Compounding negative effects of leaf litter absence and belowground competition from an invasive spring ephemeral on native spring ephemeral growth and reproduction

Restoring invaded ecological communities requires understanding whether native declines are driven directly by invader competition versus changes to abiotic conditions that are associated with invader presence. In forest understory communities, declines in spring ephemerals could result from invasion or alternatively by loss of leaf litter layers, from which native spring ephemerals evolved to emerge annually. Here, we investigate how native spring ephemerals (Erythronium spp.) are affected by leaf litter absence and direct competition with the functionally similar invader lesser celandine (Ficaria verna). The absence of a litter layer alone reduced Erythronium asexual reproduction (corm biomass), with reductions magnified when celandine was also present. Celandine impacts occurred despite celandine having limited aboveground growth under common garden conditions, indicating belowground competition with celandine is sufficient to elicit substantial impacts on Erythronium. This reduced clonal growth would eventually hinder Erythronium sexual reproduction because flowering only occurs in large individuals. Both Erythronium species responded similarly to experimental conditions: in response to shading from litter, Erythronium produced larger, heavier petioles and invested less in leaf blade tissue as a proportion of total shoot tissue. Though reduced leaf blade investment was correlated with reduced corm growth, the net benefits of litter on growth outweighed any negative effects of this biomass allocation tradeoff. These results demonstrate how, although direct competition from invasive plants can impact co-occurring natives, invader-induced changes to local environmental conditions also have important indirect effects. Management to support native spring ephemerals should include preventing losses of or restoring depleted forest litter layers as well as reducing invasive competition.

Restoring invaded ecological communities requires understanding whether native declines are driven directly by invader competition versus changes to abiotic conditions that are associated with invader presence. In forest understory communities, declines in spring ephemerals could result from invasion or alternatively by loss of leaf litter layers, from which native spring ephemerals evolved to emerge annually. Here, we investigate how native spring ephemerals ( Erythronium spp.) are affected by leaf litter absence and direct competition with the functionally similar invader lesser celandine ( Ficaria verna ). The absence of a litter layer alone reduced Erythronium asexual reproduction (corm biomass), with reductions magnified when celandine was also present. Celandine impacts occurred despite celandine having limited aboveground growth under common garden conditions, indicating belowground competition with celandine is sufficient to elicit substantial impacts on Erythronium. This reduced clonal growth would eventually hinder Erythronium sexual reproduction because flowering only occurs in large individuals . Both Erythronium species responded similarly to experimental conditions: in response to shading from litter, Erythronium produced larger, heavier petioles and invested less in leaf blade tissue as a proportion of total shoot tissue. Though reduced leaf blade investment was correlated with reduced corm growth, the net benefits of litter on growth outweighed any negative effects of this biomass allocation tradeoff. These results demonstrate how, although direct competition from invasive plants can impact co-occurring natives, invader-induced changes to local environmental conditions also have important indirect effects. Management to support native spring ephemerals should include preventing losses of or restoring depleted forest litter layers as well as reducing invasive competition.

functional traits, root competition, invasion impact, ecological restoration, Erythronium, Ficaria verna

Declines in native understory plant communities have occurred and are predicted to continue due to invasion, herbivore pressure, land use change, and climate change (Augspurger and Buck 2017; Miller and Gorchov 2004, Bialic-Murphy et al. 2020; Lee et al. 2024). When components of the forest understory are replaced by near-monocultures of invasives, competition from newly dominant species is usually presumed to be the cause, even though alternative drivers can spur native declines while simultaneously creating opportunities for invasive species establishment (MacDougall and Turkington 2005). Restoring such degraded communities requires an accurate understanding of the factors that actually led to native declines in the first place. Spring ephemerals constitute just 3-4% of all native temperate forest understory herb species (Yancy et al. 2024; Spicer et al. 2020) but are important because they support specialist pollinators and reduce nutrient losses at a time of year when vegetation cover is low and leaching and runoff rates are correspondingly high (Muller and Bormann 1976; but see Rothstein 2000). One potential driver of spring ephemeral decline whose importance may be overlooked is the loss of forest litter layers. In deciduous forests of eastern North America, persistent litter has been reduced due to long-term invasion by non-native earthworms (Frelich et al. 2019; Reed et al. 2023) and shrubs, the latter of which commonly have nutrient-rich, quickly decomposing leaves (Arthur et al. 2012; Ashton et al. 2005; Jo et al. 2017). Native spring ephemerals in this region evolved to emerge through deep litter layers in early spring (Muller 1975, 1978), a strategy with both expected benefits and costs. Deep litter layers cast shade and are physical barriers to emergence for some species (Sydes and Grime 1981). Yet, leaf litter benefits plants by increasing moisture retention (Donath and Eckstein 2008, Dion et al. 2017) and reducing frost damage by buffering soil temperatures (Sydes and Grime 1981, Baltzinger et al. 2012; Loydi et al. 2014; Tessier 2022). Leaf litter also affects nutrient availability, either directly by increasing supply (Loydi et al. 2014) or indirectly by facilitating fine root growth (Fisk et al. 2004). Overall, we expect net positive effects of litter layer presence, and thus negative effects of its absence, on spring ephemeral growth and survival in most cases (e.g., Baltzinger et al. 2012; Dion et al. 2017). Here, we investigate how native spring ephemerals are affected by direct competition with a functionally similar invader and by leaf litter absence. The invader in question, lesser celandine ( Ficaria verna, syn. Ranunculus ficaria, hereafter celandine), occurs in large, dense patches across deciduous forests of eastern North America (Axtell et al. 2010). Though introduced in 1867, it has only recently (ca. 21 st century) colonized large areas and thus raised concerns over a potential ability to competitively exclude natives (Axtell et al. 2010). Like Erythronium, celandine is also a spring ephemeral which reproduces both sexually and asexually (Axtell et al. 2010; Mattingly et al. 2023). This phenotype is exceedingly rare among invasive plant species. Because celandine dies back entirely aboveground after late spring senescence, its negative impacts may be somewhat limited, despite the extremely high abundances it can reach during peak growth in spring. Three previous studies have examined celandine impacts on native performance, finding mixed results over the short-term. Celandine presence led to decreased biomass of a co-occurring native grass (Masters and Emery 2016) and reduced lifespan and seed set of a native annual forb (Cipollini and Schradin 2011). However, it is associated with increased seed maturation and pollinator visitation rates in co-occurring Claytonia virginica, presumably due to higher pollinator activity in invaded plots (Masters and Emery 2015a). The magnitude and direction of invader impacts by celandine (or any other invader) on resident natives will depend on the species and