Seediscussions,stats,andauthorprofilesforthispublicationat:https://www.researchgate.net/publication/264049385

ImpactofinvasiveearthwormsonIxodesscapularisandotherlitter-dwellingarthropodsinhardwoodforests,centralNewYorkstate,USA

ArticleinAppliedSoilEcology·December2014

DOI:10.1016/j.apsoil.2014.07.005

CITATIONS

2

READS

115

3authors,including:

JamesBurtis

CornellUniversity

5PUBLICATIONS8CITATIONS

SEEPROFILE

JosephB.Yavitt

CornellUniversity

180PUBLICATIONS5,473CITATIONS

SEEPROFILE

Allin-textreferencesunderlinedinbluearelinkedtopublicationsonResearchGate,

lettingyouaccessandreadthemimmediately.

Availablefrom:JosephB.Yavitt

Retrievedon:25August2016

Impact of invasive earthworms on Ixodes scapularis and otherlitter-dwelling arthropods in hardwood forests, central New Yorkstate, USA

James C. Burtis *, Timothy J. Fahey, Joseph B. YavittDepartment of Natural Resources Fernow Hall, Cornell University, Ithaca, NY 14853, USA

A R T I C L E I N F O

Article history:Received 6 May 2014Received in revised form 9 July 2014Accepted 13 July 2014Available online 19 July 2014

Keywords:Exotic earthworm invasionIxodes scapularisLumbricus rubellusLumbricus terrestrisNorthern hardwood forestLitter-dwelling arthropods

A B S T R A C T

Invasive earthworms alter the structure of soils in northern hardwood forests, but the quantitativeimpacts on litter-dwelling invertebrates are unclear. Litter loss should reduce the habitat space, butnutrient-rich earthworm burrows might provide food resources. We investigated the impact of invasiveearthworms on populations of Ixodes scapularis (black-legged ticks) and other litter-dwelling arthropodsto determine the impact of a reduced litter environment. We used five pairs of one-hectare sites(earthworm invaded versus reference) within four separate contiguous forests in New York state. Thepresence of earthworms decreased the density of nymphal I. scapularis by 46.1% and larval I. scapularis by29.3%. We also observed a dramatic decline in the total abundance of litter-dwelling arthropods with69.9% of the arthropod population disappearing in the presence of earthworms. Additionally, litterarthropod populations declined disproportionately to leaf litter mass reduction indicating that thequality of the remaining litter material in the earthworm sites was poor. The impact of earthworminvasion on the litter environment and implications for the position of an important disease vector(I. scapularis) within the litter ecosystem are explored.

ã 2014 Elsevier B.V. All rights reserved.

1. Introduction

Hardwood forests in Canada and the northern United States lacknative earthworms, which were eliminated during the most recentglacial period and have not colonized from further south(James and Hendrix, 2004). As a result, this area is highlysusceptible to invasion by non-native European earthwormspecies. These northern forests have a well-developed forest floor(Gates, 1982; Reynolds, 1994) and reduction of the organic horizonby earthworm feeding activity and mixing has profound effects onforest ecosystem dynamics (Bohlen et al., 2004a), including majorchanges in nutrient cycling (Bohlen et al., 2004b; Hale et al., 2005),and decreases in the biodiversity of plant and animal communities(Frelich et al., 2006; Maerz et al., 2009). Loss of the forest floor has adirect influence on the community of litter-dwelling arthropodsthat rely on leaf litter for food and shelter (Burke et al., 2011;McLean and Parkinson, 2000; Migge-Kleian et al., 2006).Population density of Ixodes scapularis (Say, 1821) (black-leggedtick), is of particular interest, given their role in the transmission of

many tick-borne diseases; including Lyme disease, Babesiosis, andAnaplasmosis (Estrada-Peña and Jongejan, 1999; Keirans et al.,1996; Spielman et al., 1985).

The effect of earthworms on soil and litter-dwellingmicroarthropods, particularly oribatid mites (Burke et al., 2011;McLean and Parkinson, 2000) and Collembola (Eisenhauer et al.,2007; Salmon et al., 2005) has varied. For example, earthwormscan create nutrient-rich microhabitats near their middens,increasing the density and richness of microarthropods (McLeanand Parkinson, 1998; Salmon, 2001), a pattern which appears to beparticularly common in European forests where earthworms arenative (Loranger et al., 1998; Maraun et al., 1999; Migge-Kleianet al., 2006; Wickenbrock and Heisler, 1997). In contrast,Burke et al. (2011) showed that invasive earthworms decreasedthe diversity and abundance of oribatid mites in a hardwood forestin New York state. The same pattern was observed for soilarthropod communities in aspen forests in the Canadian RockyMountains (Cameron et al., 2013; Eisenhauer et al., 2007). Thealteration of microarthropod communities is probably caused bychanges in the structure and chemical composition of the forestfloor following earthworm invasion (Eisenhauer, 2010).

Competition between native millipedes (Diplopoda) andinvasive earthworms may have a negative effect on both groups

* Corresponding author. Tel.: +1 914 356 5282.E-mail address: jb766@cornell.edu (J.C. Burtis).

http://dx.doi.org/10.1016/j.apsoil.2014.07.0050929-1393/ã 2014 Elsevier B.V. All rights reserved.

Applied Soil Ecology 84 (2014) 148–157

Contents lists available at ScienceDirect

Applied Soil Ecology

journal homepage: www.else vie r .com/ locate /apsoi l

(Snyder et al., 2013), but generally the effects of earthworminvasion on other native macroarthropod saprophages (isopodsand diplopods) and predators (Araneae and Chilopoda) are underreported. In the case of arthropod predators invasive earthwormsshould reduce prey availability, but some larger spiders(Nyffeler et al., 2001), and centipedes (Migge-Kleian et al.,2006) are known to feed on earthworms. This could partiallymitigate the effect of reduced prey density and reduce the impactof earthworm invasion on populations of predatory arthropods.Some arthropod predators are also known to feed on I. scapularis inlaboratory settings (Samish and Alekseev, 2001). Although thefrequency of these predation events in nature remains unknown,predator population density may have a strong impact on thedensity of I. scapularis in the field.

The survival and activity of I. scapularis are sensitive to variationin temperature and relative humidity (Bertrand and Wilson, 1996;Stafford, 1994; Vail and Smith, 1998). Therefore, earthwormsshould influence the population density of I. scapularis because,like other litter-dwelling mites, they are dependent upon themicrohabitat provided by the forest floor (Lindsay et al., 1999). Themanual removal of leaf litter has been shown to be an effectivemanagement method for reducing populations of I. scapularis(Schulze et al.,1995). Similar effects have been demonstrated usingcontrolled burns to remove the leaf litter (Stafford et al., 1998), butthe long-term effects of litter reduction or removal on tickpopulations is unknown.

We hypothesized that the density of I. scapularis and otherlitter-dwelling arthropods would be lower in forests with invasiveearthworms. We examined five sites in New York state withvarying leaf litter characteristics and litter-dwelling arthropods toprovide a robust assessment of these hypotheses.

