinid tribes Fenusini

and Heterarthrini have been introduced into eastern North America from Europe since 1920 , (Britton 1924,1925; Dowden 194 1 ; Martin 1960; Lindquist and Thomson 1970; Smith 197 1 ; Nystrom and Evans 1989; Goulet 1992). The early birch leaf edgeminer (Messa nana Klug) and the birch leaf edgeminer [Scolioneura betuleti (Klug)] were introduced relatively recently, and remain confined to eastern Canada (Lindquist and Thomson 1970; DeClerck and Shorthouse 1985; Nystrom and Evans 1989), whereas the birch leafminer [Fenusa pusilla (Lepeletier)], the ambermarked birch leafminer [Profenusa thomsoni (Konow)], and the late birch leaf edgeminer [Heterarthrus nemoratus (Fallen)] spread to Alberta before 1970 (Drouin and Wong 1984). Our study focuses on Epusilla and P. thomsoni because they have become abundant in Alberta since the 1970s and there is public pressure to suppress their populations on urban birches.

Life history studies have been carried out in eastern North America for F. pusilla and P. thomsoni. Female F pusilla emerge in early spring, mate or reproduce parthenogenetically (Friend 1933), and oviposit in the mesophyll of young birch leaves (DeClerck and Short- house 1985). The first four larval instars feed within a growing blotch mine, and fifth-instar larvae emerge from the mine, drop to the ground, burrow into the soil, and construct pupal cells from soil and a mucilaginous secretion (Cheng and LeRoux 1965). Fenusa pusilla has one to three generations per year in Europe (Eichorn and Pschorn-Walcher 1973), three to four per year in Qutbec (Cheng and LeRoux 1965), and one to two per year in Alberta (Drouin and Wong 1984). The life history of P. thomsoni is similar to that of E pusilla, except that adults emerge in mid-summer, only females are known, there are six larval instars, and there is only one generation per year (Martin 1960; Eichorn and Pschorn-Walcher 1973; Togashi 1981; Drouin and Wong 1984; Martineau 1984; Schonrogge and Altenhofer 1992). Outbreak levels of both species were recorded often in eastern North America between 1950 and 1990 (Haack and Mattson 1993).

Many leafminers are confined to feeding within one leaf for their larval lifespan, and thus are susceptible to intra- or interspecific competition (Faeth 1991). Leafminers that do not compete use little leaf area (Faeth 1985) or are maintained at low levels by natural enemies (Faeth and Simberloff 1981). Fenusa pusilla and P. thomsoni may experience competition because both use much of the leaf to complete development (Friend 1933; Martin 1960) and both occur at high densities in Alberta (Drouin and Wong 1984).

Our study was made to expand on earlier observations of the life histories of birch- leafmining sawflies in Alberta, to help develop a pest monitoring and management plan in Edmonton, and to provide background for experiments examining birch-leafmining sawfly ecology. Birches in Alberta and the Northwest Territories were sampled to study changes in species presences and distributions of birch-leafmining sawflies since earlier reports, which showed that only F. pusilla, P. thomsoni, and H. nemoratus occurred in Alberta, and only in areas near Edmonton and Jasper (Drouin and Wong 1984; Quednau 1984). Seasonal activities of E pusilla and P. thomsoni were examined to provide background for experiments on mortality of immature birch-leafmining sawflies (Digweed 1995) and to plan timing of introduction of exotic biological control agents (unpublished data). Three methods of trapping adult leafminers were compared to determine the best method for monitoring pestiferous urban populations of birch-leafmining sawflies. We assessed the potential for competition within and between populations of Fpusilla and P. thomsoni to determine which competitive interactions (if any) could be further examined experimentally.

Materials and Methods The distributions and abundances of birch-leafmining sawflies were assessed

through a survey of native birch species (Betula papyrifera Marshall, B. neoalaskana

Volume 129 THE CANADIAN E"TT0MOLOC;ISI

-

United States

FIG. 1. Map of the 57 sites surveyed for birch leaf-mining sawflies in Alberta and the Northwest Tenitories during 1992-1994. @, leafminers collected; 0, no leafminers found.

Sargent, B. glandulosa A. Michaux, and B. occidentalis Hooker), introduced species (B. pu- bescens Ehrhart and B. pendula Roth), and probable hybrids of these. Fifty-seven sites were surveyed in Alberta and the Northwest Territories (Fig. 1) during the first week of September in one or more of the years 1992-1994, although some sites were surveyed during July or August as well. One mature tree having representative damage was examined at each site. The information collected from each tree included tree species and height, species of birch-leafmining sawflies present, and a visual estimate of the percentage of leaves infested with birch-leafmining sawflies.

Life histories of F. pusilla and P. thomsoni were studied on six (1992) or five (1993) mature birch trees from urban settings in Edmonton, Alberta (53O3 lfN, 113O3 l'W), and from a natural forest at the George Lake Field Station, approximately 80 km northwest of Edmonton (ca. 53"57'N, 114O06'W). All trees occurred on flat sites, but the urban trees were open-growing, surrounded by cut lawns or flowerbeds, whereas trees at George Lake were part of a poplar-spruce mixed forest canopy having a natural litter layer. In addition, activity of adults was monitored on five mature trees on the campus of the University of Alberta in Edmonton during 1994, on eight mature trees just south of the campus during 1995, and on

322 THE CANADIAN ENTOMOLOGIST MarchIApril 1997

six young trees (approximately 6 years old) at Sunstar Nurseries, Edmonton, during 1995. Only the eight trees sampled south of the campus of the University of Alberta during 1995 had been chemically treated for birch-leafmining sawflies during the 3 years previous to sampling; they received dimethoate treatments in 1994. Voucher specimens of adult birch- leafmining sawflies were deposited in the Strickland Museum, Department of Biological Sciences, University of Alberta.