outcome measured, so additional study is needed to examine its full range of ecological effects. Two species of North American native ephemerals, Erythronium albidum and E. americanum (Liliaceae) , commonly co-occur with celandine in invaded sites (Masters and Emery 2015a; the authors, personal observations). This genus represents 24-30 species of long-lived perennials reproducing primarily asexually through ephemeral subterraneous runners (Fig. 1) that each forms a corm at its terminus (Muller 1978). The parent corm and all runners disintegrate before summer dormancy, with offspring corms persisting (Muller 1978; Fig. 1.). Individuals that do not reproduce asexually simply replenish their corm, which then persists. Between senescence in late spring and root renewal in early fall, Erythronium remain dormant (Muller 1979, 1975). Erythronium corms occur at relatively low densities in comparison to celandine’s high density of roots and rhizomes (the authors, personal observations), suggesting belowground competition may influence celandine competitive effect magnitudes. Competition for light likely also occurs, because celandine leaf-out precedes Erythronium emergence (Frey and Schmit 2017), reducing soil surface light availability by up to 98% (the authors, unpublished data). However, Erythronium light interception is not obviously reduced by celandine because when they co-occur Erythronium leaf blades often rise above celandine shoots. Erythronium leaf blades are supported by a single petiole, which is elongated by up to three times in plants emerging through celandine or leaf litter relative to plants emerging in bare soil (unpublished data, Fig. 1c). Thus, although aboveground competition may be minimal , Erythronium may still incur costs from altering biomass allocation to produce a longer petiole. S hoot growth consumes stored corm resources in the spring before leaf blade development (Muller 1979), and if Erythronium invests more resources into petiole elongation when growing amongst celandine, fewer resources may be allocated to leaf blade ( Erythronium ’s main photosynthetic tissue). This tradeoff could ultimately result in fewer resources returned to the corm for future growth and reproduction. Lastly, celandine competition may affect Erythronium emergence and flowering phenology, which has performance implications given Erythronium’s short period of annual aboveground activity. In other systems, interspecific competition induces earlier flowering (Palacio-Lopez et al. 2020), and if similar patterns occur in Erythronium then pollinator efficacy and thus its long-term population viability may be impacted. This study investigates how Erythronium albidum and E. americanum performance over a full growing season are affected by competition from lesser celandine and litter layer absence. We hypothesized that although aboveground interference from celandine and leaf litter may be similar in some ways, the independent effects of invasive celandine and the absence of litter on Erythronium would both be negative. We thus made the following predictions: (1) The absence of litter will reduce Erythronium performance, which will be further reduced when celandine is also present. (2) Shading from celandine and from litter will cause Erythronium to increase petiole and reduce leaf blade investment; we expect this will in turn reduce corm growth. (3) Litter presence and celandine presence will each delay Erythronium leaf blade and floral development.

Plant material We purchased 189 Erythronium corms from commercial suppliers: 89 E. albidum (Prairie Moon Nursery, Winona, Minnesota, USA) and 100 E. americanum (Tennessee Wholesale Nursery, Altamont, Tennessee, USA). Erythronium americanum and E. albidum corms arrived on 27 and 27 October 2022, respectively. In November 2022, we harvested actively growing celandine bulbils from a population within deciduous riparian forest in Columbus, Ohio, USA. All plant material was stored for ≤3 days at 4⁰C before transplanting. Experimental setup and growth conditions We planted individual Erythronium corms 4cm below the soil surface depth into 1L pots. This depth is common for established Erythronium in the field (Muller 1975). Pots contained an equal mixture of coarse sand (Paver Base, Quikrete, Atlanta, Georgia, USA) and propagation mix (LM-AP All Purpose Mix, Lambert, Rivère-Ouelle, Quebec, Canada). We also incorporated slow-release fertilizer pellets according to manufacturer specifications (18g fertilizer/3.8L soil; Osmocote 14-14-14 Controlled Release Fertilizer, ICL Group, Tel-Aviv, Israel). In November 2022, we buried all but the top 2cm of each pot into a raised garden bed (7.4 × 1.3m) on The Ohio State University campus (Columbus, Ohio, USA). To compare light conditions within the garden to those in a nearby forest, we measured photosynthetically active radiation (PAR) at the soil surface on 28 February 2023 using a LI-190R Sensor (LI-COR Biosciences, Lincoln, NE, USA). PAR in the garden was 330 ± 22.9 (mean ± SEM) μmol m -2 s -1, in a shady section of the forest was 177 ± 15.0, and in a gap in the forest was 378 ± 19.3. To simulate reduced light availability due to a fully closed forest canopy in mid-spring we covered all pots with a shade cloth 3m above the soil on 23 April 2023 (light reduction 71.4 ± 2.5%). Within the garden, each Erythronium was stratified into one of six rows based on initial corm biomass such that each row had an equal distribution of similarly sized plants of each species in each randomly-assigned treatment (leaf litter, celandine, or bare soil). For the leaf litter treatment (n=30 E. albidum and 33 E. americanum ), we placed 5cm of mixed red oak ( Quercus rubra ) and red maple ( Acer rubrum ) leaves on the soil surface. For the celandine treatment (n=30 E. albidum and 34 E. americanum ), we planted 5 (± 0.04) grams of recently sprouted celandine bulbils within the top 2cm of soil. To simulate a forest floor without a litter layer, we grew the remainder (n=29 E. albidum and 33 E. americanum ) in bare soil. Sample sizes differed slightly due to Erythronium availability . In late December 2022, we realized some of the intended celandine bulbils were misidentified Viola sororia rhizomes. We removed all such rhizomes and replaced them with freshly field-collected sprouted celandine on 31 December 2022 and 1 January 2023; across all pots in this treatment, the retained celandine was also removed, weighed, and re-planted to ensure all replicates had a similar amount of disturbance and equivalent celandine biomass (7 ± 0.04 grams, an increase over the original 5g to account for root growth in sprouted celandine bulbils). Because Erythronium corms had been planted 2cm below celandine in each pot (at 4cm depth), no Erythronium was disturbed during this process. To investigate how our treatments may affect soil temperatures at corm depth we filled twelve 2.9L pots with the soil mix described above and added them to the common garden on 31 December 2022. Each pot was randomly assigned to the three treatments (n=4 each), buried to the same depth as the main experiment pots. The bare soil was not modified further, the celandine treatment received 7g of live celandine bulbils planted within the top 2cm of soil, and the litter treatment received a 10cm deep mixture of red oak and red maple leaves atop the soil surface. Temperature within each pot was recorded by a HOBO MX2201 pendant logger (Onset Computer Corporation, Pocasset, Massachusetts, USA) buried at corm depth (4cm below the soil surface). Temperature was recorded at hourly intervals between 12 January and 9 May 2023. Data collection Growth and reproduction Immediately before planting, fresh biomass of each Erythronium corm was weighed (range 0.20-1.64g, mean 0.56 ± 0.20 g). Because plants were dormant or had just become active by planting time, these values represent initial belowground biomass. Within treatments, each Erythronium was randomly assigned at planting to one of two harvests: shoot maturity (n=94) or post-senescence (n=95). The shoot maturity harvest enabled us to investigate biomass allocation patterns before senescence, when shoot