2. Methods

2.1. Site descriptions

This research was conducted in four separate forest sites in NewYork state, three in the Finger Lakes region and one in theAdirondacks. The three Finger Lakes sites were Arnot Teaching andResearch Forest (42!1605900N, 76!3801800W), Hammond Hill StateForest (42!2603300N, 76!1800900W), and Ringwood Preserve(42!2605800N, 76!2200900W). The Adirondack site was in the MooseRiver Plains Wild Forest (43!4600000N, 74!4200000W). At each sitepairs of plots were located in close proximity to each other withearthworms present or absent in one of each pair, and with similarforest composition and soil. All sites were heavily deer browsed,with little to no understory vegetation. There were five pairs ofsites in total (two pairs in Arnot). The reasons for the presence orabsence of earthworms in each site are complex and not fullyunderstood (Suárez et al., 2006; Stoscheck et al., 2012).

The Finger Lakes sites were located on the glaciated AlleghenyPlateau, between 450 and 570 m elevation. The mean temperatureranges from 22 !C in the summer to "4 !C in the winter, withannual precipitation of approximately 1000 mm. There is signifi-cant snow cover in the winter. Soils are acidic Dystrochrepts andFragiochrepts (pH 4.2–5) derived from glacial till (Arnot/HammondHill), or outwash (Ringwood). Forests are mature, second growthAllegheny northern hardwood forests dominated by sugar maple(Acer saccharum), American beech (Fagus grandifolia), easternhemlock (Tsuga canadensis) and a mix of other species. Detaileddescriptions are available elsewhere for Arnot Forest (Fain et al.,1994), Hammond Hill State Forest (Dempsey et al., 2011), andRingwood Preserve (Welbourne, 1979).

Moose River Plains Wild Forest is a 33,321 ha forest owned andmaintained by New York state and management strategies can varywidely. The mean temperature ranges from 18 !C in the summer to

"15 !C in the winter, and annual precipitation is approximately1300 mm. There is significant snow cover throughout the winter.Our sites were located in a maple-basswood rich mesic forest. Thearea was intensively logged from the mid-19th century through themid-20th century, when it was sold to New York state. The soils arederived primarily from glacial till, but they are only slightly acidic(pH 5.0–5.5) owing to the presence of metasedimentary rocks inthe underlying glacial till. The elevation of our sites rangesfrom 550 to 600 m (NYSDEC, 2011). The dominant tree specieswere eastern hemlock (T. canadensis), American basswood(Tilia americana), American beech (F. grandifolia), sugar maple(A. saccharum), and red oak (Quercus rubra).

2.2. Experimental design

There were five pairs of 1 ha plots established within the fourseparate contiguous forests described above. A 64 point samplinggrid was laid out on each of the ten plots. Each point on the grid wasseparated by 15 m. The distance between each pair of plots wasbetween 50 and 200 m; this distance varied depending on theavailability of suitable earthworm invaded areas adjacent tosuitable reference sites. Paired plots were kept at least 50 m apartto limit the movement of ticks hosted on small mammals betweensites.

2.3. Earthworm sampling

Earthworms were collected in October 2012 using the hotmustard extraction method (Lawrence and Bowers, 2002) at eightpoints on each grid. 80 g of powdered mustard seed was placedin 8 L of water and mixed. The mixture was poured into a50 cm # 50 cm area. Earthworms were collected with forceps asthey came out of the soil attempting to escape the irritating effectof the mustard. They were then brought back to the laboratory andstored in a 10% formalin solution until they were sorted by speciesor genus (juveniles) using the available guides (Reynolds, 1977;Dindal and Schwert, 1990).

2.4. Tick sampling

Black-legged ticks were collected in May/June and August/September in 2012 and May/June in 2013 using a drag samplingmethod (Schulze et al., 1997). A 1 m2 white corduroy cloth wasdragged over the leaf litter, and every 30 m the cloth was examinedfor ticks. The cloth was examined for a minimum of 10 min every30 m. Sampling transects were randomly selected and no transectwas sampled more than once every two weeks. Paired sites weresampled on the same day between 9:00 AM and 4:00 PM, as tickactivity is known to show temporal variation (Schulze et al., 2001).In 2012 each pair of sites was sampled a total of eight times (450 m2

per sample), four times in May/June and four times in August/September; three pairs of sites were re-sampled three times inMay/June of 2013 (40,050 m2 total). The spring and late summerdragging periods allowed us to capture the peak activity of thenymphal and larval life stages, respectively. In 2013 nymphs werecollected from the two Arnot sites and Hammond Hill. TheAdirondack and Ringwood sites were unavailable for sampling in2013. Nymphs and larvae were placed in a vial partially filledwith moistened plaster and returned to the laboratory wherethey were identified to species (Keirans and Litwak, 1989;Keirans et al., 1996).

2.5. Litter and arthropod sampling

Litter-dwelling arthropods were collected twice per site in2012, first in early July, and again after leaf fall in late October. Eight

J.C. Burtis et al. / Applied Soil Ecology 84 (2014) 148–157 149

collection points on each site were spread out evenly throughoutthe 1 ha area. At each point a 50 cm # 50 cm sample of the Oiand Oe horizon (hereafter referred to as forest floor) was collectedand a wet weight was recorded. The Oa horizon was absent onearthworm sites and was not collected on reference sites to avoidsampling a different arthropod community. The full litter sampleswere placed under a 40 W light source for seven days in Berlesefunnels which were connected to a vial containing 70% ethanol(Coleman et al., 2004). Samples with a wet litter mass over 400 gwere split up between multiple Berlese funnels. Arthropods werepreserved in ethanol until they could be sorted to order. Once thearthropods were extracted leaf material from each funnel wasseparated from the woody biomass and dried at 65 !C for anadditional 48 h. A dry weight was recorded for the woody biomassand the leaf material from each sample. These numbers werecombined to give the dry weight of the forest floor which wassubtracted from the wet weight yielding the water content for eachsample.

The arthropod samples were sorted into 26 orders using theavailable guides (Gibb and Oseto, 2006; Krantz, 2009; Sierwaldet al., 2014; Thyssen, 2010), and the number of individuals ineach order was recorded. Larvae for the orders Coleoptera,Diptera, Lepidoptera, and Trichoptera were counted separatelyfrom their adult form. We did not classify arthropods intoecological functional groups because many litter arthropods areknown to be generalists (Maraun et al., 2003; Halaj et al., 2005),which combined with an incomplete knowledge of their dietaryhabits makes identifying their functional roles extremelydifficult and unreliable (Krantz, 2009; Setälä, 2002; Setälä andAarnio, 2002). The data from the eight samples on each sitewere combined to estimate the number of individuals per m2.We also divided the number of individuals per sample by the drymass of the litter sample to estimate the number of individualsper gram of litter.