Statistical comparisons of catches among locations, years, and leafminer species were made using the GLM procedure in SAS (SAS Institute Inc. 1988) assuming a = 0.05. Statistical differences were deemed 'marginal' if 0.05 < p < 0.10. All means are expressed

. as 'mean + standard error'. Departures from normality in the data were corrected using square-root or logarithmic transformations (Zar 1984), or Blom's normal rank scores (SAS Institute Inc. 1988).

In 1992, pyramidal emergence traps were used to monitor adult activity [modified from Martin (1977)l. Two sides of the traps were wooden and two were made of fine mesh ('no-see-um' style irregular weave; no opening larger than 0.1 mm2), and trap basal area was 0.22 m2. Insects were captured in a clear, ventilated plastic bottle inverted at the top of the trap. A clear plastic funnel glued into the lid of the bottle minimized escapes back down to the soil. Two traps were placed randomly within the dripline under each tree, one located permanently and one moved to a new, random location each week to ensure capture of possible multiple generations. Trap bottles were replaced two (George Lake) or three (Edmonton) times per week, and all adult leafmining sawflies were identified and counted.

During 1993 and 1994, sticky traps were used instead of emergence traps to monitor adult activity. In 1993, traps were 9.0-cm-diameter translucent white plastic lids smeared with Darina AX axle grease [modified from Kalcounis et al. (1992)l. In 1994, 10.0- by 15.0-cm commercial yellow sticky traps (Phero Tech Inc., Delta, BC) were used. One trap was hung from a branch approximately 2.0 m above the ground on each tree, and traps were changed one (1994) or two (1993) times per week. Traps were hung in a new randomly determined position around the tree crown each week to reduce the potential bias of sun exposure differences. Adults of trapped birch-leafmining sawflies were identified and counted.

Because different methods were used to trap adults in different years, it was impossible to partition catch variation into trap and year effects during 1992-1994. To standardize estimates of adult activity, and to determine the best trapping method for monitoring urban populations of birch-leafmining sawflies, all three methods were used on the eight mature trees south of the university campus during 1995. Numbers of E pusilla collected on yellow sticky traps from these trees were also compared with numbers from identical traps on young trees at Sunstar Nurseries, to evaluate differences in sex ratios of F. pusilla between young and mature trees.

To describe the seasonal activities of larval E pusilla and P. thomsoni, leaf samples were collected in Edmonton and at George Lake during 1992 and 1993. One branch having at least 15 leaves was collected from each of the upper, middle, and lower portions of each tree crown once per week. Branches were selected to contain a maximum number of leafminer larvae, and so did not accurately estimate relative distributions of larvae at different levels in the tree crown. However, we assume that sampling reflected relative differences in population levels between locations and years. Larval density estimates were standardized as 'larvae per leaf', calculated using all leaves from only those weeks in which live larvae were found (i.e. leaves having the potential to be larval habitat). The following data were collected for all leaves: leaf age (expanding or fully expanded); percentage damaged by F. pusilla or P. thomsoni, estimated using 11 percentage classes (0%, lo%, 20%, ..., 100%); number, species, and stage of leafminer eggs and larvae; number of parasitized leafminer eggs and larvae; and number of dead leafminer eggs and larvae. Leaf area was approximated

Volume 129 THE CAVADIAN ~ M O L O G I S T 323

during 1993 by using leaf lengths (base to tip along the midrib) and widths (at the widest point, usually near the base) in the following regression equation: area = 0.64 (length x width) -70.78 (r2 = 0.95, n = 50 leaves, 10 from each of five trees).

The potential for intraspecific competition within F. pusilla and P. thomsoni was esti- mated by calculating the leaf area used for development by individual larvae in Edmonton during 1993. This area was estimated by multiplying leaf area by the proportion of the leaf damaged by mining. Only leaves less than 100% mined upon completion of larval feeding, and containing only larvae that successfully completed their development, were included in this calculation. This was done to prevent underestimation of the leaf area required per larva, because larvae from completely mined leaves or that died prematurely may not have fed fully. Subsequently, these estimates were compared with the area available per leafminer larva on all occupied leaves, to estimate the percentage of larval F. pusilla and P. thomsoni having less than average leaf area available, and therefore potentially competing for leaf resources.

All seasonal patterns of activity were initially plotted on both calendar- and degree-day axes, but are presented only on calendar-day axes here. Degree-days were calculated using the modified sine wave method of Allen (1976) with a threshold of 5°C (DD,,,). The use of thresholds from 1 to 12°C did not greatly affect the fit of theoretical development curve models to empirical data for either F pusilla [R2 values for non-linear regression: range, 0.698 (8°C) to 0.728 (10°C); at S°C, 0.7201 or P. thomsoni [range, 0.902 (12°C) to 0.911 (1°C); at 5"C, 0.9091 (Digweed 1995). Given such low variation in R2 values, the threshold of 5°C was chosen because it is commonly used for insects (Pruess 1983), and because use of high thresholds results in slow degree-day accumulations that reduce model resolution (Lysyk 1989; Volney and Cerezke 1992). Degree-days were accumulated starting from 1 April because, on average, only 4.4 DD,, accumulate from 1 January to 31 March in Edmonton (Environment Canada 1992-1995). Meteorological data were provided by En- vironment Canada from the Edmonton Municipal Airport (Environment Canada 1992- 1995) and from Sion, Alberta (4 km south of George Lake).