biomass would diminish. Shoot maturity was evidenced by a slight yellowing of the leaf tip (Muller 1978). For this harvest, aboveground dry biomass was the aboveground petiole (hereafter petiole) plus leaf blade, weighed separately. Biomass of the belowground portion of the petiole (from the soil surface to where the petiole connects to the corm) was excluded from analyses because it was not a component of our central hypothesis. Belowground dry biomass at shoot maturity consisted of the parent corm plus any runners, weighed separately. We separated root biomass from runner biomass, but for Erythronium grown with celandine, roots could not be separated by species. This is unlikely to affect our analyses, because for 58 of 62 Erythronium grown in litter or bare soil treatments roots composed less than 2% of total belowground biomass. The post-senescence harvest enabled us to determine total corm resource acquisition (or expenditure) during the Erythronium growing season. We conducted this harvest between 3 and 8 June 2023, at which point each Erythronium shoot had fully senesced, signifying the beginning of summer dormancy. By this point, only dormant offspring corms were alive, and all other plant material (including parent corm, runners, and all aboveground biomass) had senesced. The fresh weight of each individual corm was recorded. Dry weight was not obtained because these plants were retained for future experiments. Offspring number at shoot maturity was quantified by counting the parent corm (1) or the total number of runners attached to it. If no runners were present, the parent corm was recorded as the only offspring because the Erythronium parent corm replaces itself when no runners are present by forming a new corm slightly below the old corm (Muller 1975, 1979). When runners are present, however, reserves from the parent corm are transferred to new corms or runners; in this case, the parent corm becomes depleted and dies at the end of the season (Muller 1979, 1975). Each runner attached to the parent corm therefore indicates future production of an offspring corm. Offspring number at the post-senescence harvest was quantified by counting the total number of corms; at this point all parent corms had died and only offspring corms remained. Relative growth calculation \begin{equation} \log_{10}(estimated\ dry\ biomass)=\ {{(log}_{10}\left(\text{fresh\ biomass}\right)\ \times 1.01124)-0.67210)}\nonumber \\ \end{equation} With this estimate, we could assess relative growth on a dry weight scale. Hereafter, initial biomass refers to these estimated dry biomass values. Relative growth for each Erythronium was calculated as final dry biomass minus initial biomass, divided by initial biomass. We estimated initial biomass of the corm as described above; however, because of phenological differences between the two harvests, final biomass was quantified using different methods. For the shoot maturity harvest, all aboveground and belowground biomass was dried and weighed. For the post-senescence harvest, final biomass was estimated from the equation above (since only new corms remained at this point). In all cases, relative growth <0 signifies a net biomass loss. To ensure our inferences were robust to how relative growth was estimated, we also calculated relative growth using log response ratios. Results did not differ, so we present only the former analyses here. Biomass allocation At the shoot maturity harvest, leaf blade and petiole lengths and widths were measured to estimate leaf area and petiole volume and then dried at 60℃ and weighed. Petiole biomass data were lost for 23 individuals, so they were dropped from biomass allocation analyses. We approximated leaf blade area as the area of an ellipse and petiole volume as the volume of a right cylinder. To investigate how treatments affected tissue construction, we calculated specific leaf blade area (g/cm 2 ) and petiole biomass per unit volume. We quantified investment in leaf blade tissue as a proportion of total aboveground biomass (leaf blade biomass divided by total shoot biomass [leaf blade + petiole]). We expected this measure, which we refer to as leaf blade investment, would decline with increased relative allocation to petiole biomass. Lastly, we quantified belowground biomass allocation as root mass ratio (belowground biomass divided by total biomass). Phenology Erythronium shoot development was monitored every 3-10 days between the first emergence (3 January 2023) and shoot maturity of all but one individual (6 April 2023). For each individual, the dates of shoot emergence, leaf maturation, flower emergence, and petal opening were recorded. Shoot emergence was defined as the first date upon which the shoot had emerged from cover (i.e., litter, celandine, or soil). Leaf maturation occurred when the leaf was fully expanded as denoted by a convex shape of the blade (Muller 1978, 1979) accompanied by advanced development of chlorophyll (leaves visibly changed color from mottled brown and green to solid bright green). Flower emergence occurred when any part of the inflorescence was visible (usually the emergence of the peduncle). When the individual petals of the inflorescence were distinguishable, petal opening was recorded. From these dates, we derived four phenological responses of interest. Shoot emergence was recorded as the day of year that event occurred. Days to flower emergence was the number of days between shoot emergence and flower emergence, and petal opening period was the number of days between flower emergence and petal opening. Similarly, leaf maturation period was the number of days between shoot emergence and leaf maturation. We assessed the timing of celandine aboveground growth and thus its potential to influence abiotic conditions and Er y thronium growth or phenology. Absolute cover of celandine shoots and absolute cover of celandine shoots at or above the height of Erythronium were visually estimated at 7-10 day intervals between 17 January and 6 April 2023. Celandine was the only vegetation in the pot other than Erythronium, and the only Erythronium biomass at the soil surface was its petiole base (<1mm diameter). Statistical analysis All analyses were executed using R version 4.2.2 (R Core Team, 2022). We calculated inferential statistics using type II sums of squares in the package “car” (Fox and Weisberg, 2019), incorporated random effects as needed using “lme4” (Bates, 2015), and conducted post-hoc pairwise comparisons using “emmeans” with false discovery rate (FDR) adjustments (Lenth, 2023). Unless otherwise indicated, we present estimates of variability as standard error of the mean. Relative growth, offspring number and biomass, and all biomass allocation metrics were compared among treatments using linear models with treatment, species, and the treatment × species interaction as fixed factors and initial biomass as a covariate to account for variation in starting corm size. To investigate if relative growth was moderated by leaf investment, we fitted a linear model with leaf investment, treatment, species, the treatment × species interaction, and initial biomass as predictors. Offspring weight distributions were compared across treatments using two-sample Kolmogorov-Smirnov tests. All these analyses were run separately by harvest. Shoot emergence date, leaf maturation period, flower emergence, and petal opening period were included as response variables in linear models with planting treatment, species, the treatment × species interaction, and initial biomass as predictors. We also assessed the probability of flowering as a function of these predictors using binomial regression with a logit link function. Because only one E. americanum in the bare soil treatment had a recorded petal opening, the treatment × species interaction effect on petal opening could not be included in the model for petal opening period. For each temperature logger, the range, minimum, and maximum temperatures were calculated for each day of deployment. These daily temperature metrics were included in linear models with planting treatment as a main effect and day and logger as random effects.

