Decline of Soybean Aphid (Homoptera: Aphididae) Egg Populations ...

POPULATION ECOLOGY

Decline of Soybean Aphid (Homoptera: Aphididae) Egg Populations from Autumn to Spring on the Primary Host, Rhamnus cathartica J. A. WELSMAN,1 C. A. BAHLAI,1 M. K. SEARS,1

AND

A. W. SCHAAFSMA2

Environ. Entomol. 36(3): 541Ð548 (2007)

ABSTRACT Soybean aphid, Aphis glycines Matsumura (Homoptera: Aphididae), is a severe pest of soybeans in North America. Soybean aphid populations cycle between a secondary summer host, where populations reproduce parthenogenetically and a primary host, where populations overwinter as eggs. In North America, the secondary host is soybean, and the primary hosts are Rhamnus cathartica L. (Rhamnaceae) and R. alnifolia LÕHer. A location with abundant populations of soybean aphid on R. cathartica was identiÞed near Guelph, Ontario, Canada, in October 2004, and eggs on trees were counted at multiple sites within that location each autumn and spring over the next 2 yr. Dynamics of naturally occurring soybean aphid populations on the primary host were assessed with respect to (1) decline of overwintering eggs from autumn to spring, (2) development of spring populations on R. cathartica, and (3) development of soybean aphid populations on soybean immediately adjacent to overwintering sites. Counts of aphid eggs declined by ⬇70% between autumn and spring sampling periods in 2004 Ð2005. SigniÞcant differences in counts of aphid eggs relative to sampling height were observed in the canopy of R. cathartica. No edge effects were observed in the development of soybean aphid populations in soybeans adjacent to overwintering sites in this study. Very few eggs were collected at the same study location in the autumn of 2005, and no aphid eggs were collected from samples taken in the spring of 2006. Egg counts taken in the autumn of 2006 were intermediate in number relative to counts taken in the autumn of 2004 and 2005. KEY WORDS Aphis glycines, buckthorn, overwintering

Soybean aphid, Aphis glycines Matsumura (Homoptera: Aphididae), colonizes soybean, Glycine max L. (Merr), and has been the cause of considerable economic losses since its introduction to North America at the end of the 20th century (Ragsdale et al. 2004). Soybean aphid populations cycle between a secondary summer host, where populations reproduce parthenogenetically, and a primary host, where populations overwinter as eggs (Ragsdale et al. 2004, Wu et al. 2004). In North America, the secondary host is soybean and the primary hosts are Rhamnus cathartica L. and R. alnifolia LÕHer (Voegtlin et al. 2004b). Gynoparae and oviparae have been observed on Frangula alnus P. Mill, although overwintering on that species has not been conÞrmed (Voegtlin et al. 2004b). R. cathartica is most abundant north of the 41st parallel (Ragsdale et al. 2004). Decreasing temperature and declining plant quality are thought to initiate migration of gynoparous females from soybean to the overwintering host (Wu et al. 2004). On the overwintering host, alate males originating from soybean mate with oviparous females produced by gynoparae (Wu et al. 2004). Eggs are 1 Department of Environmental Biology, University of Guelph, Guelph, Ontario, Canada. 2 Corresponding author: Department of Plant Agriculture, Ridgetown Campus, University of Guelph, Ontario, Canada (e-mail: [email protected]).

deposited on the margins of buds in late October to mid-November in the midwestern United States, around the time of leaf drop (Ragsdale et al. 2004). Aphid eggs are yellow-green on oviposition, but later become shiny and black (Leather 1980, 1981, 1990, Wade and Leather 2002). The Þrst generation of aphids after eclosion consists of a stem mother, an apterous fundatrix (McCornack et al. 2005). Three or more generations of soybean aphids are produced by parthenogenesis on the primary host (Liu et al. 2004). On or after the third generation, primary alates are produced that migrate to the secondary host (Ragsdale et al. 2004). The most common natural enemies surveyed on soybean include Harmonia axyridis Pallas, Orius insidiosus (Say), and Leucopis spp. (Fox et al. 2004). Parasitism of soybean aphid by Braconidae and Aphelinidae on soybean has been observed in midwestern United States, although parasitoids seem to be less common than predators (Brewer and Noma 2004, Ellingson 2004). Little is known regarding the diversity of natural enemies that interact with soybean aphid populations on the primary host(s). In autumn 2004, a location outside of Guelph, Ontario, Canada, was found with large populations of soybean aphid gynoparae, androparae, and oviparae on R. cathartica. Abundant egg populations on these trees offered a rare opportunity to study overwinter-

