Key words: aquatic invertebrate herbivore, biological control, Finger Lakes, freshwater herbivory, Euhrychiopsis lecontei, survey, screening.

Transcription

1 Aquatic Ecology 31: , c 1998 Kluwer Academic Publishers. Printed in Belgium. Decline of the invasive submersed macrophyte Myriophyllum spicatum (Haloragaceae) associated with herbivory by larvae of Acentria ephemerella (Lepidoptera) Robert L. Johnson, Elisabeth M. Gross ; and Nelson G. Hairston, Jr. Section of Ecology & Systematics, Corson Hall, Cornell University, Ithaca, NY, , USA; ( rlj5@cornell.edu); Authors for correspondence, Present address: Limnological Institute, University of Konstanz, D Konstanz, Germany Accepted 6 May 1997 Key words: aquatic invertebrate herbivore, biological control, Finger Lakes, freshwater herbivory, Euhrychiopsis lecontei, survey, screening Abstract Myriophyllum spicatum, an exotic submersed macrophyte causing serious lake management problems throughout much of North America, decreased markedly in biomass in Cayuga Lake, NY, USA, since the beginning of the 1990s. Over the same period, however, the total biomass of all species of submersed macrophytes did not decline, and native macrophytes gained in abundance. The aquatic moth larva, Acentria ephemerella, was first observed on milfoil plants in Cayuga Lake in However, due to its cryptic habit the larva may have been present prior to that year. When the density of these grazers is high, herbivory by Acentria causes severe damage to the apical meristem of M. spicatum. This moth larva and another milfoil herbivore, Euhrychiopsis lecontei are widespread in 26 lakes surveyed in New York State; they are present in 25 and 24 lakes, respectively. Estimates of Acentria larval densities in summer in Cayuga Lake are 27 to 100 m,2, and a quantitative survey of larvae hibernating in milfoil stems revealed mean densities of 500 m,2 in late fall in Seneca Lake. In laboratory experiments, Acentria larvae feed on a wide variety of macrophytes commonly found in New York State. Although Acentria is not a specialist feeder, its life cycle is closely tied to M. spicatum through the moth s use of apical tips and stems for summer and winter refuges; thus deleterious damage to other macrophytes is low. Introduction The introduced submersed macrophyte Myriophyllum spicatum L. (Haloragaceae, Eurasian watermilfoil or short milfoil) frequently replaces native aquatic vegetation, dominating shallow lakes and is a major nuisance for recreation, fishing, and aesthetics (Reed, 1977; Grace & Wetzel, 1978; Smith & Barko, 1990). Milfoil was introduced to the United States most probably in the early 1940s (Couch & Nelson, 1986; Smith & Barko, 1990). Reports of earlier occurrences as early as the beginning of this century (see, e.g., Reed, 1977) are attributable to misidentification of the native milfoil, Myriophyllum sibiricum Komarov (syn. M. exalbescens Fernald), since both species have similar morphological characteristics (Aiken, 1981). Management of milfoil in New York State and elsewhere has yielded little long-term success, although different management methods such as herbicides, mechanical harvesting, water drawdowns (Sheldon & Creed, 1995), and unspecific biocontrol with triploid grass carp (Ctenopharyngodon idella Val.; R.L. Johnson, pers. comm.) have been used. In fact, M. spicatum is the least preferred food for grass carp given a choice of different submersed macrophytes (Pine & Andersen, 1991; McKnight & Hepp, 1995). Myriophyllum spicatum became established in Cayuga Lake in the mid 1960s (R.L. Johnson, pers. comm.), and had attained high abundances at the shallow north and south ends of the lake. Beginning in 1987, milfoil declined markedly. The decline was concurrent with the appearance of the aquatic moth

2 274 larva Acentria ephemerella Denis and Schiffermüller (= Acentria nivea Olivier; Passoa, 1988). Severe herbivore damage caused by these moth larvae has been observed since 1991 on milfoil shoots, especially on the apical meristem. Although herbivory has traditionally been considered to have little impact on the growth of submersed macrophytes (Hutchinson, 1975; Wetzel, 1983), more recently, vertebrate and invertebrate grazers have been implicated in influencing macrophyte biomass, productivity, and species composition (Lodge, 1991; Newman, 1991; Lodge et al., in press). Thus, the milfoil-acentria association offers a good opportunity to study freshwater plant-herbivore interactions. Further, using a naturally occurring invertebrate to control a nuisance aquatic weed may provide a long-term solution to a vexing lake-management problem at low environmental risk and financial cost. Three aquatic invertebrates occurring in North America have been investigated as possible control agents for Eurasian watermilfoil: the chironomid midge larva Cricotopus myriophylli Oliver (Mac- Rae et al., 