The vast majority (94%, 177 of 189) of our planted Erythronium survived. Erythronium americanum and E. albidum responded similarly to all treatments ( P> .12 for the effect of species and interactions with species across all analyses), so for simplicity we omit mention of these similarities for the remainder of the Results (see Supplemental Information for figures showing species-specific responses). Treatment Effectiveness Leaf litter presence affected temperature fluctuations, buffering daily temperature range (Fig.S1a) and maxima (Fig. S1b; P <.01 for both responses), although treatments did not affect daily minima (Fig. S1c; P =.14) nor averages (Fig. S1d; P =.79). Daily temperature ranges between bare soil and celandine treatments did not differ but were >50% greater in magnitude than temperature ranges in litter (in bare soil, 6.09 ± 0.18℃; in celandine, 6.26 ± 0.19℃; and in litter, 3.94 ± 0.12℃; P <.01; Fig. S1a). Daily maxima, similarly, did not differ between bare soil and celandine treatments but were 12-15% higher than those in litter (bare soil, 11.9 ± 0.34℃; celandine, 12.1 ± 0.35℃; and litter, 10.5 ± 0.29℃; FDR P <.01; Fig. S1b). By the end of Erythronium ’s active growth phase, celandine growth aboveground was less developed than in nearby field populations, presumably due to our relatively late re-planting date. Mean celandine cover at the time of Erythronium emergence was 38.8 ± 0.69%, whereas celandine cover at or above the height of Erythronium was just 11.3 ± 0.36%. This likely resulted in weaker aboveground interactions between celandine and Erythronium relative to field conditions. Growth and reproduction Erythronium growth in litter and bare soil was always greater than when grown with celandine, though differences between litter and bare soil treatments depended on Erythronium ’s developmental stage (Fig. 2; Table 1). At shoot maturity, Erythronium in litter and bare soil had on average nearly tripled in biomass relative to initial biomass (litter treatment final 0.339 ± 0.033g versus initial 0.127 ± 0.011g; bare soil final 0.302 ± 0.025 versus initial 0.115 ± 0.010g), an increase that did not differ between them (FDR P= 0.830) but was greater than when grown with celandine (final 0.208 ± 0.018g versus 0.113 ± 0.011g initial; Fig 2a). Post-senescence, however, relative growth in litter was nearly twice as high as in bare soil (FDR P< .01; Fig. 2b; Table 1), and Erythronium in celandine had experienced a net loss of corm biomass (Fig. 2b). Reproductive output when measured as offspring number was not affected by treatment at either harvest (Table 1). However, total reproductive biomass (Table 1) and per-corm biomass post-senescence were both affected, as illustrated by comparing distributions of individual offspring corm weights (Fig. 3). Nearly every Erythronium offspring corm was small (<0.05g) when parent plants were grown with celandine, whereas the other two treatments enabled the production of larger corms (Kolmogorov-Smirnov P <.001 for both pairwise comparisons). These distributions were similar for plants growing in bare soil or with litter ( P =0.092). Erythronium total offspring biomass was more than halved by celandine presence and decreased by 30% in the absence of litter (litter treatment 1.170 ±0.087; bare 0.821 ± 0.069; celandine 0.456 ± 0.045g; FDR P <.001 for both comparisons; Table 1). Biomass allocation Erythronium in litter had heavier petioles than when growing in either celandine or bare soil (Fig. 4a), with similar petiole biomass in the latter two treatments (FDR P= .864; Table 2), and similar leaf blade biomass across treatments (Fig. 4b). In contrast, petiole biomass per unit volume trended higher in celandine than in bare soil (FDR P =.074; overall treatment effect P= 0.034), though it did not differ between litter and any other treatment (Fig. 4c). Neither specific leaf area nor root mass ratio differed among treatments (Figs. 4d, 4f; Table 2). The increase in Erythronium petiole biomass when grown in litter, coupled with a lack of treatment effects on leaf biomass, also meant that leaf blade investment (blade biomass as a proportion of total shoot biomass) was lower in leaf litter than in bare soil (FDR P <.05) and trended lower than in celandine (9.3% and 6.9% reductions relative to bare soil and celandine, respectively; Fig. 4e; Table 2). Increased leaf blade investment was associated with greater corm growth, as expected, although this relationship was marginally significant (β=4.280; P= .064; model R 2 =0.408; treatment and initial biomass main effects both P< .01; Fig. 5). We found no support for this relationship varying by treatment. Phenology and Flowering Leaf litter absence hastened Erythronium shoot emergence by 14-16 days relative to other treatments but prolonged subsequent leaf maturation by the same duration (Table 3; Figs. 6a, 6b). A one standard deviation (0.059g) increase in initial biomass was associated with a 16-day earlier shoot emergence but a 5-day longer leaf maturation period (β=-275.93 and 83.17, respectively; P <.0001; Figs. 6a, 6b). Days to flower emergence was hastened by 14-15 days when litter was absent (a marginally significant effect; Table 3, Fig. S2a). Petal opening period was also prolonged by 14-16 days when litter was absent relative to both other treatments (marginally significant; Table 3, Fig. S2b). Flowering phenology was unrelated to initial biomass (Figs. S2c, S2d). Of the 179 surviving Erythronium, only 12% flowered. Erythronium with larger initial biomass were more likely to flower ( P .14). With only one exception, all flowering Erythronium had an initial biomass ≥ 0.196g (Fig. 3; Fig. S2c). Of all offspring corms produced by the end of our experiment, 34% of those whose parental corm was grown in litter grew to be larger than this putative biomass threshold, whereas only 9% of corms in bare soil and 2% (one individual) in celandine did (Fig. 3), suggesting sexual reproduction the following year would be much more likely where litter occurred and almost nonexistent where litter was absent and celandine present.