0046-225X/07/0541Ð0548$04.00/0 䉷 2007 Entomological Society of America

542

ENVIRONMENTAL ENTOMOLOGY

Vol. 36, no. 3

ing in southwestern Ontario, a region where such large egg populations had not been observed previously (T. Baute, personal communication). The following hypotheses were tested: (1) aphid egg densities on R. cathartica in spring are signiÞcantly lower than autumn densities; (2) the quantity of eggs collected at various heights in the tree canopy will differ; (3) of natural mortality factors co-occurring with A. glycines populations on the primary host in spring, predaceous arthropods are more common than parasitic forms; (4) the development of A. glycines populations in soybean Þelds adjacent to overwintering sites will not display an “edge effect”; (5) signiÞcant yearly variability in autumn egg counts will be observed at the study location. Materials and Methods A location with populations of naturally occurring soybean aphids on R. cathartica was identiÞed near Guelph, Ontario, Canada, in October 2004 (GPS: N 43.59, W 80.27; Fig. 1). The location measured ⬇740 by 430 m and included three agricultural Þelds planted to soybean in 2004 and 2005, as well as meadow/pasture undergoing secondary succession and a pine plantation (Fig. 1). Winter wheat was planted in the autumn of 2005. For purposes of sampling, the location was divided into southern and northern halves with a stream serving as a boundary. Egg counts were made over 2 yr (15 November 2004; 23 March 2005; 19 October 2005; 17 November 2005; 15 March 2006, and 23 November 2006). Four transects were established running perpendicular to the stream. Forty sample sites were identiÞed (10 per transect) in the autumn of 2004, and egg density was assessed at each site by destructive sampling. For this purpose, 10 branch cuttings ⬇10 cm in length were taken at heights ⬇0.5, 1, and 2 m above ground level. Branch cuttings were bagged, and egg numbers were counted in the laboratory under a dissecting microscope. All egg counts subsequent to 23 March 2005 involved the collection of Þve branch cuttings from the three previously sampled heights at each site. Five of the sampling sites were omitted from the 2006 spring survey because of inaccessibility (ßooding) or poor condition of the plant material (i.e., dead or dying trees). In 2004 Ð2005, data loggers (Onset Computer, Bourne, MA) were maintained at heights of 1.5Ð2 m in the canopy of R. cathartica to monitor hourly mean temperatures at sites distributed across the study location (Fig. 1). Variability in mounting heights was related to availability of plant material capable of supporting data loggers. Additional data loggers were also maintained at approximate heights of 0.5 and 2.0 m (Fig. 1). To assess impacts of natural mortality factors such as predators and parasitoids on spring aphid populations, exclusion barriers (ÔNo-See-UmÕ mesh; KaplonSimon Venture Textiles Division, Braintree, MA) were placed over some aphid colonies beginning 3 May 2005. Cylindrical No-see-um mesh enclosures were 35

Fig. 1. A study location with populations of A. glycines (Homoptera: Aphididae) on R. cathartica (Rhamnales: Rhamnaceae) was identiÞed near Guelph, Ontario, Canada, in October 2004 (GPS: N 43.59, W 80.27). Experiments took place over 2 yr, beginning 15 November 2004 to July 2005 and 19 October 2005 to 23 November 2006. Forty sampling sites were distributed across four transects. A river served as a boundary between southern and northern sampling groups.

cm in length and 12 cm in diameter, with a central supportive plastic loop sewn into the fabric. Enclosures were fastened at each end with adjustable plastic cable-ties, forming a self-supporting, protective structure. Three colonies were observed (assigned to treatments AÐC) weekly at each of at least nine of the sites previously sampled for eggs. Colony A was counted and immediately enclosed in mesh, and its population was assessed by destructive sampling 1 wk later. Colony B was counted on week 1, counted and enclosed in mesh on week 2, and harvested to be placed in a growth cabinet on week 3 at L:14 D:10, 23 ⫾ 1⬚C. Parasitism that occurred in colony B during the Þrst week should have been evident during the 2 subsequent wk in which the colony was protected by a mesh enclosure. Colony C was left as an unprotected control counted on week 1 and sampled on week 2. Coccinellids observed foraging within 5 m of any colony were noted at each site for each observation date. In 2005, when soybeans were planted on the three Þelds across the study location, aphid populations were assessed at various distances into each soybean Þeld beginning at a point on a tree line containing R. cathartica infested with A. glycines, to determine dispersal patterns into soybean from adjacent stands of buckthorn. Three transects were established running 50 m into a soybean Þeld from a single point on the Þeld edge (Fig. 1). At intervals of 5 m along each transect, a single plant was rated for A. glycines density after a rating scale by Difonzo (2002). Aphid numbers were estimated on the stem, the middle leaßet of the top uppermost fully unfolded trifoliate, the middle leaßet of a trifoliate half-way up the plant, and the