Results

Species Distributions. Birch-leafmining sawflies were found at most of the 57 sites surveyed in Alberta and the Northwest Territories (Fig. 1). Profenusa thomsoni and F. pusilla were the most prevalent species (collected from 38 and 42 sites, respectively); H. nemoratus was rare (collected from seven sites). The latter was particularly abundant near Slave Lake and Lac la Biche (also reported in Cerezke and Brandt 1993). Neither M. nana nor S. betuleti was found at any sites surveyed.

In northern Alberta, a few larvae of F. pusilla, but none of P. thomsoni, were found at three sites between Peace River and High Level (Fig. 1). However, in the town of High Level, both F. pusilla and P. thomsoni occurred on young ornamental birches. Because neither species was found at naturally forested sites nearby, the sawflies were probably imported to the town with the ornamental trees. No birch-leafmining sawflies were found north of High Level during 1992 or 1993, although in 1994, P. thomsoni was detected at Yellowknife, Northwest Territories (62"30'N, 114'29'W).

Both F pusilla and P. thomsoni were found on all birch species examined. Larvae were able to develop successfully even in the small leaves of bog birch, B. glandulosa (see Wilkinson 1990 for leaf illustrations).

Species Abundances. Profenusa thomsoni was the most abundant species of birch- leafmining sawfly in central Alberta (Fig. 2a), and was significantly more abundant than E pusilla during 1992, and in Edmonton during 1994- 1995 (Fig. 2a; Table 1; most of the variation in the 'Species' source in 1994- 1995 occurred during 1994). The number of adult leafminers trapped per tree was higher in 1994 (330.4 f 55.5) than in 1995 (32.6 f 43.9)

THE CANADIAN ENTOMOLOGIST March/April 1997

- Emergence traps

Trapping method

FIG. 2. Mean (f SE) relative abundances of Fenusa pusilla and Profenusa thomsoni; (a) Edmonton, 1992-1995, and George Lake, 1992-1993 (sampling methods: 1992, emergence traps; 1993, white sticky traps; 1994 and 1995, yellow sticky traps); (h) Edmonton, 23 May to 25 July 1995 (using three trapping methods). Above each column is the number of trees sampled, followed by the total number of birch-leafmining sawflies collected (in parentheses).

on yellow sticky traps, and in Edmonton (14.6 f 4.1) than at George Lake (1.1 + 4.1) in 1993 on white sticky traps (Table 1). In addition, the ratio of the abundances of F. pusilla and P. thomsoni in Edmonton was similar to that at George Lake in both 1992 and 1993 (Fig. 2a; 'Location x Species' interaction terms in Table 1 were not statistically significant).

The three trapping methods employed concurrently during 1995 differed in their ability to capture birch-leafmining sawflies. Significantly more adult leafminers were trapped per tree using yellow sticky traps (32.6 f 4.1) than using white sticky traps (1.8 f 4.1) or emergence traps (0.8 f 4.1), and marginally more I? pusilla (13.5 f 3.4) were caught than P. thomsoni (10.0 f 3.4) (Fig. 2b; Table 1). Differences among trapping methods may be partially explained by the small captures in emergence traps and on white sticky traps during 1995 (see parenthetical values at the tops of columns in Fig. 2b). The ratios of the abundances of F. pusilla and P. thomsoni were marginally statistically different among the three trapping methods (Table 1, 'Trapping method x Species' interaction term), with P. thomsoni being

Volume 129 THECANAMAN ENmVOLOGlST 325

TABLE 1. TWO-factor analyses of variance evaluating differences in catches of adult Fenusa pusilla and Profenusa thomsoni at Edmonton and George Lake, Alberta, 1992-1995

Source df MS F P

1992, emergence traps (square-root-transformed)

Location 1 6.47 1.33 0.263 Species 1 103.25 21.12 0.0002 Location x Species 1 0.20 0.04 0.842 Error 20 4.89

1993, white sticky traps (log-transformed)

Location Species Location x Species Error

Edmonton only, 1994 and 1995, yellow sticky traps (square-root-transformed)

Year Species Year x Species Error

Edmonton only, 1995, all three trapping methods (log-transformed)

Trapping method Species Trapping method x Species Error

more prevalent on yellow sticky traps than on white sticky traps or in emergence traps (Fig. 2b). No adults of H. nernoratus, M. nana, or S. betuleti were trapped in Alberta in any year.

Sex ratios of F. pusilla varied depending on year, trapping method, and tree age. The proportion of females varied among years in Edmonton [I992 emergence traps, 0.41 f 0.06 (n = 2 trees); 1993 white sticky traps, 0.20 f 0.08 (n = 4 trees); 1994 yellow sticky traps, 0.51 f 0.03 (n = 5 trees); 1995 yellow sticky traps, 0.59 f 0.07 (n = 8 trees)]. However, these differences may be confounded with those resulting from the different trapping methods used. When the three methods were used concurrently during 1995, differences in the proportion of females among them were not statistically significant (ANOVA: MS = 0.23, F = 2.44, p = 0.12; emergence traps, 0.81 f 0.15; white sticky traps, 0.39 f 0.12; yellow sticky traps, 0.59 f 0.11). Also in 1995, asignificantly lowerproportion of females was caught on young trees at Sunstar Nurseries (0.28 f 0.02) than on mature trees near the university campus in Edmonton (0.57 f 0.08) (two-tailed two-sample t-test, p = 0.008, df = 12). This difference may have been exaggerated by differential attraction of males to yellow sticky traps, because the difference from laboratory rearings was not statistically significant, although a lower proportion of females still occurred on younger trees (0.37 f 0.12) than on mature trees (0.51 f 0.15) (two-tailed two-sample t-test, p = 0.501, df = 6). Sex ratios were not calculated at George Lake because only males were caught, probably as a result of the low capture rates (five individuals in each of 1992 and 1993).