We document clear reductions in Erythronium performance by leaf litter absence alone, a treatment designed to simulate long-term forest invasion from earthworms and non-native shrubs with rapidly decomposing leaves. These performance declines were even greater when litter absence was coupled with direct competition from celandine. Thus, although direct invader competition impacts co-occurring natives, invader-induced changes to local conditions also have important indirect effects – in this case, with implications for sexual reproduction and thus long-term population viability. These results highlight the importance of a dual management strategy, ideally reducing celandine while also restoring leaf litter in forests where spring ephemerals occur. Direct belowground competition with invader reduces native growth and fecundity Existing studies have found reduced fecundity in natives grown amongst celandine (Cipollini and Schradin 2011, Masters and Emery 2015a). We show celandine also reduces corm growth and asexual reproduction in long-lived natives. Though celandine presence did not impact Erythronium offspring number, celandine did reduce Erythronium total asexual reproductive biomass. (Figs. 2b and 3). Belowground competition was probably the major driver of these declines because by Erythronium shoot maturation, celandine roots occupied most of the belowground pot volume, despite leafing out several weeks after Erythronium emergence and never growing large enough to shade it. Also, aboveground biomass of celandine at harvest was not correlated with any metric of trout lily performance (data not shown). The negative effects of celandine on Erythronium reported here are thus conservative estimates of its impact in the field, where belowground competition is likely to be even more pronounced because of accumulated celandine root biomass, and where aboveground competition would be expected because celandine leaf-out precedes that of Erythronium by several weeks, leading to crowding aboveground. Our results highlight the importance of documenting belowground competition effects, which for herbaceous species are generally greater than aboveground competition (reviewed by Kiær et al 2013). Erythronium flowering likelihoods were greater in plants that began the growing season as larger corms, consistent with prior observations (Muller 1979). Celandine-induced reductions in Erythronium offspring size prior to summer dormancy (which approximates autumnal biomass for the next year) would thus presumably reduce future flowering (see Fig. 3). In combination with widespread leaf litter losses, our results suggest that celandine invasions reduce Erythronium flowering likelihoods to nearly zero. This has clear implications for the long-term population viability of Erythronium and the specialist pollinators they support. Although these inferences do contrast with a lack of treatment effects on flowering probability reported above, we attribute this latter result to the fact that flowering during our experiment was influenced primarily by initial corm biomass (before treatments began), rather than by growth during the experiment. Though Erythronium survival across treatments was high in our one-year experiment, this duration may have been insufficient to detect declines in such long-lived perennials. Miller and Gorchov (2004) made a similar argument about the potential negative effects of invasive Lonicera maackii on native forest herbs, despite detecting no direct effects on survival. Longer-term experiments and demographic studies (e.g., Bialic-Murphy et al. 2019) are needed to substantiate or refute inferences about invasion impacts from shorter-term experiments. Leaf litter absence reduced native performance independently of invader competition Our finding that leaf litter presence greatly benefitted Erythronium reflects the effects of litter alone, independent from factors that are often confounded with litter dynamics in field studies (e.g., earthworm presence and the composition of neighboring plant communities; Wood and Bornstein 2011). These are likely underestimates of the true benefits to Erythronium under field conditions, where larger expanses of litter should cause greater effects than documented in pot experiments. Also, our relatively short-term experiment may not have fully captured effects of microbial activity associated with leaf litter. Though microbial effects can develop within months (Habtewold et al. 2020), those in uninvaded or newly restored forests develop over decades (Sun and Badgley 2019, Banning et al. 2011; Sun et al. 2017). The mechanisms by which leaf litter affected conditions in our experiment included reductions in diurnal temperature maxima and fluctuations (Fig. S1), adding to a growing body of evidence for its temperature buffering effects (Synes and Grim 1981, Donath and Eckstein 2008, Loydi et al. 2014; Dion et al. 2017). Litter presence also increases moisture retention (Donath and Eckstein 2008), and reduced temperature maxima could contribute to reduced moisture losses from evapotranspiration. Where leaf litter remains, such effects could lessen the impact on understory flora of late-spring heatwaves, which are increasing in frequency and intensity due to climate change (Calvin et al. 2023). The temperature dynamics and the changes in Erythronium phenology we observed under litter show strikingly similar patterns (Figs. 6, S1, S2), strongly suggesting that delayed Erythronium shoot emergence and initiation of flower development in the presence of litter resulted from litter-induced buffering of temperature fluctuations. The implications of these phenological shifts for Erythronium were apparently negligible in our study, because delays in leaf or flower initiation were coupled with shorter maturation periods such that plants reached maturity at the same time across treatments. Because frost-damaged Erythronium shoots senesce earlier (Tessier, 2022), we speculate that delayed phenology could potentially be adaptive by reducing early-spring frost damage. Positive effects of leaf litter in our study occurred despite inducing biomass allocation shifts that could negatively affect Erythronium growth and reproduction. Similar to classic etiolation as a shade-avoidance response (Darwin and Darwin 1880), our Erythronium responded to litter cover by elongating shoots. Contrary to expectations based on other species (Eickmeier and Schussler 1993; Norby et al. 2023), the Erythronium response to shading from leaf litter did not lead to reduced petiole biomass per unit volume. Instead, our Erythronium apparently allocated biomass into petiole tissue that would otherwise have been allocated to leaf blade. Either response could be costly by reducing relative investment in photosynthetic tissue, leading to fitness reductions in the presence of litter as observed in the spring ephemeral Obolaria virginica (Wood and Bornstein 2011). Yet, despite leaf blade investment being positively associated with relative growth across treatments (Fig. 5), and despite leaf blade investment being lowest in the presence of leaf litter (Fig. 4e), Erythronium in our study grew more in the presence of leaf litter (Fig. 2b). Therefore, the negative effects of shoot elongation are more than offset by the benefits associated with leaf litter. Similar patterns occur in the spring ephemeral Anemone nemorosa, which elongates petioles under leaf litter with no net negative effects on rhizome growth (Baltzinger et al. 2012). No functional differences between E. albidum and E. americanum Many Erythronium species are locally abundant and ecologically important throughout their native ranges (Austen et al. 2018; Muller and Bormann, 1976; USDA 2014; Pupillo and Astuti, 2017), yet studies examining interspecific differences are lacking. Though our species’ ranges largely overlap in Eastern North American forests, only E. americanum extends into the extreme southeast and northeast, and E. albidum alone extends west into the prairie states (USDA 2014). We found no significant differences between these species in biomass allocation, reproduction, nor response to invaders. However, because E. albidum occurs in areas of higher soil moisture relative to E. americanum (Prince 2008), we expect that negative population-level implications due to celandine invasions, which predominate in floodplain habitats, will be greater across the range of E. albidum than E. americanum . Given their importance for ecosystem functioning and maintaining biodiversity of both plants and specialist pollinators, it is critical to understand how responses to changing conditions vary between spring ephemeral species.