June 2007

WELSMAN ET AL.: DECLINE OF SOYBEAN APHID EGG POPULATIONS FROM AUTUMN TO SPRING

middle leaßet of a trifoliate at the bottom of the plant, and a rating of 0 Ð 4 was assigned (0 ⫽ no aphids, 1 ⫽ 1Ð10 aphids, 2 ⫽ 11Ð25 aphids, 3 ⫽ 26 Ð99 aphids and 4 ⫽ 100⫹ aphids per leaßet or stem). The four ratings (stem rating and trifoliate ratings at three heights) were summed to obtain a total score out of 16. Assessments were repeated every 3Ð 4 d beginning 13 June 2005 and ending on 25 July 2005. Because R. cathartica is also a primary host for Aphis nasturtii Kaltenbach (Homoptera: Aphididae), specimens were collected and conÞrmed as A. glycines using published keys and with the assistance of systematists (Dr. R. Foottit and E. Maw) at the Canadian National Collection (CNC), at the Eastern Cereal and Oilseed Research Centre (ECORC), Agriculture and Agri-Food Canada, Ottawa, Ontario, Canada (Foottit and Richards 1993, Blackman and Eastop 2000, Voegtlin et al. 2004a). Voucher specimens are housed at the Insect Collection, located in the Department of Environmental Biology at the University of Guelph, Ontario, Canada. Statistical Analyses. All statistical analyses were conducted with SAS (SAS Institute 2003). When the assumptions of analysis of variance (ANOVA) could not be met, a nonparametric procedure, the WilcoxonMann-Whitney test, was used to make comparisons among treatment groups. Means were computed in Proc NPAR1WAY with the ANOVA option (SAS Institute 2003). Analysis of Egg Densities. Percentage change in egg counts per branch segment was calculated as percentage change ⫽ [(x2 ⫺ x1)/x1] ⫻ 100, where x1 ⫽ Þrst observation and x2 ⫽ second observation. The Wilcoxon-Mann-Whitney test was used to compare egg densities at the sampled heights in the autumn of 2004, the spring of 2005, and the autumn of 2006. Comparisons of total egg counts collected in the autumns of 2004, 2005, and 2006 were conducted using the Wilcoxon-Mann-Whitney test. Egg densities were also compared relative to south/north distribution at the study location. Analysis of Temperature Data. A repeated-measures analysis using Proc Mixed was used to test the hypothesis that the mean, maximum, minimum, and range of daily temperatures were different at the two logger heights. In the Proc Mixed model, logger height, measurement date, and the interaction of logger height and date were Þxed effects. Site effects were random. The model was assessed with several covariance structures; the covariance structure producing the lowest AkaikeÕs info criterion and SchwarzÕs Bayesian criterion was selected for use in the analysis. A TukeyÕs test was conducted to compare means when a signiÞcant effect for logger height was found. Population Growth on the Primary and Secondary Host. Weekly and overall percentage population growth rates between protected and unprotected aphid colonies on R. cathartica were compared using the Wilcoxon-Mann-Whitney test. Mean colony sizes observed at the start of each weekly caging interval were compared using the Wilcoxon-Mann-Whitney test.

543

Fig. 2. Aphid egg counts on R. cathartica sampled at three heights above ground level on 15 November 2004 and 23 March 2005 at a study location near Guelph, Ontario, Canada. *SigniÞcant declines in egg counts taken at the same sampling height on different dates (Wilcoxon-Mann-Whitney test, P ⱕ 0.05).