2.6. Statistical methods

All analyses for this study were performed using packages in theR statistical software. Normality and homoscedasticity were testedusing Shapiro–Wilk tests and Bartlett tests, respectively.Arthropod abundance (individuals/100 g of litter) and litter datawere square root transformed. No transformations were applied tothe earthworm or I. scapularis density data. Arthropod density(count/m2) data were also not transformed. Earthworm densitywas analyzed across the five sites with a one-way ANOVA. Litterand arthropod abundance data were analyzed using linear mixedeffects models. Arthropod density data were analyzed using aseries of negative binomial nonlinear mixed effects models. Season(summer/fall) and earthworm presence (earthworm/reference)were the fixed factors included in these mixed models. Site (1–5)

was included as a random factor for all mixed models. All p-valuesfrom the arthropod analyses were adjusted to account for multiplecomparisons using Benjamini and Hochberg’s false discovery ratemethod (Benjamini and Hochberg, 1995). The relationshipbetween litter mass and total arthropod count was determinedusing a linear mixed model with season and earthworm presenceincluded as fixed factors. The data from each tick drag sample oneach site were combined to represent a tick density per 450 m2.The density of larval I. scapularis was analyzed using a linear mixedeffects model with earthworm presence as a fixed factor; fornymphal density year (2012/2013) was also included as a fixedfactor.

To explore the impact of season, site and earthworm presenceon arthropod communities we used NMDS ordinationsconstructed with the “Vegan” package in R. NMDS is a robustordination technique, which is effective for biological communitydata (McCune et al., 2002). We created a dissimilarity matrix withthe Bray–Curtis distance measure and chose three-dimensionalsolutions as the best representation of the dissimilarities based onthe stress values of the ordinations. We constructed one ordinationto explore whether the average community composition of eachsite was clustered by earthworm presence and season. Wedetermined whether the arthropod communities were clusteredsignificantly by earthworm presence and season using a permuta-tional MANOVA, with 1000 permutations of the distance matrix.We built four additional ordinations to determine the effect oflitter variables on the arthropod communities of each ofthe 180 collections. We split the collection data into four groups:(1) “Summer Earthworms”, (2) “Summer Reference”, (3)“Fall Earthworms”, and (4) “Fall Reference”. Environmental vectorswere fit to these ordinations. P-values were determined using1000 permutations of the data. This allowed us to determinewhether the non-woody litter mass, woody biomass, or percentmoisture of the litter collections affected the communitycomposition of each of the four groups differently.

3. Results

3.1. Earthworm density

The mean density of earthworms per m2 across all earthwormsites was 70.2 ($4.52). Six species were identified in ourcollections. There was no significant difference in the total numberof earthworms (F4,35 = 0.703, P = 0.595) across all the earthwormsites (Table 1). The dominant adult species by biomass on all fivesites were Lumbricus terrestris (Linnaeus, 1758), Lumbricus rubellus(Hoffmeister, 1843), Aporrectodea trapezoides (Dugés, 1828), andOctolasion tyrtaeum (Savigny, 1826), and juvenile Lumbricus werevery abundant. Dendrodrilus rubidus (Savigny, 1826) was present inArnot site one and in the Adirondack site. Aporrectodea tuberculata

Table 1The mean number of earthworms collected per m2 at each of the eight earthworm collection points on the five earthworm plots $1 SE (n = 8). Earthworm density did notdiffer significantly (P > 0.05) between sites.

Earthworm taxon Arnot 1 Arnot 2 Ringwood Hammond Hill Adirondacks

Aporrectodea tuberculata 0.00 0.00 2.00 ($) 0.76 0.00 0.00Aporrectodea trapezoides 1.00 ($) 0.65 0.50 ($) 0.50 7.00 ($) 3.44 8.00 ($) 2.00 1.00 ($) 0.65Dendrodrilus rubidus 0.50 ($) 0.50 0.00 0.00 0.00 1.00 ($) 0.65Octolasion tyrtaeum 12.0 ($) 2.39 8.50 ($) 2.20 7.00 ($) 1.25 6.50 ($) 2.61 2.50 ($) 1.05Lumbricus terrestris 5.50 ($) 1.05 2.50 ($) 0.73 7.5 ($) 1.76 2.50 ($) 1.05 5.00 ($) 1.96Lumbricus rubellus 1.00 ($) 0.65 10.5 ($) 2.26 3.50 ($) 1.92 5.00 ($) 1.25 1.50 ($) 0.73Juvenile Lumbricus 21.5 ($) 3.77 24.0 ($) 4.47 26.0 ($) 5.76 38.0 ($) 7.96 16.0 ($) 2.27Juvenile Aporrectodea 5.50 ($) 3.54 3.00 ($) 1.65 14.5 ($) 3.29 17.0 ($) 4.39 13.5 ($) 3.85Juvenile Octolasion 29.0 ($) 10.0 11.0 ($) 3.44 4.00 ($) 2.00 3.50 ($) 1.92 22.5 ($) 5.75Total count: 76.0 ($) 12.4 60.0 ($) 10.9 71.5 ($) 9.75 80.5 ($) 9.60 63.0 ($) 8.20

150 J.C. Burtis et al. / Applied Soil Ecology 84 (2014) 148–157

(Eisen,1874) was present in Ringwood Preserve. All points sampledon the reference sites were earthworm free, with the exceptionof one point at Hammond Hill with an average earthworm densityof 2.5 ($2.5) earthworms per m2.

3.2. Forest floor collections

The forest floor mass was significantly greater in reference sitesthan in sites with earthworms (F1,153 = 15.94, P = <0.001), italso differed between seasons (F1,153 = 35.70, P = <0.001). Thedifferences between earthworm and reference sites were associ-ated with non-woody litter material (F1,153 = 43.98, P < 0.001) as weobserved no effect of earthworm presence on woody litterbiomass. Litter moisture content was not significantly impactedby the presence of earthworms (Fig. 1).

3.3. Arthropod community

In total we counted 113,031 arthropods across all orders fromour Berlese sample extractions. There were 26 taxonomic orderspresent. Orders were further split into 29 categories including thelarval and adult life stages of Coleoptera, Diptera, Lepidoptera, andTrichoptera. Sixteen of these categories were found in at least 50%of our samples. We term these the “common” categories; theimpact of earthworm presence and season were analyzed

(Tables 2 and 3). There were 11 “uncommon” categories whichwere not present in high numbers including: Diptera (adult)(47.5% of samples), Diptera (larvae) (43.8%), Isopoda (6.9%),Geophilomorpha (35.0%), Lithobiomorpha (33.8%), Lepidoptera(adult) (11.9%), Mantodea (0.63%), Opiliones (13.1%), Orthoptera(4.4%), Polydesmida (8.8%), Protura (7.5%), Spirobolida (26.3%), andTrichoptera (larvae) (9.4%). These data were included in theordinations and total arthropod counts, but not independentlyin mixed models.

The overall effects of earthworms on arthropod density werefairly consistent (Table 2) across all orders: many commonarthropod orders were significantly less abundant in the presenceof earthworms. The mean total arthropod density was 4336 per m2

in the reference plots and 1326 per m2 in the earthworm plots.There were no orders that increased in the presence of earth-worms. Of the 16 most common arthropod categories there werefive that were not significantly affected by the presence ofearthworms: Hemiptera, Julida, Psocoptera, Symphypleona, andThysanoptera. The other 11 common categories showed asignificantly lower abundance in the presence of earthworms.The abundance of arthropods in the fall was generally greater thanin the summer (Appendix A).