1990; Kangasniemi, 1983), the weevil Euhrychiopsis lecontei Dietz (Creed & Sheldon,1995), and the aquatic moth larva Acentria ephemerella ( A. nivea : Buckingham & Ross, 1981; Painter & McCabe, 1988). Extensive studies have been performed to explore the suitability of E. lecontei as a biocontrol agent for milfoil in Vermont, USA (Sheldon & Creed, 1995). E. lecontei, which is native to North America, has expanded its host range from the native milfoil, M. sibiricum, to the invasive Eurasian watermilfoil and may further shift host due to increased specificity for M. spicatum (Solarz & Newman, 1996). Acentria was first recorded in Montreal, Canada in 1927 (Sheppard, 1945). Since then it has been reported from Ontario, New York, Massachusetts, and Wisconsin (Batra, 1979; Buckingham & Ross, 1981), Vermont (Creed & Sheldon, 1994a), Minnesota and Wisconsin (Newman & Maher, 1995) and as far west as the Missouri River valley in Iowa (Scholtens & Balogh, 1996). It has been associated with milfoil declines in the Kawartha Lakes, Ontario (Painter & McCabe, 1988) and may also be responsible in part for a milfoil decline in Brownington Pond, Vermont (Creed & Sheldon, 1994a). Although native to Europe (Berg, 1942), Acentria is apparently widespread in the northeast of North America now. Since this larva is very cryptic it may have been overlooked at other locations. In order to explore the role of Acentria in milfoil declines in New York State, long-term data sets on macrophyte densities in the Finger Lakes region (Oglesby et al., 1976; Johnson, 1995) were used. We assessed changes in total submersed macrophyte biomass and community changes associated with the disappearance of milfoil over the period During 1994, the abundance of Acentria larvae taken from Cayuga and Seneca lakes was evaluated. In addition, an extensive survey of 26 lakes in the Finger Lakes- Lake Ontario watershed and some adjacent lakes was performed in 1996 to investigate the distribution of herbivores that live on milfoil, especially Acentria and E. lecontei. Since Acentria is not a specialist feeder on milfoil, macrophyte damage observed in the field was compared with results of diet choice experiments carried out under experimental conditions. Materials and methods Cayuga Lake is a glacially-scoured mesotrophic lake of 61 km length, with a mean width of 2.8 km and a mean depth of 54.4 m (Birge & Juday, 1914; Oglesby, 1978). It is the second largest of the Finger Lakes in central New York. The lake morphometry supports significant macrophyte communities within the littoral zones of the northern and southern basins. The study area in both basins includes sites sampled in previous investigations. The long-term data sets used include 1970 (Vogel, 1973), 1972 and 1973 data (Oglesby et al., 1976) from the southern basin and data from the northern basin (Peverly et al., 1974). Data from the northwestern part of the lake for were collected by Miller (1988, 1989, 1990) and transformed from his reported wet weights by multiplying the wet weights by 0.1 to obtain dry weights for comparison. This conversion falls within the range given by Wetzel (1983), who cites a value of 88% mean water content and a range of 85 to 92% for submersed macrophytes. For the southern basin in and the northern basin in , submersed aquatic plant samples were collected between August 1 and September 15 when milfoil reached its maximum seasonal biomass. All other samples had been collected at a comparable time of the year, between the last week of July and end of August (Vogel, 1973; Peverly et al., 1974; Oglesby et al., 1976; Miller, 1988, 1989, 1990). Macrophytes were sampled for analysis of biomass and species diversity at eleven sites in the southwestern corner of the southern basin, and two sites in the northwestern corner of the northern basin of Cayuga Lake. All sites had a mean water depth of 2 3 m. The indi-

3 275 vidual sites were 100 by 100 m quadrants within which plot-sampling was carried out (see Johnson, 1995, for details). Submersed macrophytes were sampled in the southern basin by hand-harvesting 5 randomly selected 1 m 2 quadrats from each of the eleven sites during , and at 20 randomly selected 0.25 m 2 quadrats from each of the same eleven sites during , yielding the same total sampled area of 55 m 2. The two northern basin sites were sampled for submersed aquatic plants by hand-harvesting 20 randomly selected 0.25 m 2 quadrats at each site. A diver using an air buoy underwater breathing apparatus hand-harvested all above sediment plant biomass from each plot. Each sample was placed in a separate plastic bag and transported in a cooler for laboratory analysis, or frozen for later separation and dry weight determination. Plant biomass was first washed to remove sediment and algae from the plant surface and then separated by species for each plot. The separated plant species