326 THE CANAO~AS ENNJMOLOGIST MarchJApril 1997

TABLE 2. Two-factor analyses of variance evaluating differences in numbers of larvae per leaf between Edmonton and George Lake, Alberta, in 1992 and 1993. All data were transformed to Blom's normal rank scores for analysis

(SAS Institute Inc. 1988)

Source df MS F P

Fenusa pusilla

Location Year Location x Year Error

Location Year Location x Year Error

There were significantly more larval P. thomsoni per leaf in samples collected during 1992 (0.52 f 0.04) than during 1993 (0.13 + 0.04), and from Edmonton (0.38 f 0.04) than from George Lake (0.27 f 0.04) (Table 2). The mean number of larval F. pusilla per leaf was also greater in Edmonton (0.19 f 0.06) than at George Lake (0.10 f 0.06), but the difference was not statistically significant (Table 2). The actual differences in abundance of larval leafminers between locations were probably greater than indicated by these data because leaves containing larvae were actively sought, and these were much harder to find at George Lake than in Edmonton. Only six larval H. nemoratus were found at the two study locations between 1992 and 1995.

Seasonal Activities. First-generation adults of F. pusilla became active in mid- to late May in Edmonton (approximately 200 DD,,), and persisted until mid- to late June (approximately 600 DD,,) (Fig. 3). The few adults trapped at George Lake were captured from early June to early July.

The presence and size of subsequent generations of F. pusilla varied among years in Edmonton. In 1992, some trees had a small second generation in late July (Fig. 3). There was only one generation in 1993, but in both 1994 and 1995, a large second generation began in early July, and a smaller third generation occurred in mid-August (Fig. 3). More genera- tions occurred in years having higher degree-day accumulations in June and July: the highest was in 1995 (728 DD,,, three generations), followed by 1994 (71 1 DD,,, three generations), 1992 (706 DD,,, two generations), and 1993 (624 DD,,, one generation). There was no clear relationship between number of generations of F. pusilla and either the amount of precipi- tation from 1 April to 3 1 July (1992, 158.3 mm; 1993, 252.2 mm; 1994, 229.8 mm; 1995, 15 1.1 mm) or the number of days experiencing precipitation during these months (1992,44; 1993, 52; 1994, 49; 1995, 52). Differences among the three trapping methods may have affected the number of generations observed because emergence trap catches during June and July of 1995 reflected only one generation, whereas yellow sticky traps (Fig. 3) and white sticky traps revealed two generations.

Variation in the number of generations of F. pusilla was also observed among locations within years. In 1992, one generation was detected on mature trees at George Lake, two generations occurred on mature trees in Edmonton (Fig. 3), and three generations were incidentally noted on young trees at an Edmonton nursery.

Volume 129 THE CANADL4N ENTOMOLOGIS?

I , , , , , , , . . ; 6 9 2 6 \ 98&,?.,9 P 1 Q b +

&"J.'+d~3p.4,3p.2,&4Q,~2.~9B",bQ>Q' +d 4 .@ ,+, ,$ ,. ,.?'p" +9 8 Date

FIG. 3. Seasonal activities of adults of Fenusapusilla and Profenusa thomsoni during 1992-1995 in Edmonton, Alberta, as determined by emergence traps (1992), white sticky traps (1993), or yellow sticky traps (1994 and 1995).

Larval F pusilla were first seen in Edmonton in early June (approximately 350 DD,,) and persisted in mines until late July (approximately 1000 DD,,,) (Fig. 4) . Activity occurred slightly later at George Lake, coinciding with a later bud burst there. In 1992, a small second generation of larvae was evident in Edmonton, peaking on 7 August (Fig. 4). In 1993, a bimodal distribution of larvae occurred in Edmonton in the early summer (Fig. 4) , but it is unlikely that this represents two generations because only one generation of adults was observed (Fig. 3) and a relatively low number of degree-days accumulated during June and July of 1993 (see above).

Adults of P. thomsoni became active in Edmonton from early to late June (between approximately 400 and 550 DD,,), and persisted until early August (approximately 1000 DD,,) (Fig. 3). Unlike F. pusilla, P. thomsoni consistently had only one generation (Fig. 3). Larvae were first seen in Edmonton in the first week of July (approximately 700 DD,,) and most finished development by late August (approximately 1200 DD,,) (Fig. 4). Similar to F. pusilla, adult and larval activities commenced roughly 2 weeks later at George Lake (Fig. 4). Some late larvae were still feeding within yellowing leaves in late September, and probably died before completing their development.