Based on the results presented here, and the widespread occurrence of litter reductions and invasive celandine across our region, we would expect to see continued population declines in these Erythronium species as well as their specialist pollinators. This suggests management of celandine is warranted, contrasting with reports that celandine presence can increase seed set in nearby Claytonia virginica (Masters and Emery 2015a). We speculate that negative impacts from celandine and other invasives may be especially underestimated in spring ephemeral communities due to the combination of their short period of annual aboveground activity and tendency towards long-lived perenniality that may yield subtle changes over short timeframes. Our results also highlight the importance of litter layer preservation, for example by replacing invaders that have quickly decomposing leaves with natives that produce more recalcitrant leaf litter. Deep litter layers (~20cm) are also detrimental to celandine reproduction (Masters and Emery 2015b), which could additionally help by introducing feedbacks that reduce its competitiveness against native ephemerals. Tables and Figures Table 1. Effects of treatment, species, and initial biomass on Erythronium growth and reproduction. Bolded results are significant at P <0.05. | Relative growth (harvest at shoot maturity; n=56) | Treatment | 6.67 | <.001 | | Species | 2.54 | .117 | | | Treatment x species | 1.47 | .239 | | | Initial biomass | 12.52 | <.001 | | | Relative growth (harvest post-senescence; n=84) | Treatment | 28.45 | <.001 | | Species | 0.00 | .976 | | | Treatment x species | 0.37 | .694 | | | Initial biomass | 21.87 | <.001 | | | Offspring number (harvest at shoot maturity; n=83) | Treatment | 0.05 | .951 | | Species | 0.68 | .412 | | | Treatment x species | 0.35 | .704 | | | Initial biomass | 0.97 | .329 | | | Offspring number (harvest post-senescence; n=87) | Treatment | 1.24 | .294 | | Species | 0.03 | .861 | | | Treatment x species | 0.60 | .553 | | | Initial biomass | 0.45 | .507 | | | Total offspring weight (g) (harvest post-senescence; n=84) | Treatment | 30.62 | <.001 | | Species | 0.04 | .835 | | | Treatment x species | 0.21 | .811 | | | Initial biomass | 2.32 | .131 | Table 2. Effects of treatment, species, and initial biomass on Erythronium biomass allocation, based on harvest at shoot maturity. Bolded results are significant at P <0.05. | Petiole biomass (g) (n=66) | Treatment | 7.44 | .001 | | Species | 0.71 | .404 | | | Treatment x species | 0.01 | .990 | | | Initial biomass | 0.32 | .576 | | | Leaf blade biomass (n=68) | Treatment | 1.74 | .183 | | Species | 0.70 | .408 | | | Treatment x species | 0.09 | .915 | | | Initial biomass | 0.23 | .635 | | | Petiole biomass per volume (g/cm 3 ) (n=33) | Treatment | 3.87 | . 034 | | Species | 1.69 | .204 | | | Treatment x species | .427 | .657 | | | Initial biomass | .071 | .792 | | | Specific leaf area (g/cm 2 ) (n=62) | Treatment | 0.36 | .698 | | Species | 1.10 | .299 | | | Treatment x species | 0.76 | .475 | | | Initial biomass | 0.09 | .769 | | | Root mass ratio (n=56) | Treatment | 0.92 | .407 | | Species | 1.91 | .174 | | | Treatment x species | 0.36 | .700 | | | Initial biomass | 1.07 | .305 | | | Leaf blade investment (leaf blade biomass: total shoot biomass; n=62) | Treatment | 4.07 | .023 | | Species | 0.23 | .630 | | | Treatment x species | 0.09 | .919 | | | Initial biomass | 0.02 | .883 | Table 3. Effects of treatment, species, and initial biomass on Erythronium phenology. Bolded results are significant at P <0.05, and italicized results are marginally significant at 0.05< P <0.10. | Shoot emergence date (n=178) | Treatment | 8.67 | <.001 | | Species | 0.61 | .436 | | | Treatment x species | 0.46 | .633 | | | Initial biomass | 92.60 | <.001 | | | Leaf maturation period (n=176) | Treatment | 43.96 | <.001 | | Species | 1.86 | .175 | | | Treatment x species | 0.02 | .983 | | | Initial biomass | 45.73 | <.001 | | | Days to flower emergence (n=20) | Treatment | 3.38 | . 066 | | Species | 0.00 | .974 | | | Treatment x species | 0.934 | .418 | | | Initial biomass | 0.08 | .780 | | | Petal opening period (n=18) | Treatment | 2.93 | .096 | | Species | 0.42 | .531 | | | Initial biomass | 0.08 | .781 | Figure 1. (a) Morphology of Erythronium albidum and E. americanum . The petiole runs from the base of the leaf blade to the corm and is pigmented in the portion occurring above the soil surface. Three runners are seen growing from the corm on the right (with shoot removed). (b) The extensive root system of lesser celandine. (c) A naturally occurring E. albidum individual with petiole elongated to 10cm, presumably as a response to the surrounding celandine. Figure 2. Relative growth of Erythronium when grown under a cover of leaf litter, in bare soil, or with lesser celandine (where litter was also absent). Growth was recorded (a) at shoot maturity and (b) after senescence. Error bars represent 95% CI. The grey lines at 0 indicate final biomass equal to initial biomass. Letters represent significant pairwise differences within a given panel. Figure 3. Density plot of individual offspring biomass for Erythronium under the three planting treatments. Dashed line indicates the putative biomass threshold required for Erythronium flowering (0.196g), based on flowering incidence from corms with known biomass in the preceding growing season. Figure 4. Erythronium biomass allocation patterns when grown under the three planting treatments, including (a) petiole biomass, (b) leaf blade biomass, (c) petiole biomass per volume, (d) specific leaf area, (e) leaf blade investment and (f) root mass ratio. Error bars represent 95% CI. Letters represent significant pairwise differences within a given panel. Note that in panel (c) the pairwise difference between bare soil and celandine was marginally significant ( P= .074). Figure 5. Erythronium relative growth as a function of leaf blade investment, plotted separately for each planting treatment. Regression lines reflect parameter estimates from the best-fit model. Figure 6. Erythronium phenology variation under the three planting treatments and as a function of initial biomass, including (a) shoot emergence date and (b) leaf maturation period. Results presented in the inset figures are treatment means and 95% CI, with letters indicating significant pairwise differences between treatments.

