Study Species
Arion subfuscus was introduced from Europe to North America in the early 19th century (Chichester and Getz 1973). It is a polyphagous herbivore that has been described as “having the greatest impact upon natural communities” of any northeastern U.S. mollusk (Chichester and Getz 1969) and is the most abundant slug in New England (French 2012). Adults were collected from fallen logs at a forest edge in South Kingstown, RI (USA), in April-September 2022 for the seed germination assays and May-June 2023 for the seed/seedling experiment. Slugs were maintained in the lab on a diet of iceberg lettuce (Lactuca sativa var. capitata) and leaf litter in 21–24° C terrariums that were regularly misted to maintain humidity.
Brassica nigra is an herbaceous plant native to the Mediterranean (where it coevolved with A. subfuscus) and grown globally. Seeds were sourced from Outsidepride Seed Source, LLC (Independence OR USA); A. subfuscus is also found in Oregon (Burke 2013). Seeds were kept in a dark drawer, away from sunlight, prior to the experiment. B. nigra seeds came in two distinct color morphs, orange and black; we used only black morph seeds in our work to facilitate the visual assessment of germination.
Germination assays (2022)
Starting in April 2022, we conducted five replicate experiments assessing whether exposure to A. subfuscus mucus affected B. nigra germination speed (time to germination) and rate (percent of seeds that germinate). Each experiment was conducted in a series of 90 mm petri plates lined with Whatman grade 598 qualitative 90 mm white filter paper (Millipore Sigma, Burlington MA USA) and wet with 2 mL distilled water. Petri plates assigned to the mucus treatment had one pre-weighed adult slug added to each of them, while petri plates assigned to the control treatment had water only. Once the slugs were added, both slug and control plates were interspersed and held in a dark cabinet at 20–24°C.
After 24 hours, all slugs were removed and returned to their terrariums. Once the slugs were removed, a piece of Ahlstrom-Munksjö black qualitative grade 8613 90 mm filter paper (Ahlstrom-Munksjö Paper, Jönköping Barnarpsgatan Sweden) was placed in each petri plate over the white filter paper and wet with 2 mL distilled water. The black color of the filter paper facilitated visual assessment of when the white radicle emerged from each seed. Ten B. nigra seeds were then placed on top of the black filter paper in each dish. The mucus and control dishes were then interspersed and held in a dark cabinet at ambient temperature. Depending on the experiment, each petri plate was checked either twice (9 am and 9 pm; April experiment) or three times (9 am, 3 pm, and 9 pm; May, October, and December experiments) daily for seed germination. The checks continued for 5–6 days until all seeds had germinated or germination had ceased. The assay was performed a total of five times. The April experiment had 20 petri plate replicates per treatment, the May experiment had 40 replicates per treatment, the October experiment had 30 replicates per treatment, and the December experiment had 20 replicates per treatment.
Seed and seedling experiment (2023) (Supplemental Fig. 1): Prior to the start of the experiment, 130 2.5x2.5 cm styrofoam pots (Carolina Biological Supply Company, Burlington NC USA) were filled with MiracleGro moisture control potting mix (Scotts-MiracleGro Inc., Marysville OH USA). The pots were placed over a mesh screen to allow drainage and prevent pots from being exposed to water draining from other pots.
Germination assay
The 2023 germination assay portion of the seed and seedling experiment was conducted identically to the 2022 assays except as follows. Each petri plate in the mucus treatment had two adult slugs added to it, after which all petri plates were kept in the dark. There were ten control petri plates and ten mucus petri plates. After 24 hours the slugs were removed and 20 B. nigra seeds were placed in each control and treatment petri plate. Eighteen hours after the seeds were added to the petri plates, and prior to any germination, three randomly selected seeds were removed from each of the 20 petri plates for use in the competition assay (described below). Starting 24 hours after the B. nigra seeds were added and continuing every six hours afterwards, the 17 seeds remaining in each plate were checked for germination (defined as a visible radicle). The initial 24-hour sampling point was determined using data from the 2022 assays that checked B. nigra germination at 16, 18, 20, and 24 hours and found essentially no germination at the first three time points (6 of 2482 germinated seeds; 0.24% of total). Plates were checked until the first ten of seventeen seeds germinated in each petri plate (72 hours post).
Competition assay
In each of 30 styrofoam pots, a randomly selected seed from the control treatment was planted in one corner and a seed from the mucus treatment was planted in the opposite corner (a total of two seeds per pot). Each side of the pot was labelled with its treatment and originating petri plates. Pots were watered and rotated as described above.
After three weeks, the two plants in each pot were carefully removed and rinsed to remove excess soil and separate their root systems. After the plants were soil-free, they were patted dry. Total biomass was measured, and then the plant was sectioned to measure total above- and below-ground biomass.
‘Seed exposure’ and ‘seed and seedling exposure’ treatments
The first ten germinated seeds in each dish were each planted in individual styrofoam pots filled with potting mix and watered to soil saturation beforehand. Pots were placed in front of a window for sunlight, watered every twelve hours, and turned 90° each day to ensure even sunlight exposure.