were oven dried at 105 C for 48 h to determine dry weight. The mean biomass was determined for each species within a site, and from these data, the mean biomass for the northern and southern basins were calculated. Data are presented in grams dry weight per square meter. Numbers of Acentria larvae were determined from samples taken for macrophyte biomass in 1994 at the southern end of Cayuga Lake. Tightly sealed bags containing the macrophyte samples were opened at 24 and 48 h. The low oxygen stress that the sealed bags created induced moth larvae to crawl off the macrophytes, permitting them to be counted, and hence larval density to be estimated. In November 1994, fifteen 0.25 m 2 plots from Seneca Lake (the largest of the New York Finger Lakes) were screened for Acentria larvae hiding in milfoil stems and on other macrophytes. Each stem was carefully checked under a dissecting microscope and all larvae extracted from the stem. This procedure yielded a more precise estimate of the numbers of Acentria present than the sealed-bag procedure. Twenty-six lakes in 16 counties in central New York State were surveyed for the presence of aquatic insect herbivores on milfoil. Lakes were sampled between July and October 1996, with most lakes visited in August and September. Some lakes were sampled several times during the summer. Twenty-five randomly picked tips (25 cm long) of milfoil were collected from four different sites at each lake. Additionally, 25 complete stems (tip to root) were taken at one of the four sites to account for herbivores occurring other than at the apical tip. Each stem was sealed in a small plastic bag and collected material was transported back to the lab in a cooler and kept at 4 C until screened for invertebrates. The time between sampling and screening did not exceed 4 days, with the exception of four of the 35 sampling sites. Invertebrates in these samples, which were frozen for up to two weeks, were still intact and easy to identify. Each stem was carefully screened under a dissecting microscope and all invertebrates present were recorded. Larvae for the feeding experiment were collected from macrophytes (M. spicatum or Ceratophyllum demersum L.) originating from either Seneca Lake or Cayuga Lake. They were kept individually in 55 mm plastic petri dishes with 5 ml of aged tap water and a whorl of milfoil until experiments were started. Larvae were transferred to plates with fresh food and water at least once a week, and maintained in a controlled environment chamber at 15 C and 14:10 h light:dark cycle. Preliminary studies showed that feeding or starving the larvae before the start of these experiments did not affect the outcome, so larvae were fed on M. spicatum prior to the experiments to make sure they were active and feeding. For the feeding trials, plant parts of different macrophyte species were offered to the larvae for 24 h. The following macrophyte species commonly found in Cayuga Lake and surrounding area were used: Ceratophyllum demersum L., Chara vulgaris L., Elodea canadensis Michaux, Heteranthera dubia (Jacquin) MacMillan, Myriophyllum spicatum L., Nasturtiumofficinaleauctt. (= Rorippa nasturtiumaquaticum (L.) Hayek), Nitella sp., Potamogeton crispus L., Potamogeton pectinatus L, RanunculustrichophyllusChaix and Utricularia sp. After 24 h the larvae were transferred into fresh plates with M. spicatum. Feeding activity was estimated by the number of fecal pellets produced. Ten individual larvae were used per treatment. Feeding trials with each macrophyte were repeated at least twice. Additionally, we recorded whether or not the larvae built retreats using the macrophyte offered. Results Milfoil biomass has declined dramatically in Cayuga Lake from the late 1980s to the present (Figure 1). The relative biomass of milfoil in the lake declined from between 58% and 95% in the 1970s and 1980s to 12% at the south and to 0% at the north end in 1996 (Figures 1c and d). The very low macrophyte biomass in 1972 at the south end is related to a tropical storm

4 276 event (Oglesby et al., 1976). Absolute milfoil biomass declined at the southern end from over 160 g m,2 in the 1970s to 16.5 g m,2 in 1996 (Figure 1a), and disappeared completely from the sampling quadrants at the northern end in 1996 after peaks of 325 g m,2 in 1988 and 763 g m,2 in 1974 (Figure 1b). Over the same period, the total biomass of all submersed macrophytes at both ends of the lake fluctuated between 26 and 435 g m,2 (with the exception of the very high milfoil densities in 1973 and 1974 at the northwest corner), without any consistent trend toward increase or decrease, particularly during the period of milfoil decline ( , Figures 1a and b). Native species were present in low abundance throughout the milfoil invasion. The milfoil decline in Cayuga Lake has been accompanied by an increase in native macrophyte