A repeated-measures analysis using Proc Mixed was used to test the hypothesis that soybean aphid populations in soybean Þelds immediately adjacent to R. cathartica did not display an edge effect. Tests of residuals on transformed data [y ⫽ ln(1/(x ⫹ 0.5)] were conducted to test the assumptions of ANOVA. A Shapiro-Wilkes test was conducted using Proc Univariate on the calculated studentized residuals of experimental data to test normality of the errors. Proc Plot was used to plot residual against predicted values for the response variable (aphid rating) and to plot residuals by date, site, and sampling distance. In the Proc Mixed model, distance from the transect origin (0, 5, 10. . . 50 m), sampling date, and the interaction of distance and date were Þxed effects. Site effects were random. The model was assessed with several covariance structures; the covariance structure producing the lowest AkaikeÕs info criterion and SchwarzÕs Bayesian criterion was selected for use in the analysis. A TukeyÕs multiple means comparison test was conducted to compare means when there were signiÞcant Þxed effects. A type I error rate (␣) of 0.05 was selected to make decisions of signiÞcance in all statistical tests. Results Decline in Egg Counts. Mean egg counts per branch segment declined by 70% between the autumn (4.7 eggs per branch segment on 15 November 2004) and spring (1.4 eggs per branch segment on 23 March 2005) sampling dates. Between the autumn and spring sampling periods of 2004 and 2005, egg densities declined by ⬇83% at 2 m, 49% at 1 m, and 64% at 0.5 m (Fig. 2). On 15 November 2004, egg densities at 2 m were greater than at 0.5 m on branch segments collected from the northern half of the study location

544


Vol. 36, no. 3

Fig. 3. Aphid egg counts on R. cathartica sampled at three heights above ground level on 15 November 2004 and 23 March 2005 at a study location near Guelph, Ontario, Canada. Egg counts of a different letter taken at different sampling heights in the same sample grouping (samples from northern or southern halves of the study location) are signiÞcantly different (Wilcoxon-Mann-Whitney test, P ⱕ 0.05).

(Fig. 3). On 23 March 2005, egg densities at 1 m were greater than densities at 0.5 m on branch segments collected from the northern half of the study location (Fig. 3). On 23 November 2006, no differences in egg counts were observed relative to sampling height or source (i.e., southern/northern half of study location). Continuous temperature data sets for the period 18 November 2004 to 23 March 2005 were recovered from six sites with loggers at heights of 1.5Ð2.0 m. Continuous data sets could not be recovered from the remaining sites because of vandalism, battery failure, or incomplete data sets. Average temperatures recorded by Hobo data loggers for the months beginning 18 November and ending 23 March 2005 were 2.5 (November), ⫺4.7 (December), ⫺8.6 (January), ⫺6.0 (February), and Ð5.8⬚C (March). The lowest temperature of the winter was recorded in January (Ð33.3⬚C). Complete data sets from loggers recording hourly temperatures at two heights (0.5 and 2.0 m) from 5 January to 3 March 2005 were recovered from three sites (sites: North, B1; North, B2; South, A2; Fig. 1). Mean daily temperatures and minimum daily temperatures at 0.5 and 2.0 m differed signiÞcantly during the interval 26 January to 3 March 2005, although the difference in temperatures at both heights for each measure was ⬍1⬚C. On 15 November 2004, 5,585 eggs were collected, and on 23 March 2005, 1,686 eggs were collected. On 19 October 2005, 13 eggs were observed, and 4 eggs were collected on 17 November 2005. No eggs were collected in samples taken the following spring (15 March 2006). A total of 250 eggs were collected on 23 November 2006. The total number of eggs collected in each year of the study (2004, 2005, and 2006) differed signiÞcantly. Population Development on the Primary Host in Spring. Aphid colonies observed on 3 May 2005 were signiÞcantly larger than those observed in the subse-

quent 2 wk (beginning 10 and 18 May; Fig. 4). By the last 2 wk of sampling (beginning 24 and 30 May), mean aphid colony size was no greater than that observed on 3 May (Fig. 4). When analyzed with a Wilcoxon two-sample test, no differences in weekly population growth rates were observed among unprotected and enclosed aphid colonies for the Þrst 4 wk of exclusion treatments (03, 10, 18, and 24 May 2005). In the last week of sampling (30 May to 8 June 2005), unprotected colonies decreased at a greater rate compared with enclosed colonies. Mean weekly percentage change in enclosed populations overall was 136.7%, which differed signiÞcantly from that of unprotected populations, in which negative growth overall was exhibited (⫺3.67%). Natural Enemy Surveys. No adult or larval parasitoids were collected during the study. Three species of Coccinellidae were observed foraging in the vicin-

Fig. 4. Mean size of aphid colonies (weekly n ⱖ 18 distributed across ⱖ9 weekly sites within the study location) observed on R. cathartica L. during May, 2005 at a location near Guelph, Ontario, Canada. Observation dates of a different letter are signiÞcantly different (Wilcoxon-MannWhitney test, P ⱕ 0.05).