One week after the start of the experiment, half of the seedlings grown from mucus-treated seeds were assigned to receive additional mucus exposure (‘seed and seedling exposure’); the other half of the seedlings were not exposed (‘seed exposure’).
To generate the additional slime cue for each ‘seed and seedling exposure’ pot, two Arion were placed into a petri plate containing 0.32 g sifted (425 µm sieve) potting mix wetted with 1.3 mL distilled water. After 24 hours, the slugs were removed, and the soil mixed to evenly incorporate the slime into the soil. A 0.3 g aliquot of mucus-containing soil was then placed on top of the existing soil in a ‘seed and seedling exposure’ pot and the pot watered immediately afterwards. This procedure was repeated for each ‘seed and seedling exposure’ pot on days 7, 9, and 12 after planting, with new petri plates and slugs used for each application.
Seedlings in the ‘seed exposure’ and 'control’ treatments did not receive additional soil during growth. To control for the effect of soil addition, we added a fourth ‘control and soil’ control treatment in which a set of control plants received 0.3 g aliquots of wetted no-mucus soil on top of the existing soil on days 7, 9, and 12. All plants were watered as above for a total of three weeks after planting.
Slug choice bioassay
Three weeks after the start of the experiment, choice bioassays were done in which Arion were allowed to choose between a cotyledon from a ‘control’ plant and one from either a A) ‘seed exposure’ plant or B) ‘seed and seedling exposure’ plant. There were 15 replicates for the ‘control’ versus ‘seed exposure’ comparison, and 17 replicates for the ‘control and soil’ versus ‘seed and seedling exposure’ comparison.
For each bioassay, a cotyledon was cut from a seedling and its fresh weight recorded; we attempted whenever possible to pair similarly sized cotyledons. A 90 mm petri plate was sprayed once with distilled water using a perfume atomizer before a control and treatment cotyledon were placed next to each other on the left and right halves of the dish. The control cotyledon was placed on the left side in odd-numbered replicates and the right side in even-numbered replicates.
One day prior to the start of the bioassays, mature A. subfuscus were removed from their terrarium and held individually in a wetted petri plate. After being starved for 24 hours, each slug was weighed and then placed in the middle of the petri plate equidistant from the two cotyledons. The petri plate was then closed, and each slug allowed to forage. After two hours, the slug was removed, and both cotyledons reweighed. Data from replicates in which the slug did not feed on either cotyledon was excluded from analysis.
Statistical Methods
Germination assays
Because the germination assays in 2022 used slightly different methods than those in 2023, we analyzed them separately (although we note that combining them does not alter our conclusions). To compare differences in the proportion of seeds in each petri plate that germinated, we used a generalized linear mixed model with a binomial response distribution. To evaluate the effect of slug mucus on the rate of germination, we used survival analysis (McNair et al. 2010). We used restricted mean survival time (Zhao et al. 2015) because is it more robust in situations where treatments have strong time-dependent treatment effects (i.e., situations that can violate the proportional hazards assumption of Cox proportional hazards regression; Zhao et al. 2015). An additional benefit of restricted mean survival time is that it produces estimates over a defined period that is relevant for ecological hypotheses (whereas Cox regression produces ratios of hazard functions that may be less intuitive). Changes in germination timing that might arise due to mucus might be expected to change the degree to which seedlings experience competition, because seeds that germinate earlier experience the benefits of being able to grow before other seeds germinate (e.g., Orrock and Christopher 2010). As such, a relevant ecological period over which to evaluate germination is the amount of time that it takes the majority of seeds to germinate (and hence the window of time when germinating early can yield benefits). We define the evaluation period as the time when > 80% of seeds had germinated; this period was 72 hours for the 2022 experiments and 48 hours for the 2023 experiment.
Competition assay
We used a general linear mixed model to examine the effect of mucus exposure on the growth of individual plants grown in competitive conditions. The dependent variable we examined was total plant biomass (the sum of above- and below-ground biomass). The model included a fixed effect of treatment (no-mucus control vs. exposure to mucus as a seed and seedling). The model included random effects to account for variation due to seeds experiencing the same conditions within a petri plate. The model also included a random effect to accommodate variation arising due to the unique pair of plants grown together in each replicate pot, reflecting the paired nature of our experimental design (i.e., two plants with different treatments were grown in the same pot).
Slug choice bioassay
We examined the effect of mucus exposure on slug herbivory by comparing the proportion of plant mass remaining following the herbivory trial using linear mixed effect models. Because of slight differences in methodology, we conducted separate analyses for plants exposed as seeds and a separate analysis for plants exposed both as seeds and as seedlings. Because our assays used a paired design that presented slugs with tissue from an untreated (control) and a mucus-treated plant, we used the treatment arena as a random effect (analogous to a paired t-test).
Statistical analyses were carried out using R 4.3.2 (R Core Team 2023), the lme4 package (Bates et al. 2015) and the survRM2 package (Uno et al. 2022).