abundance (Figure 2). At the south end Elodea canadensis and Potamogeton pusillus (Figure 2a) increased, whereas at the north end Chara vulgaris, Heteranthera dubia and Vallisneria americana (Figure 2b) are dominant. In 1991, larvae of Acentria were found on milfoil shoots for the first time. At this time, the milfoil decline was becoming obvious, but we did not appreciate the potential role of these herbivores as control agent of milfoil until The density of Acentria larvae from macrophytes collected for the biomass measurements in southern Cayuga Lake in 1994 was estimated at m,2 in midsummer and only 4 5 m,2 at the beginning of September. Quantitative screening for hibernating Acentria in Seneca Lake yielded densities between 216 and 824 m,2 at the beginning of November Acentria is present year-round in Cayuga and Seneca Lakes and can be found in high numbers, primarily on Myriophyllum spicatum. In Cayuga Lake, Acentria feeds preferentially on milfoil, damaging especially the apical meristem, and thus substantially affecting its growth. Acentria feeding on Ceratophyllum, Elodea or species of Potamogeton seldom destroy the apical tips of those macrophytes and thus have little or no measurable effect on growth. However, feeding experiments in which larvae were not given a choice of plants to consume revealed that a broad range of macrophytes is palatable to this herbivore (Table 1). At the same time, a few macrophyte species were not consumed at all (Chara vulgaris, Nitella sp., Nasturtium officinale, Heteranthera dubia, Ranunculus trichophyllus). Acentria did not typically build retreats on the macrophytes that were rejected as food, however in a few instances, some larvae attached to the offered macrophyte to gain shelter (Table 1). Since milfoil started to decline, Elodea canadensis and Potamogeton pusillus have increased in biomass at the southwest corner of Cayuga Lake (Figure 2a). The northwest corner exhibits a different macrophyte community structure with Chara vulgaris, Heteranthera dubia and Vallisneria americana being the dominant macrophytes now (Figure 2b). Feeding activity on Elodea was only moderate in the experiments (Table 1) which is in accordance with Elodea gaining importance in southern Cayuga Lake (Figure 2a). High feeding ability on Ceratophyllum demersum and some species of Potamogeton under laboratory conditions does not reflect the situation in the field (Figure 2). The lakes surveyed for presence of milfoil herbivores were located principally in the Finger Lakes region of NewYork State. Both Acentria and the weevil E. lecontei are widespread: Acentria was found at 31 of 35 sites and in 25 of the 26 surveyed lakes, whereas the weevil occurred at 29 of 35 sites and in 24 out of 26 lakes (Table 2). At many places one or the other of these herbivores dominated in numbers with codominance being rare (R.L. Johnson & E.M. Gross, unpubl.). The midge larvae Cricotopus myriophylli were found on milfoil collected from the Experimental Ponds at Cornell University, and although many Chironomid midges were found on milfoil in the lakes surveyed, their taxonomic identification is pending. Discussion Myriophyllum spicatum has declined substantially in both the northern and southern basins of Cayuga Lake since In the 1970s milfoil reached very high biomass(upto763gm,2 ) in the range of dense milfoil beds observed elsewhere (280 to 1150 g m,2, Grace & Wetzel, 1978). This plant apparently gained dominance over other macrophytes in Cayuga Lake following the effects of tropical storm Agnes in 1972 (Oglesby et al., 1976). The growth structure of milfoil (i.e., canopy formation) and its ability to disperse by shoot fragments probably enabled this species to survive the storm-related decrease in transparency and increase in suspended sediment. No data were collected between 1974 and 1987, but during this period milfoil may have already started to decline. It was, however, still the dominant macrophyte between 1987 and 1989 with proportions of 30 to 75% of total aquatic plant biomass (Figures 1 and 2).

5 277 Figure 1. Biomass of M. spicatum and other submersed macrophytes in Cayuga Lake, and (a) and (c) Southwest corner, (b) and (d) Northwest corner. (a) and (b) display the absolute biomass of all submersed macrophytes. All macrophytes except milfoil are represented by the white part of the bar, the milfoil biomass is shown as the cross-hatched part of the bar. Note the different scaling of the y-axis, data are given in (g m,2 ). (c) and (d) show milfoil biomass as % of total submersed aquatic macrophyte biomass for both sites. At times marked with N no data were collected. Figure 2. Submersed macrophyte composition in Cayuga Lake after the decline of M. spicatum. (a) Southwest corner, (b) Northwest corner. Data are shown as relative biomass (%), the absolute biomass of all macrophytes can be found in Figure 1a.