Potential for Competition. There is little possibility of direct competition for leaf resources between F. pusilla and P. thomsoni; few leaves possessed mines of both species (expressed as a percentage of larva-bearing leaves; Edmonton: 1992, 0.23%; 1993, 1.28%; George Lake: 1992, 0.05%; 1993, 0.07%). However, there is potential for intraspecific

THE CAPIADIAF EWOMOLOGIST

MI F. pusllla, Edmonton 40, P. thomsoni, Edmonton

60 F. pusilla, George Lake 1 50 P. thomsoni, George Lake 1

Date

FIG. 4. Seasonal activities of larval Fenusa pusilla and Profenusa thomsoni during 1992 and 1993 at Edmonton and George Lake as determined by leaf sampling.

competition for both E pusilla and P. thomsoni. The mean amount of leaf area required per final-instar larva was significantly higher for P. thomsoni (276 f 13 mm2) than forE pusilla (146 f 13 mm2) (ANOVA on log-transformed data, MS[Species] = 2.15, F = 70.15, p = 0.001, df = 1,4). Using these estimates, 17.8% and 13.0% of the larval populations of E pusilla and P. thomsoni, respectively, had a below-average food supply available to them during 1993, and therefore may have experienced intraspecific competition. Although leaf areas were not measured during 1992, intraspecific competition among larval P. thomsoni was probably more severe than during 1993 because larvae were more abundant in 1992 (see above and Table 2).

Discussion

Species Abundances and Distributions. Profenusa thomsoni was the most abundant birch- leafmining sawfly in central Alberta during 1992- 1995 (Fig. 2), and it now appears to be the most abundant species in parts of eastem Canada (Hurley and Magasi 1995). However, P. thomsoni is rare in Europe (Eichom and Pschorn-Walcher 1973; Pschorn-Walcher and Altenhofer 1989; Schonrogge and Altenhofer 1992; Kenis et al. 1995) where it may be effectively controlled by parasitoids. Generalist chalcidoid parasitism rates of larval P. thom- soni in Europe can reach 27-46% (Schonrogge and Altenhofer 1992; Kenis et al. 1995).

Sex ratios observed for E pusilla in this and other studies (Friend 1933; Cheng and LeRoux 1965) suggest that 50% females is typical. However, our data from 1995 show that

Volume 129 THE CANADIAN ENWMOLOGIST 329

males may be more abundant on young trees having larger populations (mean catch on yellow sticky traps was 36 f 14 adult E pusilla per mature tree on the university campus, and 2290 f 689 adults per young tree at sunstar nurseries). Male-biased sex ratios are rare in sawflies, and their adaptive significance is unclear (Craig and Mopper 1993). Perhaps female E pusilla lay fewer fertilized (female) eggs when crowded, to reduce the population growth rate in subsequent generations.

It has been suggested that P. thomsoni is native to North America (Schonrogge and Altenhofer 1992) because a possible specialist parasitoid of P. thomsoni was reared in Ontario (Lathrolestes sp., Martin 1960) but not in Europe, where almost all other fenusine leafminers are parasitized by members of the ichneumonid genera Lathrolestes and Grypo- centrus (Pschom-Walcher and Altenhofer 1989). However, it seems more probable that P. thomsoni is introduced, because distribution patterns observed in northern ~ l b e r t a and the Northwest Territories (Fig. 1) suggest that birch-leafmining sawflies are still expanding their ranges in western Canada. In addition, leafmining sawfly damage was first noted on birch in eastern Canada after 1925 (Cheng and LeRoux 1965), and in Alberta only by the 1970s (Drouin and Wong 1984).

~eterarthrus-nemoratus was rare in Alberta during 1992- 1995, although Drouin and Wong (1984) found it to be common in Edmonton during 1978- 1979, second in abundance to E pusilla. Heterarthrus nemoratus can be common in eastern North America (Dowden 1941; Hall 1992), but is generally rare in Europe (Eichom and Pschorn-Walcher 1973). The decrease in populations of H. nemoratus in Alberta may have resulted from competition with P. thomsoni because both species consume mature leaves in late summer (Drouin and Wong 1984). Alternatively, populations may have been reduced by native parasitoids, such as the chalcidoid Chrysocharis nitetis (Walker), which efficiently exploited rare larval H. nemora- tus during 199 1 - 1995 (Digweed 1995).

The concurrent use of the three trapping methods during 1995 showed that they differ in ability to catch birch-leafmining sawflies,and therefore comparison of adult activity data collected using these different methods should be done cautiously. Emergence traps elimi- nated behavioural biases in catches, such as differential attraction of one sex or species to traps, but yellow sticky traps yielded higher numbers of adults, and are preferable when sampling small populations or for maximizing species representation. Yellow sticky traps are also less expensive and less likely to be vandalized than emergence traps, making them a better trap for monitoring urban birch-leafmining sawfly populations.

Seasonal Activities. Drouin and Wong (1984) did not observe more than two generations of E pusilla per year in Alberta, but our study establishes that three generations per year are possible, and shows significant among-year and among-tree variation in voltinism (Fig. 3). This variation probably resulted from a complex interaction between temperature and the prevalence of immature long-shoot leaves later in the summer. More new long-shoot leaves were casually observed in 1994 and 1995, both of which had three generations of F. pusilla and greater June-July degree-day accumulations, than in 1993, which had only one generation and lower June-July temperatures. In 1992, young nursery trees had a perpetual supply of new long-shoot leaves, and had three generations of E pusilla, whereas mature trees at George Lake lacked new leaves after ~ u n e and had only one generation. Chichester (1994) found that the number of adults of both E pusilla and E dohrnii (Tischbein), the European alder leafminer, emerging per generation was proportional to the quantity of acceptable foliage available. However, it is not clear how immature E pusilla assess the availability of later-occurring, immature long-shoot leaves. Perhaps faster larval develop- ment time or chemical cues from the host experienced while feeding prevent larvae from entering diapause, thus allowing them to emerge as adults later the same summer. Many sawflies have flexible voltinism patterns and diapause schedules which are affected by temperature, photoperiod, and host plant quality (GCri et al. 1993). However, these factors

330 THE C.4N.4Dl.4N ENTDMOLWIST MarchIApril 1997

differ in effect among sawfly and host plant species, and their interactions are generally poorly understood (GCri et al. 1993).