NMDS ordination was used to evaluate the effects ofearthworms and season on the composition of the litter arthropodcommunity (Fig. 2). The ordination was based on all 26 arthropod

Fig. 1. Bars represent the mean +1 SE (n = 8) of the litter collections per m2 between the five sites split by season and earthworm presence. A1 = Arnot 1, A2 = Arnot 2,HH = Hammond Hill, RP = Ringwood Preserve, and ADK = Adirondacks. Total dry litter was significantly reduced in the presence of earthworms (P < 0.05). The losses wereassociated with leaf material rather than woody material.

J.C. Burtis et al. / Applied Soil Ecology 84 (2014) 148–157 151

orders that were identified (larvae and adults were included in asingle category for these analyses). The ordination reached a stablesolution after 50 runs and had a stress value of 5.49. This indicates astrong relationship as an ordination with a stress value below 10 isconsidered to be a good representation (McCune et al., 2002). TheR2 value was 0.982. The similarity of the arthropod communities oneach site, depending on season and earthworm presence, is shownin an ordination plot in Fig. 2. Using a permutational MANOVAwith 1000 permutations we observed that arthropod communitycomposition was significantly clustered both by season (F1,17 = 5.50,P value = 0.001) and depending on earthworm presence(F1,17 = 12.11, P value = 0.001).

3.4. Density of Ixodes scapularis

In 2012 a total of 129 I. scapularis nymphs were collected; 114 inMay and June, and an additional 15 nymphs in August andSeptember. In 2013, 104 nymphs were collected in May and June.The density data from the May and June collections in 2012 and2013, during the peak of nymphal activity, were used to comparethe density of nymphs between the earthworm and reference sites.In 2012 we collected a total of 654 I. scapularis larvae, 24 in May and

June, and 630 in August and September. The density data from the2012 August and September collections were used to examinedifferences in the larval densities between earthworm andreference sites. There were no I. scapularis detected on theAdirondack sites, so that pair of sites was excluded from theanalyses of tick density. The density of questing nymphs wassignificantly higher on the reference sites than the earthworm sites(F1,41 = 11.80, P = 0.003). The mean density of nymphs was higher inthe 2013 collections than in the 2012 collections (F1,41 = 9.78,P = 0.003). The mean 2012 larval density was also significantlylower in the earthworm sites than in the reference sites(F1,24 = 5.32, P = 0.030) (Fig. 3). Site was not a significant factor inany of our analyses of tick density when the Adirondack sites wereexcluded.

3.5. Litter environment and arthropod community

We constructed four separate ordinations (one for each clustershown in Fig. 2) allowing us to fit environmental variables toarthropod communities (Fig. 4). The final ordination for thesummer earthworm plots had a stress value of 11.3 (R2 = 0.987), forsummer reference a stress value of 6.28 (R2 = 0.996), for fall

Table 2Mean arthropod density per m2 ($)1 SE (n = 80) for the “common” arthropod categories compared across earthworm and reference sites. Total density is the sum of all28 arthropod categories that were collected. Data regarding seasonal differences are in Table 4. Results from negative binomial nonlinear mixed effects models are listed. All p-values have been adjusted for multiple comparisons using Benjamini and Hochberg’s false discovery rate method.

Arthropod taxon Earthworms Reference F-value1,153 Pr

Araneae 33.15 ($) 3.68 78.80 ($) 5.92 22.24 <0.001Coleoptera (adult) 5.80 ($) 1.32 19.75 ($) 2.22 25.93 <0.001Coleoptera (larvae) 8.70 ($) 1.67 27.25 ($) 3.78 16.76 <0.001Entomobryomorpha 274.3 ($) 50.35 527.4 ($) 49.63 9.44 0.003Hemiptera 12.95 ($) 3.00 13.00 ($) 2.49 0.38 0.562Hymenoptera 13.55 ($) 3.05 29.70 ($) 7.02 9.95 0.002Julida 7.45 ($) 2.33 10.00 ($) 2.02 1.23 0.210Lepidoptera (larvae) 11.90 ($) 1.85 20.40 ($) 2.62 10.61 <0.001Mesostigmata 95.60 ($) 9.34 402.3 ($) 35.48 62.03 <0.001Poduromorpha 85.05 ($) 18.03 378.9 ($) 53.74 17.17 <0.001Pseudoscorpionida 14.90 ($) 2.83 34.05 ($) 4.30 12.61 <0.001Psocoptera 15.05 ($) 3.68 24.55 ($) 3.90 2.29 0.060Sarcoptiformes 580.5 ($) 97.77 2458.8 ($) 203.74 33.52 <0.001Symphypleona 72.00 ($) 10.40 98.45 ($) 17.88 0.11 0.775Thysanoptera 16.25 ($) 2.51 27.05 ($) 4.96 1.48 0.184Trombidiformes 14.2 ($) 2.41 68.6 ($) 7.61 69.83 <0.001Total density 1288.0 ($) 156.3 4277.05 ($) 298.7 56.82 <0.001

Table 3Mean arthropod abundance per 100 g of litter collected ($)1 SE (n = 80) for the “common” arthropod categories compared across earthworm and reference sites. Totalabundance is the sum of all 28 arthropod categories that were collected. Data regarding seasonal differences are in Table 5. Results from linear mixed effects models are listed.All p-values have been adjusted for multiple comparisons using Benjamini and Hochberg’s false discovery rate method.

Arthropod taxon Earthworms Reference F-value1,153 Pr

Araneae 5.01 ($) 0.64 8.84 ($) 0.65 12.15 0.049Coleoptera (adult) 0.86 ($) 0.19 2.27 ($) 0.27 23.14 0.027Coleoptera (larvae) 1.10 ($) 0.20 3.04 ($) 0.43 22.37 0.018Entomobryomorpha 37.29 ($) 5.34 60.16 ($) 5.79 7.80 0.074Hemiptera 3.14 ($) 1.28 1.51 ($) 0.29 0.18 0.730Hymenoptera 2.46 ($) 0.66 3.29 ($) 0.72 3.53 0.089Julida 0.99 ($) 0.27 1.34 ($) 0.31 0.99 0.317Lepidoptera (larvae) 2.12 ($) 0.37 2.61 ($) 0.32 6.17 0.089Mesostigmata 13.70 ($) 1.17 45.63 ($) 3.71 21.72 0.027Poduromorpha 10.32 ($) 2.15 38.48 ($) 5.54 10.60 0.059Pseudoscorpionida 1.99 ($) 0.38 3.86 ($) 0.50 7.64 0.078Psocoptera 2.65 ($) 0.66 3.52 ($) 0.60 4.17 0.076Sarcoptiformes 76.78 ($) 11.77 280.92 ($) 23.69 33.74 0.018Symphypleona 13.65 ($) 1.69 11.00 ($) 2.94 0.01 0.912Thysanoptera 2.75 ($) 0.46 3.59 ($) 0.75 0.38 0.645Trombidiformes 2.12 ($) 0.36 8.51 ($) 1.21 53.22 <0.001Total abundance 177.54 ($) 23.54 488.38 ($) 40.34 35.99 0.018

152 J.C. Burtis et al. / Applied Soil Ecology 84 (2014) 148–157

earthworms a stress value of 8.96 (R2 = 0.992), and for fall referencea stress value of 9.98 (R2 = 0.99).

In all cases both site and the mass of non-woody litter had astrong impact on the community composition of arthropods. Wealso found that the amount of woody biomass had a significantimpact on the summer arthropod community in the earthworm

site that was not present in the fall. In the summer percentmoisture (r2 = 0.55, P = 0.001), non-woody litter mass (r2 = 0.403,P = 0.001), and woody litter biomass (r2 = 0.353, P = 0.001) all had asignificant effect on the arthropod community composition onthe earthworm sites. In the summer on reference sites, only on-woody litter mass (r2 = 0.635, P = 0.001) had a significant impact. Inthe fall only non-woody litter mass had a significant impact(r2 = 0.388, P = 0.001) on the earthworm sites. In the fall on thereference sites the percent moisture (r2 = 0.402, P = 0.002) andnon-woody litter mass (r2 = 0.306, P = 0.004) had a significant effecton community composition.

The total arthropod density was positively correlated with littermass in our samples (F1,153 = 44.26, P = <0.001), and to furtherevaluate the effect of earthworms on litter habitat quality wecalculated arthropod abundance per 100 g of litter. The totalabundance of arthropods was significantly lower in the earthwormsites than in the reference sites, with means of 183.5 per 100 g oflitter and 495.9 per 100 g of litter, respectively. The abundanceof six of the arthropod categories were significantly lower in thepresence of earthworms (Table 3), indicating that the changes inabundance exhibited in the density data are not due to litter massreduction alone, but also to changes in the quality of the remaininglitter habitat.

4. Discussion

We observed significantly lower densities of litter-dwellingarthropods and I. scapularis on earthworm-invaded compared toreference forests in central New York state. We also foundthat arthropod abundance, as measured per gram of litter, waslower in earthworm sites than reference sites. This suggeststhat the remaining litter in the earthworm invaded sites was notof equal quality to that in the reference sites and could notsupport the equivalent arthropod population. Although wecannotconclusively ascribe cause and effect, we explore severalpossible mechanisms to explain these density decreasesbelow.

Fig. 2. NMDS ordination of the arthropod communities. Permutational MANOVAresults indicate significant (P < 0.05) clustering by earthworm presence(earthworm/reference) and season (summer/fall). The stress value of the ordinationis 5.49, and R2 = 0.983.

Fig. 3. The bars represent the mean +1 SE for the density of questing larvae and nymphs on the earthworm and reference sites per 450 square meter collection. The density ofquesting nymphs and larvae show a significant (P < 0.05) decrease in the presence of earthworms.

J.C. Burtis et al. / Applied Soil Ecology 84 (2014) 148–157 153

4.1. Earthworms and microarthropods

The reduction of the abundance of litter-dwelling micro-arthropods in the presence of earthworms suggests thatearthworms are altering the litter in ways that impact manytrophic levels within the litter arthropod community. Abundancesof the Collembolan orders, Poduromorpha and Entomobryomor-pha, were significantly less in the presence of earthworms(Table 2), possibly because of the reduced availability of fungi inthese sites (Dempsey et al., 2011), which may decrease thispreferred food source of Collembola (Jørgensen et al., 2005). Theimpact of earthworms was particularly strong on litter-dwellingmites. Mite populations declined both in the number of individualsper m2, and the number of individuals per 100 g of leaf litter(Tables 2 and 3). This decrease indicates that earthworms not onlyreduce habitat availability for mites, but also appear to decreasethe quality of the remaining habitat. This change in habitat qualityis probably explained by alterations of litter structure associatedwith earthworm feeding activity (Eisenhauer, 2010), as well as theselective feeding habits exhibited by many species of epigeic andanecic earthworms that feed preferentially on leaf material with a

low C:N ratio (Belote and Jones, 2009); earthworm taxa (L. rubellusand L. terrestris) from these feeding groups dominated theearthworm community on all the sites (Table 1). In generalmicroarthropod density is strongly dependent on leaf litteravailability (Hasegawa et al., 2013; Sayer et al., 2010; Takeda,1987) and quality (Yang et al., 2007) because many taxa rely onlitter as their primary source of food (Saitoh et al., 2011;Sayer, 2006).

4.2. Earthworms and macroarthropods

The effect of earthworm invasion on macroarthropods was lessconsistent than for microarthropods. Only two orders of macro-arthropods were significantly impacted by the presence ofearthworms after correcting for differences in litter mass (Table 3).The affected orders were Araneae and Coleoptera. Coleoptera is avery diverse order with a wide variety of feeding habits (White,1983), but Araneae are the most common non-flying, obligatearthropod predators. The trophic interactions of litter communitiesare complex and variable (Ponsard et al., 2000), but altering thefood sources of detritivores is known to have a cascading effect on

Fig. 4. Ordinations for the arthropod collections by earthworm presence and season fitted with vectors which show the effect of percent moisture, woody biomass and leafmaterial mass on the arthropod community composition. Longer lines indicate stronger relationships, significant vectors are bolded and denoted with the following symbol(*).

154 J.C. Burtis et al. / Applied Soil Ecology 84 (2014) 148–157

arthropod predators (Chen and Wise, 1999), and that effect mayhave caused the lower abundance of arthropod predators that weobserved in the presence of earthworms.

The effect of earthworms on the macroarthropod communitycould have affected tick populations, which are significantly loweron earthworm than reference plots. The incidence of predation ofI. scapularis by arthropod predators is unknown, although there arespecies of arthropod predators that are known to prey onI. scapularis in a laboratory setting (Samish and Alekseev, 2001).Although Araneae exhibited a significant decrease in the presenceof earthworms, it is possible that those remaining predatorsincreasingly feed on I. scapularis as an alternative food source,thereby reducing their population. Additionally, it has beenobserved that in a system with a simple litter structure arthropodpredators exert a stronger top-down force than they do in a systemwith a complex litter structure (Kalinkat et al., 2013). Experimentalmanipulations of the arthropod food web would be needed toevaluate these possible explanations.

Populations of the most common order of macroarthropoddetritivores, Julida (millipedes; class: Diplopoda), were comparablein the earthworm and reference sites. Other studies havefound significant impacts of earthworm invasion on millipedepopulations both in the lab and in the field, and these impacts havebeen attributed to direct competition (Snyder et al., 2009, 2011,2013). The different responses between mites and millipedes mayhave arisen because mites are more reliant on the presence offinely shredded leaf material (Eisenhauer, 2010), whereas milli-pedes are large enough to shred larger fragments (Snyder et al.,2013). It is also possible that millipedes are able to move betweenthe earthworm and reference sites, given the comparatively largedispersal distances covered by some larger millipedes (Baker,1978;Hopkins and Read, 1992).

4.3. Earthworms and the litter environment

Another mechanism that may contribute to decreased densityof ticks and other arthropods is the impact of earthworms on littermicrohabitat availability. Earthworms are known to reduce andalter the microhabitat availability for many microarthropods,including mites (Eisenhauer, 2010). This is a likely mechanism bywhich earthworms are affecting the density of I. scapularis. There is

evidence that populations of I. scapularis are immediately reducedby the manual removal or reduction of leaf litter (Schulze et al.,1995; Stafford et al., 1998). It is unclear whether there are long-term effects of litter removal on the survival of I. scapularis, but itssurvival is strongly correlated with microclimate suitability(Williams and Ward, 2010; Bertrand and Wilson, 1996). Specifical-ly, I. scapularis survival is negatively impacted by severalconditions, including increased summer temperature, variationsin humidity and higher vapor pressure deficits. Leaf litter mass wassignificantly reduced by the presence of earthworms, matchingobservations in other studies in northern hardwood forests (Haleet al., 2005). We also observed that both litter mass and moisturehad a strong effect on the litter arthropod community as a whole,particularly in the summer (Figs. 3 and 4). It seems likely that thedecrease in litter mass would reduce the availability of humidmicroclimates, but further work on this mechanism is needed.