June 2007


545

Fig. 5. Aphis glycines (Homoptera: Aphididae) populations in soybean assessed by rating (0Ð16) at various distances from R. cathartica stands in Þelds (n ⫽ 3), at a study location near Guelph, Ontario, Canada. A. glycines populations were not signiÞcantly different at any distance from the Þeld edge for any sampled date (13 June to 25 July 2005).

ity of A. glycines colonies: Coccinella septempunctata L. (3 observations), Harmonia axyridis Pallas (12 observations), and Propylaea quatuordecimpunctata L. (5 observations). The date of the Þrst observation for each species was 3, 10, and 18 May 2005, respectively. Development of A. glycines Populations in Soybean Fields Adjacent to Overwintering Sites. Aphis glycines populations on soybean Þelds at the study location increased throughout the observation period beginning 13 June 2005 and ending 25 July 2005. Aphid populations along transects did not differ at any distance from the Þeld edge (Fig. 5). Discussion Oviposition and Egg Survival. In this study, a 70% decline in aphid eggs was observed between the autumn and spring sampling periods in 2004 Ð2005. This level of mortality among overwintering eggs is comparable with that observed for other aphid species. Gange and Llewellyn (1988) have reported total winter egg mortality of 65% for the alder aphid, Pterocallis alni De Geer, and overwintering mortality among eggs of Rhopalosiphum padi L. range from 70 to 81% (Leather 1980, 1981). Reports of mortality for overwintering eggs of Acyrthosiphon pisum (Harris) range from 70 (Bronson 1935) to 83% (Dunn and Wright 1955). Mortality factors responsible for decline in egg abundance could include predation and exposure to winter temperatures. In a study in which R. padi L. eggs were protected against predation and counted weekly throughout winter and early hatch, total mortality in the protected group was 35% compared with 81% mortality in unprotected populations (Leather 1981). Dunn and Wright (1955) have suggested that rain could also contribute to mortality by dislodging eggs and promoting fungal infection. Determining the relative contribution of rain and natural enemies to total mortality among A. glycines eggs will require that exclusion experiments similar to those by Leather (1981) be conducted.

January 2005 had the lowest average daily temperatures (⫺8.6⬚C), as well as the lowest recorded temperature (⫺33.3⬚C) of that winter. The supercooling point (SCP) for A. glycines eggs, the temperature below which ice crystals form in tissues and instantaneous death occurs, is approximately ⫺34⬚C (Salt 1961, Somme 1982, McCornack et al. 2005). A proportion of the eggs in this study may have been exposed to temperatures approaching the SCP for A. glycines. Although statistically signiÞcant differences in mean and minimum daily temperatures recorded by loggers at 0.5 and 2.0 m were found, it is unlikely that such differences are biologically signiÞcant, because both measures differed ⬍1⬚C at each height. Data loggers were situated such that measurements of temperatures in the air spaces within the tree canopy were recorded. There is a possibility that temperatures experienced at the branch surface or budÐ branch interface are different from temperatures experienced in the air spaces between branches. Such “microclimate” effects could render certain areas within the canopy more conducive to survival of aphid eggs. Whether such microclimate effects vary throughout the tree canopy will require more exact measurements of temperature at the surface of the branch and/or bud. Variation in Egg Counts at Sampled Heights. Egg densities at 2 m on 15 November 2004 were greater than those sampled at 0.5 m in the northern half of the study location. This may reßect the pattern of aphid immigration into the R. cathartica canopy in autumn by alates. The southern and northern halves of the study location differed with respect to the quantity of sampling sites located on a Þeld edge (Þve sites on Þeld edges in northern half of study location and 15 in southern half; Fig. 1). Alates immigrating into trees located on a Þeld edge could have greater access to the entire tree canopy, whereas alates immigrating into areas of denser vegetation might initially land in the upper canopy. Favret and Voegtlin (2001) have suggested that the closed canopy of woods could interfere with the passage of alates into lower regions of the canopy.