Austen, E. J., Lin, S.-Y., & Forrest, J. R. K. (2018). On the ecological significance of pollen color: A case study in American trout lily (Erythronium americanum). Ecology, 99 (4), 926–937. https://doi.org/10.1002/ecy.2164 Ashton, I. W., Hyatt, L. A., Howe, K. M., Gurevitch, J., & Lerdau, M. T. (2005). Invasive Species Accelerate Decomposition and Litter Nitrogen Loss in a Mixed Deciduous Forest. Ecological Applications, 15 (4), 1263–1272. Arthur, M. A., Bray, S. R., Kuchle, C. R., & McEwan, R. W. (2012). The influence of the invasive shrub, Lonicera maackii, on leaf decomposition and microbial community dynamics. Plant Ecology, 213 (10), 1571–1582. Augspurger, C. K., & Buck, S. (2017). Decline in herb species diversity over two decades in a temperate deciduous forest in Illinois. The Journal of the Torrey Botanical Society, 144 (4), 392–405. Axtell, A. E., DiTommaso, A., & Post, A. R. (2010). Lesser Celandine ( Ranunculus ficaria ): A Threat to Woodland Habitats in the Northern United States and Southern Canada. Invasive Plant Science and Management, 3 (2), 190–196. https://doi.org/10.1614/IPSM-D-09-00044.1 Baltzinger, M., Archaux, F., & Dumas, Y. (2012). Tree litter and forest understorey vegetation: A conceptual framework to understand the effects of tree litter on a perennial geophyte, Anemone nemorosa. Annals of Botany, 109 (6), 1175. https://doi.org/10.1093/aob/mcs047 Banning, N. C., Gleeson, D. B., Grigg, A. H., Grant, C. D., Andersen, G. L., Brodie, E. L., & Murphy, D. V. (2011). Soil Microbial Community Successional Patterns during Forest Ecosystem Restoration. Applied and Environmental Microbiology, 77 (17), 6158–6164. https://doi.org/10.1128/AEM.00764-11 Bialic-Murphy, L., Brouwer, N. L., & Kalisz, S. (2020). Direct effects of a non-native invader erode native plant fitness in the forest understory. Journal of Ecology, 108 (1), 189–198. https://doi.org/10.1111/1365-2745.13233 Cipollini, K. A., & Schradin, K. D. (2011). Guilty in the Court of Public Opinion: Testing Presumptive Impacts and Allelopathic Potential of Ranunculus ficaria. The American Midland Naturalist, 166 (1), 63–74. https://doi.org/10.1674/0003-0031-166.1.63 Darwin, C., & Darwin, F. (1880). The Power of Movement in Plants. John Murray, London. Bates, D., Maechler, M., Bolker, B., & Walker., S. (2015). Fitting Linear Mixed-Effects Models Using lme4. Journal of Statistical Software, 67(1), 1-48. doi:10.18637/jss.v067.i01. Donath, T. W., & Eckstein, R. L. (2008). Grass and oak litter exert different effects on seedling emergence of herbaceous perennials from grasslands and woodlands. Journal of Ecology, 96 (2), 272–280. https://doi.org/10.1111/j.1365-2745.2007.01338.x Eickmeier, W. G., & Schussler, E. E. (1993). Responses of the Spring Ephemeral Claytonia virginica L. to Light and Nutrient Manipulations and Implications for the “Vernal-Dam” Hypothesis. Bulletin of the Torrey Botanical Club, 120 (2), 157–165. https://doi.org/10.2307/2996945 Fisk, M. C., Fahey, T. J., Groffman, P. M., & Bohlen, P. J. (2004). Earthworm Invasion, Fine-root Distributions, and Soil Respiration in North Temperate Forests. Ecosystems, 7 (1), 55–62. https://doi.org/10.1007/s10021-003-0130-3 Frelich, L. E., Blossey, B., Cameron, E. K., Dávalos, A., Eisenhauer, N., Fahey, T., Ferlian, O., Groffman, P. M., Larson, E., Loss, S. R., Maerz, J. C., Nuzzo, V., Yoo, K., & Reich, P. B. (2019). Side-swiped: Ecological cascades emanating from earthworm invasions. Frontiers in Ecology and the Environment, 17 (9), 502–510. https://doi.org/10.1002/fee.2099 Frey, M. N., & Schmit, J. P. (2017). Early-Season Treatment of Fig Buttercup (Ranunculus ficaria). Invasive Plant Science and Management, 10 (2), 191–200. https://doi.org/10.1017/inp.2017.13 Fox, J., and S. Weisberg. 2019. An R Companion to Applied Regression, Third ed. Thousand Oaks CA: Sage. https://socialsciences.mcmaster.ca/jfox/Books/Companion/ Habtewold, J. Z., Helgason, B. L., Yanni, S. F., Janzen, H. H., Ellert, B. H., & Gregorich, E. G. (2020). Litter composition has stronger influence on the structure of soil fungal than bacterial communities. European Journal of Soil Biology, 98, 103190. https://doi.org/10.1016/j.ejsobi.2020.103190 Jo, I., Fridley, J. D., & Frank, D. A. (2017). Invasive plants accelerate nitrogen cycling: Evidence from experimental woody monocultures. Journal of Ecology, 105 (4), 1105–1110. https://doi.org/10.1111/1365-2745.12732 Kiær, L. P., Weisbach, A. N., & Weiner, J. (2013). Root and shoot competition: A meta-analysis. Journal of Ecology, 101 (5), 1298–1312. https://doi.org/10.1111/1365-2745.12129 Lee, B. R., Yancy, A. J., & Heberling, J. M. (2024). Phenological Escape and Its Importance for Understory Plant Species in Temperate Forests. International Journal of Plant Sciences, 185 (4), 321–333. https://doi.org/10.1086/729439 Lenth, R.V. 2023. emmeans: Estimated Marginal Means, aka Least-Squares Means. R package version 1.8.5. Loydi, A., Lohse, K., Otte, A., Donath, T. W., & Eckstein, R. L. (2014). Distribution and effects of tree leaf litter on vegetation composition and biomass in a forest–grassland ecotone. Journal of Plant Ecology, 7 (3), 264–275. https://doi.org/10.1093/jpe/rtt027 Masters, J. A., & Emery, S. M. (2015a). The showy invasive plant Ranunculus ficaria facilitates pollinator activity, pollen deposition, but not always seed production for two native spring ephemeral plants. Biological Invasions, 17 (8), 2329–2337. https://doi.org/10.1007/s10530-015-0878-3 Masters, J. A., & Emery, S. M. (2015b). Leaf Litter Depth has only a Small Influence on Ranunculus ficaria (Ranunculaceae) Biomass and Reproduction. The American Midland Naturalist, 173 (1), 30–37. Masters, J. A., & Emery, S. M. (2016). Do