6 278 Table 1. Feeding and retreat building of Acentria ephemerella larvae on selected submersed macrophytes. The feeding rate was determined as fecal pellets (fp) produced per hour. Each macrophyte species was tested at least twice with 10 larvae per experiment. Data are presented as means 1 SE of the single experiments (n). Since M. spicatum was run as a control each time, the number of replicate experiments is higher for this species Species Feeding rate (mean 1 SE) n Retreat [fp h,1 ] building Myriophyllum spicatum Potamogeton crispus Potamogeton pectinatus Elodea canadensis Ceratophyllum demersum Utricularia sp Heteranthera dubia Ranunculus trichophyllus Nasturtium officinale , Chara vulgaris , Nitella sp , In 1991, milfoil shoots in Cayuga Lake exhibited severe herbivore damage which led to our discovery of the presence of Acentria larvae. Unlike previous years of high M. spicatum biomass, milfoil shoots have not reached the water surface to form canopies since Herbivore damage occurs mainly on the apical meristem, arresting stem elongation and forcing milfoil to invest in side shoots. Under high herbivory, these side shoots are in turn damaged, further stunting stem growth. Due to the large number of plots sampled for the macrophyte biomass survey, a more thorough screening for Acentria and other herbivores could not be performed. Acentria densities of 27 to 100 m,2 in Cayuga Lake found during 1994 are likely to be substantial underestimates because only larvae that crawled off the sampled macrophytes were counted.recent studies indicate that this method censuses only about 10 20% of all larvae present (R.L. Johnson, unpubl.). Finally, densities of hibernating Acentria from one location in Seneca Lake averaged 500 m,2 ( m,2 ), and similar densities of 300 m,2 were found for A. ephemerella on Zostera marina in Danish coastal waters (Hedal & Schmidt, 1992). There may, in addition, be seasonal variation in larval density related to the life-cycle of this moth. Larvae use milfoil leaves to build retreats in summer and hide inside milfoil stems in the winter. After milfoil declined to very low densities in 1995, larvae overwintered in special hibernacula on Ceratophyllum demersum. C. demersum is heavy enough to withstand the strong autumnal wave action in Cayuga and Seneca Lakes that removes most other submersed macrophytes from shallow areas. Furthermore, this macrophyte stays metabolically active in winter (Spencer & Wetzel, 1993) and may provide an oxygen supply for the overwintering larvae. Acentria is assumed to be an invading species from Europe (Batra, 1977). It may have entered North America along the St. Lawrence River and from there invaded the North American Great Lakes region, where it has been found as far west as Minnesota and Wisconsin (Newman & Maher, 1995) and the Missouri River valley in Iowa (Scholtens & Balogh, 1996). The results of our survey indicate that Acentria is present in the majority of lakes in central New York State and is thus more ubiquitous than previously assumed. The larvae have probably been overlooked in the past because they are cryptic, especially in the first or second instars. Taken together, these findings suggest that herbivory by Acentria is a major cause of the decline of M. spicatum in Cayuga Lake. Acentria is associated with milfoil declines in other lakes as well. In the Kawartha Lakes, Canada, where milfoil declined by 95%, Painter & McCabe (1988) found the moth larvae present in exactly the same abundance (6 larvae per 10 apical tips of plants 25 cm long) as we observed in Cayuga Lake. The milfoil decline in Brownington Pond, Vermont, USA, reported by Creed & Sheldon (1994a) may also be attributable, at least in part, to the Acentria lar-

7 279 Table 2. Occurrence of Acentria ephemerella and Euhrychiopsis lecontei in lakes in central New York. Samples were taken from 35 sites in 26 lakes belonging to 16 counties. The presence or absence of a herbivore is marked with y or n, respectively. Lakes marked with an were not included in the 1996 quantitative survey but were sampled at least once within 1995 to 1997 Lake Location County Coordinates Acentria E. lecontei Cayuga Seneca Falls Seneca N, W y n Cayuga Union Springs Cayuga N, W y y Cayuga Ithaca Tompkins N, W y n Cazenovia Cazenovia Madison N, W y y Champlain Willsboro Essex N, W y y Chautauqua Chautauqua Chautauqua N, W y y Crooked Tully Onondaga N, W y y Dryden Dryden Tompkins N, W y y George Lake George Warren N, W y y Jamesville Jamesville Onondaga N, W y y Keuka Penn Yan I Yates N, W n y Keuka Penn Yan II Yates N, W n y Keuka Branchport Yates N, W y y Keuka Hammondsport Steuben N, W y y Lamoka Tyrone Schuyler N, W y y Lagrange Lagrange Wyoming N, W y y LeRoy LeRoy Genesee N, W y y Little York Little York Cortland N, W y y Moraine Hamilton Madison N, W y y Neatawanta Fulton Oswego N, W y y Ontario Fairhaven Cayuga N, W y y Otter Meridian Cayuga N, W y y Otisco Amber Onondaga N, W y y Otisco Rice Grove Onondaga N, W n n Owasco Auburn Cayuga N, W y y Panther Carterville Oswego N, W n y Salubria Bath Steuben N, W y n Seneca Geneva I Seneca N, W y y Seneca Geneva II Seneca N, W y n Seneca Watkins Glen Schuyler N, W y n Silver Perry Wyoming N, W y y Spencer North Spencer Tioga N, W y y Tully Tully Onondaga N, W y y Tully Preble Cortland N, W y y Waneta Weston Schuyler N, W y y vae that were present. In their study, Creed & Sheldon (1994a) ascribe the decline to feeding by the weevil, E. lecontei. However, the moth larvae were also present in high enough numbers to cause significant effects (this study; Painter & McCabe, 1988) and apparently caused more damage to milfoil plants than did the weevil in concomitant laboratory experiments (Creed & Sheldon, 1994a). Similar sudden disappearances of milfoil in other lakes of the Finger Lakes region of New York State are also concurrent with the presence of high numbers of Acentria larvae (R.L. Johnson, unpubl.). Alternative explanations for milfoil declines have not been found. Although nutrient limitation could cause a reduction in milfoil density, it cannot explain the events in Cayuga Lake since total macrophyte biomass should have been affected (Moran et al., 1993), but was not (Figure 1A). Similarly, Painter & McCabe (1988) excluded nutrient impacts as a reason for the milfoil decline in the Kawartha Lakes, Canada.

8 280 The weevil, Euhrychiopsis lecontei, has been suggested as an important herbivore on M. spicatum in lakes in North America. It has been found in association with milfoil in several lakes located in Vermont, Massachusetts, New York, and Connecticut in the USA, as well as in Alberta, Canada (Creed & Sheldon, 1994b). By comparison, previous records of Acentria include Minetto (Forbes, 1938), Varna (Munroe, 1947), and Brownington Pond, Vermont (Creed et al., 1992) in the USA, and Montreal (Sheppard, 1945), Lac a la Tortue, St. Anne de Bellevue, both Quebec (Munroe, 1947), Dundas Marsh, Ontario (Judd, 1947) and the north shore of Lake Erie (Judd, 1950) in Canada. Both Acentria and E. lecontei were present in nearly all of the 26 lakes we surveyed in New York State. In most instances, one of these two herbivores was dominant. For example, in Dryden Lake E. lecontei dominated and Acentria was only present at low density. The shoreline of Dryden Lake, with many deciduous trees, offers appropriate shelter for adult weevils in the autumn when they leave the water. Dryden Lake is eutrophic and milfoil often exhibits a dense epiphyte growth. Experiments have shown that Acentria avoids macrophytes with a thick epiphyte cover (E.M. Gross, unpubl.; see also Buckingham & Ross, 1981). In contrast to Dryden Lake, Acentria is virtually the only herbivore present in Cayuga and Seneca lakes; weevils have only been found in low numbers in a few sheltered bays with shallow slopes (see Table 2). Neither Cayuga nor Seneca Lake offers suitable overwintering habitat for adult weevils along its shore. Furthermore, because both Acentria and Euhrychiopsis use the apical tips of milfoil for their development, competitive interactions between the species are possible and deserve further investigation. Initial evidence from our survey of lakes within New York State indicates that lakes with a high abundance of Acentria or Euhrychiopsis exhibit severe damage to milfoil (R.L. Johnson & E.M. Gross, unpubl.). According to its feeding pattern (see Table 1) and previous studies (Berg, 1942; Batra, 1979; Buckingham & Ross, 1981) Acentria is a generalist feeder. However, its non-selective feeding on various macrophytes under laboratory conditions does not reflect the distribution of this herbivore found in the field, or the damage observed on macrophytes other than milfoil. Although principally found on M. spicatum in Cayuga Lake, Acentria also occurs on Ceratophyllum demersum, Elodea canadensis and Potamogeton richardsonii. Other macrophytes are apparently not seriously damaged by Acentria herbivory. The two ends of Cayuga Lake have developed different macrophyte species compositions after the milfoil decline. In the southern end, Potamogeton species and Elodea are nowdominant, despite the fact that Acentria can potentially feed on them. At the northern end, at least two of the dominant species, Chara vulgaris and Heteranthera dubia, are not affected by Acentria herbivory at all. In ongoing studies, we are evaluating the host choice of Acentria and its herbivore effect on the macrophyte community in Cayuga Lake. Although non-selective feeding might initially be regarded as a drawback to the use of this species for biocontrol purposes, both in Cayuga Lake and in the Kawartha Lakes (Painter & McCabe, 1988), where Acentria is abundant, native macrophytes appear healthy, diverse and abundant. Indeed, the oligophagous nature of this herbivore might be advantageous, because other less-impacted macrophytes provide a refuge for the moth larvae when milfoil is driven to low density. Although not native to North America, the widespreadoccurrenceof Acentria indicates that it is nowanestablishedspecies. Thus, usingacentria as a biocontrol for milfoil would not constitute a new introduction but only an augmentation of larval densities to effective numbers. Acknowledgements We wish to thank E. Mulligan, R. Schindelbeck, K. Switzer, D. Corson, T. Groman, J. Shaffer, P. VanDusen, K. Thomas, E. Thomas, J. Riggs, J. Toner, & K. Hester for help with the field collections and attention to detail during the hundreds of hours at the microscope or separating of plant species. The identification of Acentria ephemerella and Euhrychiopsis lecontei was confirmed by R.E. Hoebeke, Cornell Insect Collection, Ithaca, NY. R.W. Bode, NY State Museum, Albany, NY, verified the identification of the adults and larvae of Cricotopus myriophylli. A special thanks to D.R. Bouldin for his support and encouragement throughout this study. This work was funded by Hatch Project (NYC ) USDA, the Aquatic Vegetation Control Program of the Tompkins County Planning Department, the Aquatic Vegetation Control Program of the Seneca County Soil and Water Conservation District and the Finger Lakes-Lake Ontario Watershed Protection Alliance formally the Aquatic Vegetation Control Program of the Water Resources Board. These funds were allocated by the New York State Legislature to the above organizations. J.E. Skaley of the Tompkins

9 281 County Planning Department and J. Malyj with the Seneca County Soil and Water Conservation District deserve special thanks for their support, without it this study would not have taken place. References Aiken SG (1981) A conspectus of Myriophyllum (Haloragaceae) in North America. Brittonia 33: Batra SWT (1977) Bionomics of the aquatic moth Acentropus niveus (Oliver), a potential biological control agent for Eurasian watermilfoil and Hydrilla. NY Ent Soc 85: Berg K (1942) Contributions to the biology of the aquatic moth Acentropus niveus (Oliv.). Vidensk Medd Dansk Naturh Foren 105: Birge EA and Juday C (1914) A limnological study of the Finger Lakes of New York. US Bur Fish Bulletin 32: Buckingham GR and Ross BM (1981) Notes on the biology and host specificity of Acentria nivea (=Acentropus niveus). J Aquat Plant Manage 19: Couch R and Nelson E (1986) Myriophyllum spicatum. First International Symposium on Watermilfoil (Myriophyllum spicatum) and related Haloragaceae Species. The Aquatic Plant Management Society, Vicksburg, Mississippi, USA, pp Creed RP and Sheldon SP (1994a) The effect of two herbivorous insect larvae on Eurasian watermilfoil. J Aquat Plant Manage 32: Creed RP and Sheldon SP (1994b) Aquatic weevils (Coleoptera, Curculionidae) associated with Northern watermilfoil (Myriophyllum sibiricum) in Alberta, Canada. Ent News 105: Creed RP, Jr. and Sheldon SP (1995) Weevils and watermilfoil: Did a North American herbivore cause the decline of an exotic plant? Ecol Appl 5: Creed RP, Sheldon SP and Cheer DM (1992) The effect of herbivore feeding on the buoyancy of Eurasian watermilfoil. J Aquat Plant Manage 30: Forbes WTM (1938) Acentropus in America (Lepidoptera, Pyralidae). J NY Ent Soc 46: 338 Grace JB and Wetzel RG (1978) The production biology of Eurasian watermilfoil (Myriophyllum spicatum L.): A review. J Aquat Plant Manage 6: 1 11 Hedal S and Schmidt SC (1992) On the occurrence of Acentria ephemerella Den. and Schiff. Ent Meddr 60: Hutchinson GE (1975) A Treatise on Limnology. Vol. III: Limnological Botany. John Wiley and Sons, New York, NY, USA Johnson RL (1995) Monitoring aquatic vegetation in Tompkins County. Project Completion Report. SFY Tompkins County Planning Department. Ithaca, NY, USA Judd WW (1947) Acentropus niveus Olivier (Lepidoptera, Pyralidae) at Hamilton, Ontario. Can Ent 79: 119 Judd WW (1950) Acentropus niveus (Oliv.) (Lepidoptera: Pyralidae) on the north shore of Lake Erie with a consideration of its distribution in North America. Can Ent 82: Kangasniemi BJ and Oliver DR (1983) Chironomidae (Diptera) associated with Myriophyllum spicatum in Okanagan Valley Lakes, British Columbia. Can Entomol 115: Lodge DM (1991) Herbivory on fresh-water macrophytes. Aquat Bot 41: Lodge DM, Cronin G, van Donk E and Froelich AJ (in press) Impact of herbivory on plant standing crop: comparisons among biomes, between vascular and nonvascular plants, and among freshwater herbivore taxa. In: Jeppesen E, Sondergaard M, Sondergaard M and Christoffersen K (eds.). The Role of Submerged Macrophytes in Structuring the Biological Community and Biogeochemical Dynamics in Lakes. Springer Verlag, Berlin, Heidelberg, Germany MacRae IV, Winchester NN and Ring RA (1990) Feeding activity and host preference of the milfoil midge Cricotopus myriophylli Oliver (Diptera: Chironomidae). J Aquat Plant Manage 28: McKnight SK and Hepp GR (1995) Potential effect of grass carp herbivory on waterfowl foods. J Wildl Manage 59: Miller GL (1988) Effects of mechanical control of aquatic vegetation on community structure in Cayuga Lake. Department of Environmental Studies. Asheville, NC, USA Miller GL (1989) Second annual review of the effects of mechanical control of aquatic vegetation on community structure in Cayuga Lake. Department of Environmental Studies. Asheville, NC, USA Miller GL (1990) Third annual review of the effects of mechanical control of aquatic vegetation on community structure in Cayuga Lake. Department of Environmental Studies. Asheville, NC, USA Moran E, Johnson RL and Bouldin D (1993) Changes in the Cayuga Lake macrophyte community between 1970 and Project Completion Report Tompkins County Planning Department. Ithaca, NY, USA Munroe EG (1947) Further North American records of Acentropus niveus (Lepidoptera, Pyralidae). Can Ent 79: 120 Newman RM (1991) Herbivory and detritivory on freshwater macrophytes by invertebrates. A review. J N Am Benth Soc 10: Newman RM and Maher LM (1995) New records and distribution of aquatic insect herbivores of watermilfoils (Haloragaceae, Myriophyllum spp.) in Minnesota. Ent News 106: Oglesby RT (1978) The limnology of Cayuga Lake. In: Bloomfield JA (ed.). Lakes of New York. Vol. I (pp ) Academic Press, New York, NY, USA Oglesby RT, Vogel A, Peverly JH and Johnson R (1976) Changes in submerged plants at the south end of Cayuga Lake following tropical storm Agnes. Hydrobiol 48: Painter DS and McCabe KJ (1988) Investigation into the disappearance of Eurasian watermilfoil from the Kawartha Lakes, Canada. J Aquat Plant Manage 26: 3 12 Passoa S (1988) Systematic positions of Acentria ephemerella (Denis and Schiffermüller) Nymphulinae, and Schoenobiinae based on morphology of immature stages (Pyralidae). J Lepid Soc 42: Peverly JH, Oglesby RT, Vogel A and Johnson RL (1974) Aquatic weed management in the Finger Lakes. Technical Report 90. Cornell University Water Resources and Marine Science Center. Ithaca, NY, USA Pine RT and Anderson LWJ (1991) Plant preferences of triploid grass carp. J Aquat Plant Manage 29: Reed CF (1977) History and distribution of Eurasian watermilfoil in the United States and Canada. Phytologia 36: Scholtens BG and Balogh GJ (1996) Spread of Acentria ephemerella (Lepidoptera: Pyralidae) in central North America. Great Lakes Entomol 29: Sheldon SP and Creed RP Jr. (1995) Use of a native insect as a biological control for an introduced weed. Ecol Appl 5: Sheppard AC (1945) A new record for Canada (Lepidoptera). Can Ent 77: 55 Smith CS and Barko JW (1990) Ecology of Eurasian watermilfoil. J Aquat Plant Manage 28: 55 64

10 282 Solarz SL and Newman RM (1996) Oviposition specificity and behavior of the watermilfoil specialist Euhrychiopsis lecontei. Oecologia 106: Spencer WE and Wetzel RG (1993) Acclimation of photosynthesis and dark respiration of a submersed angiosperm beneath ice in a temperate lake. Plant Physiol 101: Vogel A (1973) Changes in the submerged aquatic flora at the south end of Cayuga Lake between 1929 and Master thesis. Cornell University, Ithaca, NY Wetzel RG (1983) Limnology. 2nd Ed. Saunders College Publishing, New York