In conclusion we observed a decrease in the density of nymphaland larval I. scapularis in the presence of earthworms, as well as adecrease in the density of many other litter-dwelling arthropods.Considering the strong impact earthworms had on both the litterenvironment and the arthropod community the decrease in thedensity of I. scapularis is likely due to a combination of changes inthe litter food web, possibly causing an increased rate of predation,and a reduction in microhabitat availability in the leaf litter causedby earthworms. Additionally, the impact of the presence ofearthworm on the arthropod community is likely underestimatedin our data as only the Oi and Oe horizons were collected, while theOa horizon is completely eliminated by earthworms. This studyhighlights our limited understanding of the mechanisms by whichthe litter environment may affect litter-dwelling arthropodsincluding the population density of an important disease vector.Further experimental manipulations of macroarthropod predatorpopulations and the litter environment in a natural setting will beimportant steps towards improving our understanding of theposition of I. scapularis in litter ecosystems.

Acknowledgements

The authors would like to thank Dr. Richard Ostfeld and KellyOggenfuss of the Cary Institute of Ecosystem Studies for lendingthe laboratory and field supplies required for this project. This

Table 5Mean seasonal arthropod abundance per 100 g of litter collected ($)1 SE (n = 80). Seasonal differences and results from linear mixed effects models are listed below. Allp-values have been adjusted for multiple comparisons using Benjamini and Hochberg’s false discovery rate method.

Arthropod taxon Fall Summer F-value1,153 Pr

Araneae 6.8 ($) 0.57 7.04 ($) 0.78 1.04 0.374Coleoptera (adult) 1.63 ($) 0.25 1.51 ($) 0.24 2.50 0.152Coleoptera (larvae) 2.98 ($) 0.43 1.15 ($) 0.20 25.68 <0.001Entomobryomorpha 46.55 ($) 5.19 50.89 ($) 6.19 0.01 0.954Hemiptera 3.46 ($) 1.27 1.18 ($) 0.28 10.26 0.003Hymenoptera 1.44 ($) 0.61 4.31 ($) 0.73 27.10 <0.001Julida 1.10 ($) 0.35 1.23 ($) 0.22 0.46 0.954Lepidoptera (larvae) 0.70 ($) 0.10 4.02 ($) 0.40 107.5 <0.001Mesostigmata 29.22 ($) 3.55 30.11 ($) 2.99 31.72 0.522Poduromorpha 36.61 ($) 5.46 12.19 ($) 2.59 36.75 <0.001Pseudoscorpionida 1.46 ($) 0.24 4.38 ($) 0.55 28.62 <0.001Psocoptera 0.36 ($) 0.07 5.81 ($) 0.77 140.3 <0.001Sarcoptiformes 227.89 ($) 25.94 129.84 ($) 15.15 13.25 <0.001Symphypleona 15.26 ($) 2.45 9.40 ($) 2.31 4.51 <0.001Thysanoptera 2.41 ($) 0.64 3.92 ($) 0.59 8.90 0.005Trombidiformes 7.11 ($) 1.24 3.52 ($) 0.49 11.44 0.001Total abundance 389.84 ($) 68.45 276.07 ($) 46.30 7.49 0.010

J.C. Burtis et al. / Applied Soil Ecology 84 (2014) 148–157 155

research was funded by the Kieckhefer Adirondack FellowshipProgram.

Appendix A

Tables 4 and 5.

References

Baker, G.H., 1978. The distribution and dispersal of the introduced millipede,Ommatoiulus moreletii (Diplopoda: Iulidae), in Australia. J. Zool. 185, 1–11.

Belote, T.R., Jones, R.H., 2009. Tree leaf litter composition and nonnativeearthworms influence plant invasion in experimental forest floor mesocosms.Biol. Invasions 11, 1045–1052.

Benjamini, Y., Hochberg, Y.,1995. Controlling the false discovery rate: a practical andpowerful approach to multiple testing. J. R. Stat. Soc. Ser. B Stat. Methodol. 57,289–300.

Bertrand, M.R., Wilson, M.L., 1996. Microclimate-dependent survival of unfed adultIxodes scapularis (Acari: Ixodidae) in nature: life cycle and study designimplications. J. Med. Entomol. 33, 619–627.

Bohlen, P.J., Scheu, S., Hale, C.M., McLean, M.A., Migge, S., Groffman, P.M., Parkinson,D., 2004a. Non-native invasive earthworms as agents of change in northerntemperate forests. Front. Ecol. Environ. 2, 427–435.

Bohlen, P.J., Groffman, P.M., Fahey, T.J., Fisk, M.C., Suárez, E., Pelletier, D.M., 2004b.Ecosystem consequences of exotic earthworm invasion of north temperateforests. Ecosystems 7, 1–12.

Burke, J.L., Maerz, J.C., Milanovich, J.R., Fisk, M.C., Gandhi, K.J.K., 2011. Invasion byexotic earthworms alters biodiversity and communities of litter- and soil-dwelling oribatid mites. Diversity 3, 155–175.

Cameron, E.K., Knysh, K.M., Proctor, H.C., Bayne, E.M., 2013. Influence of twoexotic earthworm species with different foraging strategies on abundanceand composition of boreal microarthropods. Soil Biol. Biochem. 57, 334–340.

Chen, B., Wise, D.H., 1999. Bottom-up limitation of predaceous arthropods in adetritus-based terrestrial food web. Ecology 80, 761–772.

Coleman, D.C., Crossley Jr., D.A., Hendrix, 2004. Fundamentals of Soil Ecology, Ed. 2Academic Press, Waltham, Massachusetts.

Dempsey, M.A., Fisk, M.C., Fahey, T.J., 2011. Earthworms increase the ratio of bacteriato fungi in northern hardwood forest soils primarily by eliminating the organicmatter. Soil Biol. Biochem. 43, 2135–2141.

Dindal, D.L., Schwert, D.P., 1990. Soil Biology Guide. Wiley, New York, New York.Eisenhauer, N., Partsch, S., Parkinson, D., Scheu, S., 2007. Invasion of a deciduous

forest by earthworms: changes in soil chemistry microflora, microarthropodsand vegetation. Soil Biol. Biochem. 39, 1099–1110.

Eisenhauer, N., 2010. The actions of an animal ecosystem engineer: identification ofthe main mechanisms of earthworm impacts on soil microarthropods.Pedobiologia 53, 343–352.

Estrada-Peña, A., Jongejan, F., 1999. Ticks feeding on humans: a review of records onhuman-biting Ixododiea with special reference to pathogen transmission. Exp.Appl. Acarol. 23, 685–715.

Fain, J.J., Volk, T.A., Fahey, T.J., 1994. Fifty years of change in an upland forest insouth-central New York: general patterns. Bull. Torrey Bot. Club 121,130–139.

Frelich, L.E., Hale, C.M., Scheu, S., Holdsworth, A.R., Heneghan, L., Bohlen, P.J., Reich,P.B., 2006. Earthworm invasion into previously earthworm-free temperate andboreal forests. Biol. Invasions 8, 1235–1245.

Gates, G.E., 1982. Farewell to North American megadriles. Megadrilogica 4, 12–77.

Gibb, T.J., Oseto, C.Y., 2006. Classification of Insects and Mites. Academic Press, SanDiego, California.

Halaj, J., Peck, R.W., Niwa, C.G., 2005. Trophic structure of a macroarthropod litterfood web in managed coniferous forest stands: a stable isotope analysis with@15N and @13C. Pedobiology 49, 109–118.

Hale, C.M., Frelich, L.E., Reich, P.B., Pastor, J., 2005. Effects of European earthworminvasion on soil characteristics in northern hardwood forests of Minnesota, USA.Ecosystems 8, 911–927.

Hasegawa, M., Okabe, K., Fukuyama, K., Makino, S., Okochi, I., Tanaka, H., Goto, H.,Mizogichi, T., Sakata, T., 2013. Community structures of Mesostigmata,Prostigmata, and Oribatida in broad-leaved regeneration forests and coniferplantations of various ages. Exp. Appl. Acarol. 59, 391–408.

Hopkins, S.P., Read, H.J., 1992. The Biology of Millipedes. Oxford: Oxford UniversityPress, 233 p.

James, S.W., Hendrix, P.F., 2004. Invasion of exotic earthworms into North Americaand other regions. In: Hendrix, P. (Ed.), Earthworm Ecology. CRC press, pp.75–88.

Jørgensen, H.B., Johansson, T., Canbäck, B., Hedlund, K., Tunlid, A., 2005. Selectiveforaging of fungi by collembolans in soil. Biol. Lett. 1, 243–246.

Kalinkat, G., Brose, U., Rall, B.C., 2013. Habitat structure alters top-down control inlitter communities. Oecologia 172, 877–887.

Keirans, J.E., Litwak, T.R., 1989. Pictorial key to the adults of hard ticks, familyIxodidae (Ixodida: Ixodoidea), east of the Mississippi river. J. Med. Entomol. 26,435–448.

Keirans, J.E., Hutcheson, H.J., Durden, L.A., Klompen, J.S.H., 1996. Ixodes (Ixodes)scapularis (Acari: Ixodidae): redescription of all active stages, distribution, hosts,geographical variation and medical and veterinary importance. J. Med. Entomol.33, 297–318.

Krantz, G.W., 2009. Habits and Habitats. In: Krantz, G.W. (Ed.), A Manual ofAcarology. Texas Tech University Press, Lubbock, Texas, USA.

Lawrence, A.P., Bowers, M.A., 2002. A test of the ‘hot’ mustard extraction method ofsampling earthworms. Soil Biol. Biochem. 34, 549–552.

Lindsay, L.R., Mathison, S.W., Barker, I.K., McEwen, S.A., Gillespie, T.J., Surgeoner, G.A., 1999. Microclimate and habitat in relation to Ixodes scapularis (Acari:Ixodidae) populations on Long Point, Ontario, Canada. J. Med. Entomol. 36,255–262.

Loranger, G., Ponge, J.F., Blanchart, E., Lavelle, P., 1998. Impact of earthworms on thediversity of microarthropods in a vertisol (Martinique). Biol. Fert. Soils 27,21–26.

Maerz, J.C., Nuzzo, V.A., Blossey, B., 2009. Declines in woodland salamanderabundance associated with non-native earthworm and plant invasions.Conserv. Biol. 23, 975–981.

Maraun, M., Alphei, J., Bonkowski, M., Buryn, R., Migge, S., Peter, M., Schaefer, M.,Scheu, S., 1999. Middens of the earthworm Lumbricus terrestris (Lumbricidae):microhabitats for micro- and mesofauna in forest soil. Pedobiologia 43,276–287.

Maraun, M., Martens, H., Migge, S., Theenhaus, A., Scheu, S., 2003. Adding to ‘theenigma of soil animal diversity’: fungal feeders and saprophagous soilinvertebrates prefer similar food substrates. Eur. J. Soil Biol. 39, 85–95.

McCune, B., Grace, J., Urban, D.L., 2002. Analysis of Ecological Communities. MjMSoftware Design, Gleneden Beach, Oregon.

McLean, M.A., Parkinson, D., 1998. Impacts of the epigeic earthworm Dendrobaenaoctaedra on oribatid mite community diversity and microarthropod abundancesin pine forest floor: a mesocosm study. Appl. Soil Ecol. 7, 125–136.

McLean, M.A., Parkinson, D., 2000. Introduction of the epigeic earthwormDendrobaena octaedra changes the oribatid community and microarthropodabundances in a pine forest. Soil Biol. Biochem. 32, 1671–1681.

Migge-Kleian, S., McLean, M.A., Maerz, J.C., Heneghan, L., 2006. The influence ofinvasive earthworms on indigenous fauna in ecosystems previously uninhabit-ed by earthworms. Biol. Invasions 8, 1275–1285.

Table 4Mean arthropod density per m2 ($)1 SE (n = 80) for the most common arthropod categories. Seasonal differences and results from negative binomial nonlinear mixed effectsmodels are listed below. All p-values have been adjusted for multiple comparisons using Benjamini and Hochberg’s false discovery rate method.

Arthropod taxon Fall Summer F-value1,153 Pr

Araneae 63.80 ($) 5.67 48.15 ($) 5.30 3.46 0.071Coleoptera (adult) 15.20 ($) 2.32 10.35 ($) 1.54 7.30 0.010Coleoptera (larvae) 26.90 ($) 3.80 9.05 ($) 1.65 25.55 <0.001Entomobryomorpha 454.3 ($) 58.69 347.5 ($) 43.44 4.82 0.034Hemiptera 18.90 ($) 3.41 7.05 ($) 1.63 9.02 0.004Hymenoptera 13.45 ($) 5.86 29.80 ($) 4.94 15.26 <0.001Julida 8.40 ($) 1.33 9.05 ($) 2.79 0.03 0.405Lepidoptera (larvae) 6.35 ($) 0.86 25.95 ($) 2.75 76.22 <0.001Mesostigmata 269.9 ($) 35.09 228 ($) 26.43 1.11 0.309Poduromorpha 363 ($) 53.23 101.0 ($) 22.19 32.28 <0.001Pseudoscorpionida 15.00 ($) 2.47 33.95 ($) 4.52 23.75 <0.001Psocoptera 2.7 ($) 0.48 36.90 ($) 4.66 130.1 <0.001Sarcoptiformes 2082.5 ($) 226.3 956.8 ($) 118.9 31.08 <0.001Symphypleona 107.5 ($) 12.60 63.00 ($) 16.15 63.58 <0.001Thysanoptera 18.90 ($) 6.21 24.40 ($) 4.50 6.74 0.012Trombidiformes 58.90 ($) 7.94 23.90 ($) 3.40 25.82 <0.001Total density 3574.7 ($) 340.6 1990.4 ($) 196.7 26.82 <0.001

156 J.C. Burtis et al. / Applied Soil Ecology 84 (2014) 148–157

Nyffeler, M., Moor, H., Foelix, R.F., 2001. Short communication spiders feeding onearthworms. J. Arachnol. 29, 119–124.

NYSDEC, 2011. Moose River Plains Wild Forest unit management plan. New YorkState Department of Environmental Conservation. Lands and Forests PreserveManagement, Albany, New York.

Ponsard, S., Arditi, R., Jost, C., 2000. Assessing top-down and bottom-up control in alitter-based soil macroinvertebrate food chain. Okios 89, 524–540.

Reynolds, J.W., 1977. Key to Sexually Mature Earthworms Found in Ontario. RoyalOntario Museum, Tornoto, Canada.

Reynolds, J.W., 1994. The distribution of the earthworms (Oligochaeta) of Indiana: acase for the post-quaternary introduction theory for Megadrile migration inNorth America. Megadrilogica 5, 13–32.

Saitoh, S., Fujii, S., Takeda, H., 2011. Evaluation of the bottom-up force ofaccumulated organic matter on microarthropods in a temperate forest floor.Eur. J. Soil Biol. 47, 409–413.

Salmon, S., 2001. Earthworm excreta (mucus and urine) affect the distribution ofspringtails in forest soils. Biol. Fert. Soils 34, 304–310.

Salmon, S., Geoffroy, J.J., Ponge, J.F., 2005. Earthworms and collembola relationships:effects of predatory centipedes and humus forms. Soil Biol. Biochem. 37, 487–495.

Samish, M., Alekseev, E., 2001. Arthropods as predators of ticks (Ixodidae). J. Med.Entomol. 38, 1–11.

Sayer, E.J., 2006. Using experimental manipulation to assess the roles of leaf litter inthe functioning of forest ecosystems. Biol. Rev. 81, 1–31.

Sayer, E.J., Sutcliffe, L.M.E., Ross, R.I.C., Tanner, E.V.J., 2010. Arthropod abundance anddiversity in a lowland tropical forest floor in Panama: the role of habitat spacevs. nutrient concentrations. Biotropica 42, 194–200.

Schulze, T.L., Jordan, R.A., Hung, R.W., 1995. Suppression of subadult Ixodesscapularis (Acari: Ixodidae) following removal of leaf litter. J. Med. Entomol. 32,730–733.

Schulze, T.L., Jordan, R.A., Hung, R.W., 1997. Biases associated with several samplingmethods used to estimate abundance of Ixodes scapularis and Amblyommaamericanum. J. Med. Entomol. 34, 615–623.

Schulze, T.L., Jordan, R.A., Hung, R.W., 2001. Effects of selected meteorologicalfactors on diurnal questing of Ixodes scapularis and Amblyomma americanum(Acari: Ixodidae). J. Med. Entomol. 38, 318–324.

Setälä, H., Aarnio, T., 2002. Vertical stratification and trophic interactions amongorganisms of a soil decomposer food web – a field experiment using 15N as atool. Eur. J. Soil Biol 38, 29–34.

Setälä, H., 2002. Sensitivity of ecosystem functioning to changes in trophicstructure, functional group composition and species diversity in belowgroundfood webs. Ecol. Res. 17, 207–215.

Sierwald, P., Bond, J.E., Shear, W.A., Hoffman, R.M., Shelley, R., 2014. Milli-PEET: Keyto millipede orders. The Field Museum of Natural History. Chicago, Illinois, USA.

Snyder, B.A., Boots, B., Hendrix, P.F., 2009. Competition between invasive earth-worms (Amynthas corticis, Megascolecidae) and native North American

millipedes (Pseudopolydesmus erasus, Polydesmidae): effects on carbon cyclingand soil structure. Soil Biol. Biogeochem. 41, 1442–1449.

Snyder, B.A., Callaham, M.A., Hendrix, P.F., 2011. Spatial variability of an invasiveearthworm (Amynthas agrestis) population and potential impacts on soilcharacteristics and millipedes in the Great Smoky Mountains National Park,USA. Biol. Invasions 13, 349–358.

Snyder, Callaham, M.A., Lowe, C.N., Hendrix, P.F., 2013. Earthworm invasion in NorthAmerica: food resource competition affects native millipede survival andinvasive earthworm reproduction. Soil Biol. Biochem. 57, 212–216.

Spielman, A., Wilson, M.L., Levine, J.F., Piesman, J., 1985. Ecology of Ixodesdammini-borne human babesiosis and Lyme disease. Annu. Rev. Entomol. 30,429–460.

Stafford, K.C., 1994. Survival of immature Ixodes scapularis (Acari: Ixodidae) atdifferent relative humidities. J. Med. Entomol. 31, 310–314.

Stafford, K.C., Ward, J.S., Magnarelli, L.A., 1998. Impact of controlled burns onthe abundance of Ixodes scapularis (Acari: Ixodidae). J. Med. Entomol. 35,510–513.

Stoscheck, L.M., Sherman, R.E., Suárez, E.R., Fahey, T.J., 2012. Exotic earthwormdistributions did not expand over a decade in hardwood forest in New Yorkstate. Appl. Soil Ecol. 62, 124–130.

Suárez, E.R., Tierney, G.L., Fahey, T.J., Fahey, R., 2006. Exploring patterns of exoticearthworm distribution in a temperate hardwood forest in south-central NewYork, USA. Landscape Ecol. 21, 297–306.

Takeda, H., 1987. Dynamics and maintenance of collembolan community structurein a forest soil system. Res. Popul. Ecol. 29, 291–346.

Thyssen, P.J., 2010. Key for identification of immature insects. In: Amendt, J., Goff, M.L., Campobasso, C.P., Grassberger, M. (Eds.), Current Concepts in ForensicEntomology. Springer Science, pp. 25–42.

Vail, S.G., Smith, G., 1998. Air temperature and relative humidity effects onbehavioral activity of blacklegged tick (Acari: Ixodidae) nymphs in New Jersey. J.Med. Entomol. 35, 1025–1028.

Welbourne, M.L., 1979. Nutrient distribution and tree growth with respect to slopeposition and wind action on a glacial outwash ridge in central New York.Master’s Thesis. Cornell University, Ithaca, New York.

White, R.E., 1983. A Field Guide to the Beetles of North America. Houghton Mifflin,Boston, Massachusetts.

Wickenbrock, L., Heisler, C., 1997. Influence of earthworm activity on the abundanceof Collembola in soil. Soil Biol. Biochem. 29, 517–521.

Williams, S.C., Ward, J.S., 2010. Effect of Japanese Barberry (Ranunculales:Berberidaceae) removal and resulting microhabitat changes on Ixodes scapularis(Acari: Ixodidae) abundances in Connecticut, USA. Environ. Entomol. 36,1911–1921.

Yang, X., Warren, M., Zou, X., 2007. Fertilization responses of soil litter fauna andlitter quantity quality, and turnover in low and high elevation forests of PuertoRico. Appl. Soil Ecol. 37, 63–71.

J.C. Burtis et al. / Applied Soil Ecology 84 (2014) 148–157 157