546


Egg count decline observed in this study might also be explained by intraspeciÞc competition for preferred oviposition sites. The alder aphid, Pterocallis alni De Geer, favors bark crevices and bud axils for oviposition (Gange and Llewellyn 1988). Similarly, R. padi deposits eggs in axils of buds of Prunus padus L.(Leather 1981). When oviposition space on the axils of P. padus buds is depleted, further oviposition by R. padi often occurs on top of existing eggs, on adjacent bark or on the bud itself (Leather 1990; as cited by Wade and Leather 2002). Eggs laid in such suboptimal locations might be more readily removed by rain, wind, or accessed by predators (Leather 1990, Wade and Leather 2002). Competition for optimal oviposition sites may have been a contributing factor to egg count decline in this study. SigniÞcant declines in egg counts were observed between sampling dates of 15 November 2004 and 23 March 2005 at all sampled heights. Although wind speed was not measured at the sampling heights studied in this investigation, it is an additional variable that could potentially inßuence mortality, because wind speed increases with increased height above ground level (Oke 1978). Population Development on the Primary Host in Spring. Although there was an overall difference in colony growth rates among enclosed and unprotected colonies, a statistically signiÞcant difference in weekly growth rates was not observed until the last round of exclusion treatments (30 May to 8 June 2005). Dispersal from the primary host by alates could have contributed to differences in population growth rates among enclosed and unprotected populations. The natural enemy complex was likely an additional contributing variable responsible for the observed differences, although the protective effect of the enclosures against environmental variables such as rain cannot be discounted. No parasitoids were observed in this study. Aphid parasitoids belonging to the family Braconidae (Hymenoptera) overwinter as last-instar larvaeÐprepupa in mummiÞed aphids (Stary 1970). For parasitism of spring aphid populations to occur, adult parasitoids must be temporally and spatially coincident with spring fundatrix populations. As a new pest to North America, it is unclear to what extent native and established parasitoids have adapted to the life cycle of the soybean aphid. In the initial stages of the hostÞnding process, female adult parasitoids may use olfactory cues for orientation (Vinson 1976). These olfactory cues may include volatiles originating from a variety of potential sources, such as the host, byproducts of the host, the plant host, and volatiles originating from interactions of the herbivore with the host plant (Titayavan and Altieri 1990, Turlings and Tumlinson 1992, Zhang et al. 1998, Bradburne and Mithen 2000). Both soybean and R. cathartica are non-native species (Voegtlin et al. 2004b). It is plausible that many parasitoid species in North America, although capable of parasitizing soybean aphid, may lack innate preferences for the olfactory cues associated with the soybean aphid and its plant hosts.

Vol. 36, no. 3

Intraguild predation by predators may also have contributed to the lack of parasitism observed in this study. Both adult and larval parasitoids can be consumed by predators (Kindlmann and Ruzicka 1992, Ferguson and Stiling 1996, Volkl and Kraus 1996, Rosenheim 1998, Meyhofer and Hindayana 2000). It would be reasonable to hypothesize that, in this study, the probability of a larval parasitoid being killed by intraguild predation might be especially high, given that aphid colonies observed were small. Development of A. glycines Populations in Soybean Fields Adjacent to Overwintering Sites. No edge effects were observed in the development of soybean aphid populations in soybean Þelds immediately adjacent to overwintering sites. This is consistent with observations made by other researchers (Ragsdale et al. 2004, Onstad et al. 2005). Observations of initial colonization of soybean Þelds by A. glycines suggest that migrant alate females make many ßights within a single Þeld, stopping for a short time to feed and deposit nymphs before moving to a new host plant (Ragsdale et al. 2004). Even if alate females leaving primary host plants Þrst visit immediately adjacent soybean plants, our study may have lacked the temporal resolution necessary to pick up such movement. Yearly Variability in Aphid Populations. Soybean aphid infestations in southwestern Ontario were Þrst documented in 2001 (Hunt et al. 2003). Since that time, year-to year variations in the severity of infestations in Ontario have been observed, with the highest populations of soybean aphid observed in 2001, 2003, and 2005 (T. Baute, personal communication). Reasons for this variability are poorly understood but could be related to any of the following factors: (1) cyclical relationships with natural enemies including H. axyridis (C. B. and M. S., unpublished data), (2) spatial and temporal patterns of insecticide application in southwestern Ontario, (3) suitability and availability of overwintering habitats and primary hosts such as R. cathartica, and (4) the size and frequency of dispersal events involving the movement of A. glycines alate populations into Ontario from potential source populations originating in neighboring states such as Michigan and Ohio (T. Baute, personal communication). Heavy aphid pressure in southwestern Ontario during the summer of 2005 was preceded by the observation of large numbers of eggs during the autumn of 2004 at both the location described in this study and across southwestern Ontario (T. Baute, personal communication). No eggs were collected in this study in the spring of 2006, and no aphids were collected during the months of April and May 2006 on R. cathartica at the study location. In this study, we estimated overwintering mortality experienced by populations of soybean aphid eggs. Because the primary host range of A. glycines in North America seems to be quite limited, the possibility of estimating next season risk to soybeans based on egg mortality is a practical possibility. If such estimates of risk were shown to be reliable, the probable return on early-season control options (i.e., neonicotinoid seed

June 2007


treatments) might be more easily judged for any given growing season and region.

Acknowledgments We thank R. Foottit and E. Maw of Agriculture and Agrifood Canada for assistance in the identiÞcation of specimens. The authors thank technicians T. Phibbs, Z. Peters, C. Smith, and L. Daust for assistance in the Þeld and in the laboratory. Support for this project was received from the Ontario Ministry of Agriculture, Food and Rural Affairs New Directions Program, Syngenta Canada, Bayer Crop Science, the Canadian Adaptation Council, the Grape Growers and Wine Councils of Ontario, and the Ontario Soybean GrowerÕs Association. We thank Dr. Chris Difonzo and the anonymous reviewers for their editorial comments, which were of great assistance in improving the manuscript.

References Cited Blackman, R. L., and V. F. Eastop. 2000. Aphids on the worldÕs crops: an identiÞcation and information guide. Wiley, Chinchester, UK. Bradburne, R. P., and R. Mithen. 2000. Glucosinolate genetics and the attraction of the aphid parasitoid Diaeretiella rapae to Brassica. Proc. R. Soc. Lond. Ser. B 267: 89 Ð95. Brewer, M. J., and T. Noma. 2004. Assessment of soybean aphid parasitoids, predatory ßies, and pathogens in cultivated habitats in soybean production areas: NCR-125 Arthropod biological control (http://www.cips.msu.edu/ ncr125/StateRpts2004MI.htm). Bronson, T. E. 1935. Observations on winter survival of pea aphid eggs. J. Econ. Entomol. 28: 1030 Ð1036. Difonzo, C. 2002. Soybean aphid in Michigan (http://www. ipm.msu.edu/CAT02 ßd/pdf/5Ð9SoybeanAphid.pdf). Dunn, J. A., and D. W. Wright. 1955. Overwintering egg populations of the pea aphid in East Anglia. Bull. Entomol. Res. 46: 389 Ð392. Ellingson, B. 2004. Soybean aphid parasitoids in Wisconsin (http://www.plantpath.wisc.edu/soyhealth/aphids/ sbanatenemy.htm). Favret, C., and D. J. Voegtlin. 2001. Migratory aphid (Hemiptera: Aphididae) habitat selection in agricultural and adjacent habitats. Environ. Entomol. 30: 371Ð379. Ferguson, K. I., and P. Stiling. 1996. Non-additive effects of multiple natural enemies on aphid populations. Oecologia (Berl.) 108: 375Ð379. Foottit, R. G., and W. R. Richards. 1993. The genera of the aphids of Canada, Homoptera: Aphidoidea and Phylloxeroidea. Ottawa: Research Branch, Agriculture Canada. Fox, T. B., D. A. Landis, F. F. Cardoso, and C. D. DiFonzo. 2004. Predators suppress Aphis glycines Matsumura population growth in soybean. Environ. Entomol. 33: 608 Ð 618. Gange, A. C., and M. Llewellyn. 1988. Egg distribution and mortality in the alder aphid, Pterocallis alni. Entomol. Exp. Appl. 48: 9 Ð14. Hunt, D., R. Foottit, D. Gagnier, and T. Baute. 2003. First Canadian records of Aphis glycines (Hemiptera: Aphididae). Can. Entomol. 135: 879 Ð 881. Kindlmann, P., and Z. Ruzicka. 1992. Possible consequences of a speciÞc interaction between predators and parasites of aphids. Ecol. Model. 61: 253Ð265.

547

Leather, S. 1981. Factors affecting egg survival in the bird cherry-oat aphid, Rhopalosiphum padi. Entomol. Exp. Appl. 30: 197Ð199. Leather, S. R. 1980. Egg survival in the bird cherry-oat aphid, Rhopalosiphum padi. Entomol. Exp. Appl. 27: 96 Ð97. Leather, S. R. 1990. The role of host quality, natural enemies, competition and weather in the regulation of autumn and winter populations of the bird cherry aphid, pp. 35Ð 44. In A. D. Watt, S. R. Leather, M. D. Hunter, and N.A.C. Kidd (eds.), Population dynamics of forest insects. Intercept Press, Andover, MA. Liu, J., K. Wu, K. R. Hopper, and K. Zhao. 2004. Population dynamics of Aphis glycines (Homoptera: Aphididae) and its natural enemies in soybean in northern China. Ann. Entomol. Soc. Am. 97: 235Ð239. McCornack, B. P., M. A. Carrillo, R. C. Venette, and D. W. Ragsdale. 2005. Physiological constraints on the overwintering potentiaal of the soybean aphid (Homoptera: Aphididae). Environ. Entomol. 34: 235Ð240. Meyhofer, R., and D. Hindayana. 2000. Effects of intraguild predation on aphid parasitoid survival. Entomol. Exp. Appl. 97: 115Ð122. Oke, T. R. 1978. Boundary layer climates. Methuen & Co., New York. Onstad, D. W., S. Fang, D. J. Voegtlin, and M. G. Just. 2005. Sampling Aphis glycines (Homoptera: Aphididae) in soybean Þelds in Illinois. Environ. Entomol. 34: 170 Ð 177. Ragsdale, D. W., D. J. Voegtlin, and R. J. O’Neil. 2004. Soybean aphid biology in North America. Ann. Entomol. Soc. Am. 97: 204Ð208. Rosenheim, J. A. 1998. Higher-order predators and the regulation of insect herbivore populations. Annu. Rev. Entomol. 43: 421Ð 447. Salt, R. W. 1961. Principles of insect cold-hardiness. Annu. Rev. Entomol. 6: 55Ð74. SAS Institute. 2003. Version 9.1. SAS Institute, Cary, NC. Somme, K. 1982. Supercooling and winter survival in terrestrial arthropods. Comp. Biochem. Physiol. 73A: 519 Ð 543. Stary, P. 1970. Biology of aphid parasites (Hymenoptera: Aphidiidae) with respect to integrated control. Dr. W. Junk N. V., The Hague. Titayavan, M., and M. A. Altieri. 1990. Synomone-mediated interactions between the parasitoid Diaeretiella rapae and Brevicoryne Brassicae under Þeld conditions. Entomophaga 35: 499 Ð507. Turlings, T.C.J., and J. H. Tumlinson. 1992. Systemic release of chemical signals by herbivore-injured corn. Proc. Natl. Acad. Sci. U.S.A. 89: 8399 Ð 8402. Vinson, S. B. 1976. Host selection by insect parasitoids. Annu. Rev. Entomol. 21: 109 Ð133. Voegtlin, D. J., S. E. Halbert, and G. Qiao. 2004a. A guide to separating Aphis glycines Matsumura and morphologically similiar species that share its hosts. Ann. Entomol. Soc. Am. 97: 227Ð232. Voegtlin, D. J., R. J. O’Neil, and W. R. Graves. 2004b. Tests of suitability of overwintering hosts of Aphis glycines: identiÞcation of a new host association with Rhamus alnifolia LÕHeritier. Ann. Entomol. Soc. Am. 97: 233Ð 234. Volkl, W., and W. Kraus. 1996. Foraging behaviour and resource utilization of the aphid parasitoid Pauesia unilachni: adaptation to host distribution and mortality risks. Entomol. Exp. Appl. 79: 101Ð109.

548


Wade, F. A., and S. R. Leather. 2002. Overwintering of the sycamore aphid, Drepanosiphum platanoidis. Entomol Exp. Appl. 104: 241Ð253. Wu, Z., Z. Sciienk-Ilamlin, W. Ziaan, D. W. Ragsdale, and G. E. Heimpel. 2004. The soybean aphid in China: a historical review. Ann. Entomol. Soc. Am. 97: 209 Ð218.

Vol. 36, no. 3

Zhang, Y., B. Guo, Z. Hou, X. Chen, and F. Yan. 1998. Olfactory orientation of the parasitoid wasp Lysiphlebus fabarum to its host food plants. Entomol. Sin. 5: 74 Ð 82. Received for publication 31 August 2006; accepted 2 February 2007.