Urban-Rural Differences. Birch leafminers were generally more abundant in Edmonton than at George Lake. The reasons for this are not clear, but Olkowski et al. (1978) found that the cedar aphid (Cinara curvipes Patch) in California was more abundant in urban habitats, possibly as a result of different climatic conditions and a lack of parasitoids. Predators exploited high-density patches of P. thomsoni more efficiently at George Lake than in Edmonton during experiments conducted in 1994 (Digweed 1995), suggesting that differ- ences in the impact of natural enemies may have contributed to urban-rural differences in birch-leafmining sawfly abundance.

Another reason for lower abundance of leafminers at George Lake may have been the higher abundance of the native birch skeletonizer, Bucculatrix canadensisella Chambers, at George Lake during 1992 and 1993 (Digweed 1995). High densities of B. canadensisella result in severe birch defoliation (Friend 1927), which can decrease the quality of birch leaves (Niemela and Tuomi 1993), and may negatively affect populations of both E pusilla and P. thomsoni. In addition, larval P. thomsoni may compete directly with those of B. canaden- sisella for leaf resources because the two species feed on birch leaves contemporaneously (Friend 1927; Blais and Pilon 1968; Digweed 1995).

Activity at George Lake consistently occurred later than in Edmonton, and this delay was not resolved by plotting the data in degree-days. However, our degree-days may have been poor estimates of conditions experienced by the sawflies because they were calculated using air (rather than soil) temperatures measured at sites remote from sampled trees (i.e. the Edmonton Municipal Airport and the town of Sion). This is supported by our observation that soil in the forest at George Lake was sometimes still partially frozen in early May, but was thawed at more open sites in Edmonton. It is likely that host cues used by leafminers to initiate emergence occurred later at George Lake because bud burst always occurred later there than in Edmonton.

Potential for Competition. There is little opportunity for direct interspecific competition between F. pusilla and P. thomsoni because their leafmining activities are quite separate spatially and temporally. This may be an artifact retained from the evolution of birch- leafmining sawflies in Europe (DeClerck and Shorthouse 1985). The genus Betula has 50 leafmining insect species associated with it in Britain, the most of any native tree genus there (Godfray 1984). Members of such a large guild should display a high degree of niche partitioning, such that interspecific competition is minimized (e.g. Pleasants 1980). The six species of birch-leafmining sawflies native to Europe occupy quite different spatio-temporal niches (Eichorn and Pschorn-Walcher 1973), and similar niche partitioning is evident in lepidopteran leafminers on birch in Europe (Stigmella spp.; see Boomsma et al. 1987). In contrast, North American birch have a much smaller leafminer fauna (16 species recorded in Ontario; Lindquist and Thomson 1970), so introduced populations of F. pusilla and P. thomsoni were probably released from strict niche constraints experienced in Europe. With more time, introduced birch-leafmining sawflies may evolve expanded niches. For example, P. thomsoni may become active earlier in the season, which could eventually permit multivoltinism.

Intraspecific competition probably occurs for both F pusilla and P. thomsoni, and at lower larval densities per leaf for P. thomsoni than for F. pusilla. The value we calculated for the amount of leaf area used by F. pusilla falls midway between that of Jones and Raske (1976) (mean: 90 mm2, n = 0) and Friend (1933) (mean + SE on unsaturated leaves: 194.8 f 17.4 mm2, n = 17). However, Martin (1960) found that larval P. thomsoni required approximately 530 mm2 to complete development, significantly more than was required in Alberta. His methods are not explicitly stated, so it is difficult to determine reasons for this

Volume 129 THE CANADIAN ENTOMOLOGIST 3,3 1

discrepancy. It is possible that differences in leaf thickness account for some of the differences among estimates.

Conclusions. This study shows that three species of introduced birch-leafmining sawflies are established in central Alberta. The multivoltine F. pusilla occurred earlier and was generally less abundant than the univoltine P. thomsoni. However, both species were probably abundant enough to experience intraspecific competition. Information we have obtained about relative abundance, seasonal activity, and trapping methods provides essen- tial background information for further experiments investigating the ecological relation- ships of these sawflies, and should aid future monitoring and management programs in central Alberta.

Acknowledgments We thank R. McQueen, M. Braun, C. Currie, H. Ckcamo, C. Klingenberg, A. Meyer,

and several summer students for technical support, advice, and much stimulating discussion. C. Saunders of the City of Edmonton Pest Management Services and several Edmonton homeowners are thanked for allowing us to study trees in their care. G. Smith and H. Cerezke shared useful information about distributions of birch-leafmining sawflies. Comments of anonymous reviewers improved this paper. This study was primarily funded through an NSERC Operating Grant to JRS and a Canadian Forest Service (Green Plan) grant to DWL. SCD was supported by NSERC (Postgraduate Scholarship), the Canadian Forest Service (Graduate Supplement), the Entomological Society of Canada (Postgraduate Scholarship), and the University of Alberta (Walter H. Johns Graduate Fellowship).

References Allen, J.C. 1976. A modified sine wave method for calculating degree-days. Environmental Entomology 5:

388-396. Blais, J.R., and J.G. Pilon. 1968. Influence of temperature and moisture on the survival of cocoons, and on adult

emergence of Bucculatrix canadensisella. The Canadian Entomologist 100: 742-749. Boomsma, J.J., H. Timmermans, C.P.M. Corvers, and J. Kabout. 1987. Monophagous leaf-mining larvae of

Stigmella (Lepidoptera: Nepticulidae) on birch: Patterns and differentiation in exploitation of the host. Holarctic Ecology 10: 206-218.

Britton, W.E. 1924. A European leaf-miner of birch. Journal of Economic Entomology 17: 601. 1925. A European sawfly leaf-miner of birch. Connecticut Agricultural Experiment Station Bulletin 265:

340-341. Cerezke, H.F., and J.P. Brandt. 1993. Forest Insect and Disease Conditions in Alberta, Saskatchewan, Manitoba,

and the Northwest Territories in 1992. Forestry Canada, Northwest Region, Northern Forestry Centre, Edmonton, Alberta, Information Report NOR-X-332: 23 pp.

Cheng, H.H., and E.J. LeRoux. 1965. Life history and habits of the birch leaf miner, Fenusa pusilla (Lepeletier) (Hymenoptera: Tenthredinidae), on blue birch, Betula caerulea grandis Blanchard, Morgan Arboretum, Qutbec, 1964. Annals of the Entomological Society of Qukbec 10: 173- 188.

Chichester, J.L. 1994. Influence of Host Phenology on Voltinism Patterns of Alder, Birch, andHawthorn Leafminers (Hymenoptera: Tenthredinidae), with Evidence for Bet-hedging Strategies in Birch and Alder Leafminer. M.Sc. thesis, Department of Entomology, Michigan State University, East Lansing, MI. 121 pp.

Craig, T.P., and S. Mopper. 1993. Sex ratio variation in sawflies. pp. 61 -92 in Wagner, M.R., and K.F. Raffa (Eds.), Sawfly Life History Adaptations to Woody Plants. Academic Press, San Diego, CA.

DeClerck, R.A., and J.D. Shorthouse. 1985. Tissue preference and damage by Fenusa pusilla and Messa nana (Hymenoptera: Tenthredinidae), leaf-mining sawflies on white birch (Betula papyrifera). The Canadian Entomologist 117: 35 1 -362.

Digweed, S.C. 1995. Effects of Natural Enemies, Competition, and Host Plant Quality on Introduced Birch Leafminers (Hymenoptera: Tenthredinidae). Unpublished M.Sc. thesis, University of Alberta, Edmonton, AB. 144 pp.

Dowden, P.B. 1941. Parasites of the Birch Leaf-mining Sawfly (Phyllotoma nemorata). USDA Technical Bulletin 757: 55 pp.

Drouin, and H.R. Wong. 1984. Birch Leaf-mining Sawflies in Alberta (Hymenoptera: Tenthredinidae). Environment Canada, Canadian Forest Sewice, Northern Forestry Research Centre, Edmonton, Alberta, Information Report NOR-X-260: vi + 13 pp.

332 THE CANADIAN ENTOMOLOGIST March/April 1997

Eichom, O., and H. Pschorn-Walcher. 1973. The parasites of the birch leaf-mining sawfly (Fenusa pusilla [Lep.], Hymenoptera: Tenthredinidae) in central Europe. Technical Bulletin of the Commonwealth Institute of Biological Control 16: 79-104.

Environment Canada. 1992-1995. Monthly Meteorological Summary, Edmonton Municipal Airport. Environment Canada, Atmospheric Environment Service, Edmonton, Alberta. April to September, 1992-1995.

Faeth, S.H. 1985. Host selection by leaf miners: Interactions among three trophic levels. Ecology 66: 870-875. - 1991. Novel aspects of host tree resistance to leafminers. pp. 219-239 in Baranchikov, Y.N., W.J. Mattson,

F.P. Hain, and T.L. Payne (Eds.), Forest Insect Guilds: Patterns of Interaction with Host Trees. USDA Forest Service General Technical Report NE-153.

Faeth, S.H., andD. Simberloff. 1981. Population regulation of aleaf-mining insect, Cameraria sp. nov., at increased field densities. Ecology 62: 620-624.

Friend, R.B. 1927. The biology of the birch leaf skeletonizer, Bncculatrix canadensisella Chambers. Connecticut Agricultural Experiment Station Bulletin 288: 395 -486. - 1933. The birch leaf-mining sawfly Fenusa pumila Klug. Connecticut Agricultural Experiment Station

Bulletin 348: 293-364, Gtri, C., J.-P. Allais, and M.-A. Auger. 1993. Effects of plant chemistry and phenology on sawfly behavior and

development. pp. 173-210 in Wagner, M.R., and K.F. Raffa (Eds.), Sawfly Life History Adaptations to Woody Plants. Academic Press, San Diego, CA.

Godfray, H.C.J. 1984. Patterns in the distribution of leaf-miners on British trees. Ecological Entomology 9: 163-168.

Goulet, H. 1992. The Insects and Arachnids of Canada, Part 20. The Genera and Subgenera of the Sawflies of Canada and Alaska: Hymenoptera: Symphyta. Research Branch, Agriculture Canada, Publication 1876: 235 pp.

Haack, R.A., and W.J. Mattson. 1993. Life history patterns of North American tree-feeding sawflies. pp. 503-545 in Wagner, M.R., and K.F. Raffa (Eds.), Sawfly Life History Adaptations to Woody Plants. Academic Press, San Diego, CA.

Hall, J.P. 1992. Forest Insect and Disease Conditions in Canada, 1992. Natural Resources Canada, Canadian Forest Service. 120 pp.

Hurley, J.E., and L.P. Magasi. 1995. Forest Pest Conditions in the Maritimes in 1994. Canadian Forest Service, Maritimes Region, Information Report M-X-194E: 55 pp.

Jones, J.M., and A.G. Raske. 1976. Notes on the biology of the birch leafminer, Fenusa pusilla (Lep.), in Newfoundland (Hymenoptera: Tenthredinidae). Phytoprotection 57: 69-76.

Kalcounis, M.C., R.D. Csada, and R.M. Brigham. 1992. Axle grease as an alternative adhesive for use on sticky traps. The Canadian Entomologist 124: 561 -562.

Kenis, M., D. You, and A. Van Averbeke. 1995. Ambermarked birch leaf-miner (Profe~~usathomsoni [Konow]). pp. 27-31 in Kenis, M., K. Carl, A. Van Averbeke, N. Otten, D. You, and E. Altenhofer (Eds.), International Institute of Biological Control, European Station, Annual Report 1995 for the Canadian Forest Service. International Institute of Biological Control, Silwood Park, Berkshire, UK.

Lindquist, O.H., and M.J. Thomson. 1970. The biology of a birch leaf miner, Messa nana (Hymenoptera: Tenthredinidae), new to Canada. The Canadian Entomologist 102: 108-1 1 1.

Lysyk, T.J. 1989. Stochastic model of eastern spruce budworm (Lepidoptera: Tortricidae) phenology on white spruce and balsam fir. Journal of Economic Entomology 82: 1161-1 168.

Martin, J.E.H. 1977. The Insects and Arachnids of Canada. Part 1. Collecting, Preparing, and Preserving Insects, Mites, and Spiders. Research Branch, Agriculture Canada, Publication 1643: 182 pp.

Martin, J.L. 1960. The bionomics of Profenusa thonzsoni (Konow) (Hymenoptera: Tenthredinidae) a leaf-mining sawfly on Betula spp. The Canadian Entomologist 92: 376-384.

Martineau, R. 1984. Insects H m f u l to Forest Trees. Multiscience Publications Ltd., Montrtal, PQ. 261 pp. Niemela, P., and J. Tuomi. 1993. Sawflies and inducible resistance of woody plants. pp. 391-407 in Wagner, M.R.,

and K.F. Raffa (Eds.), Sawfly Life History Adaptations to Woody Plants. Academic Press, San Diego, CA. Nystrom, K.L., and H.J. Evans. 1989. Biological notes on a birch leaf edgeminer, Scolione~ra betuleti

(Hymenoptera: Tenthredinidae), new to North America. Proceedings of the Entomological Society of Ontario 120: 17-24.

Olkowski, W., H. Olkowski, A.I. Kaplan, and R. van den Bosch. 1978. The potential for biological control in urban areas: Shade tree insect pests. pp. 31 1- 347 in Frankie, G.W., and C.S. Koehler (Eds.), Perspectives in Urban Entomology. Academic Press, New York, NY.

Pleasants, J.M. 1980. Competition for bumblebee pollinators in Rocky Mountain plant communities. Ecology 61: 1446-1459.

Pruess, K.P. 1983. Day-degree methods for pest management. Environmental Entomology 12: 613-619. Pschorn-Walcher, H., and E. Altenhofer. 1989. The parasitoid community of leaf-mining sawflies (Fenusini and

Heterarthrini): A comparative analysis. Zoologischer Anzeiger 222: 37-56. Qnednau, F.W. 1984. Fenusa pusilla (Lepeletier), Birch Leafminer (Hymenoptera: Tenthredinidae). pp. 291-294

in Kelleher, J.S., and M.A. Hulme (Eds.), Biological Control Programmes Against Insects and Weeds in Canada 1969-1980. CAB International, Wallingford, UK.

Volume 129 THE CANADIAN ENTOMOLOGIST 333

SAS Institute Inc. 1988. SASISTAT User's Guide, Release 6.03 Edition. SAS Institute Inc., Cary, NC. 1028 pp. Schonrogge, K., and E. Altenhofer. 1992. On the biology and larval parasitoids of the leaf-mining sawflies

Profenusa thomsoni (Konow) and P. pygmaea (Konow) (Hymenoptera: Tenthredinidae). Entomologist's Monthly Magazine 128: 99-108.

Smith, D.R. 197 1. Nearctic Sawflies. 111. Heterarthrinae: Adults and Larvae (Hymenoptera: Tenthredinidae). USDA Technical Bulletin 1420: 84 pp.

Togashi, I. 1981. The Japanese sawflies of the genus Profenusa (Hymenoptera: Tenthredinidae), with description of three new species. KontyLi, T o b o 49: 408-413.

Volney, W.J.A., and H.F. Cerezke. 1992. The phenology of white spruce and the spruce budworm in northern Alberta. Canadian Journal of Forest Research 22: 198-205.

Wilkinson, K. 1990. Trees and Shrubs of Alberta. Lone Pine Publishing, Edmonton, AB. 191 pp. Zar, J.H. 1984. Biostatistical Analysis, 2nd ed. Prentice-Hall, Englewood Cliffs, NJ. xiv + 718 pp.

(Date received: 13 June 1996; date accepted: 15 November 1996)