multiple mechanisms drive the dominance of an invasive plant ( Ranunculus ficaria, Ranunculaceae) along an urban stream? The Journal of the Torrey Botanical Society, 143 (4), 359–366. https://doi.org/10.3159/TORREY-D-15-00020.1 Mattingly, K. Z., Day, C. T. C., Rauschert, E. S. J., Tayal, A., & Hovick, S. M. (2023). Genetic and morphological comparisons of lesser celandine (Ficaria verna) invasions suggest regionally widespread sexual reproduction. Biological Invasions, 25 (2), 379–397. https://doi.org/10.1007/s10530-022-02921-4 MacDougall, A. S., & Turkington, R. (2005). Are Invasive Species the Drivers or Passengers of Change in Degraded Ecosystems? Ecology, 86 (1), 42–55. Miller, K. E., & Gorchov, D. L. (2004). The invasive shrub, Lonicera maackii, reduces growth and fecundity of perennial forest herbs. Oecologia, 139 (3), 359–375. https://doi.org/10.1007/s00442-004-1518-2 Muller, R. N. (1979). Biomass Accumulation and Reproduction in Erythronium albidum. Bulletin of the Torrey Botanical Club, 106 (4), 276. https://doi.org/10.2307/2560353 Muller, R. N. (1978). The Phenology, Growth and Ecosystem Dynamics of Erythronium americanum in the Northern Hardwood Forest. Ecological Monographs, 48 (1), 1–20. https://doi.org/10.2307/2937357 Muller, R. N., & Bormann, F. H. (1976). Role of Erythronium americanum Ker. In Energy Flow and Nutrient Dynamics of a Northern Hardwood Forest Ecosystem. Science, 193 (4258), 1126–1128. https://doi.org/10.1126/science.193.4258.1126 Muller, R. N. (1975). The natural history, growth, and ecosystem relations of Erythronium americanum Ker. in the northem hardwood forest. Ph.D. thesis, Yale University. New Haven, CT. Norby, R. J., Baxter, T., Živković, T., & Weston, D. J. (2023). Shading contributes to Sphagnum decline in response to warming. Ecology and Evolution, 13 (9). https://doi.org/10.1002/ece3.10542 Palacio-Lopez, K., King, C. M., Bloomberg, J., & Hovick, S. M. (2020). Natural selection on traits and trait plasticity in Arabidopsis thaliana varies across competitive environments. Scientific Reports, 10 (1), 21632. https://doi.org/10.1038/s41598-020-77444-w Prince, Jessica. (2008). Niche distribution and partitioning of co-occuring congeners Erythronium albidum and Erythronium americanum . Governors State University College of Arts and Sciences. Graduate Project (M.A). Library Catalog. Pupillo, P., & Astuti, G. (2017). Population structure of Erythronium dens-canis L. (Liliaceae) in the northern Apennines (Italy). Italian Botanist, 4, 1–14. https://doi.org/10.3897/italianbotanist.4.12439 R Core Team (2022). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/. Reed, S. P., Bronson, D. R., Forrester, J. A., Prudent, L. M., Yang, A. M., Yantes, A. M., Reich, P. B., & Frelich, L. E. (2023). Linked disturbance in the temperate forest: Earthworms, deer, and canopy gaps. Ecology, 104 (6), e4040. https://doi.org/10.1002/ecy.4040 Rothstein, D. E. (2000). Spring ephemeral herbs and nitrogen cycling in a northern hardwood forest: An experimental test of the vernal dam hypothesis. OECOLOGIA, 124 (3), 446–453. https://doi.org/10.1007/PL00008870 Spicer, M. E., Mellor, H., & Carson, W. P. (2020). Seeing beyond the trees: A comparison of tropical and temperate plant growth forms and their vertical distribution. Ecology, 101 (4), e02974. https://doi.org/10.1002/ecy.2974 Sun, S., & Badgley, B. D. (2019). Changes in microbial functional genes within the soil metagenome during forest ecosystem restoration. Soil Biology and Biochemistry, 135, 163–172. https://doi.org/10.1016/j.soilbio.2019.05.004 Sun, S., Li, S., Avera, B. N., Strahm, B. D., & Badgley, B. D. (2017). Soil Bacterial and Fungal Communities Show Distinct Recovery Patterns during Forest Ecosystem Restoration. Applied and Environmental Microbiology, 83 (14), e00966-17. https://doi.org/10.1128/AEM.00966-17 Sydes, C., & Grime, J. P. (1981). Effects of Tree Leaf Litter on Herbaceous Vegetation in Deciduous Woodland: II. An Experimental Investigation. Journal of Ecology, 69 (1), 249–262. https://doi.org/10.2307/2259829 Tessier, J. T. (2022). Severe frost but not shade could limit the future growing season of Erythronium americanum . Botany, 100 (3), 275–282. https://doi.org/10.1139/cjb-2021-0131 USDA-NRCS-NGCE, NPDT, & ESRI [United States Department of Agriculture, Natural Resources Conservation Service, National Council for Geographic Education, National Plant Data Team, Environmental Systems Research Institute, Inc.]. 2014. The PLANTS database. https://plants.usda.gov/home. Wood, D. L., & Bornstein, A. J. (2011). Notes on the Biology of Obolaria virginica (Gentianaceae) in Southeast Missouri, and the Effects of Leaf Litter on Emergence and Flower Production. Castanea, 76 (2), 157–163. Yancy, A. J., Lee, B. R., Kuebbing, S. E., Neufeld, H. S., Spicer, M. E., & Heberling, J. M. (2024). Evaluating the definition and distribution of spring ephemeral wildflowers in eastern North America. American Journal of Botany, 111 (5), e16323. https://doi.org/10.1002/ajb2.16323 Information & Authors Information Version history Peer review timeline Published Biological Invasions Version of Record12 Sep 2025Published Copyright This work is licensed under a Non Exclusive No Reuse License.

Authors Metrics & Citations Metrics Article Usage 259views 111downloads Citations Download citation Grace Gutierrez, Stephen Hovick. Compounding negative effects of leaf litter absence and belowground competition from an invasive spring ephemeral on native spring ephemeral growth and reproduction. Authorea. 17 January 2025. DOI: https://doi.org/10.22541/au.173711012.24594636/v1 DOI: https://doi.org/10.22541/au.173711012.24594636/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu.