Environmental associations and the paleoecological significance of the genus Pyxidicula Ehrenberg,1838 and Pyxidicula muskegii sp. nov. in the Hudson and James Bay region peatlands, Canada

Soils hold the second largest pool of biomass and represent some of the highest biodiversity per unit area on Earth (Bardgett and van der Putten 2014; Nielsen et al. 2015; Bar-On et al. 2018; Geisen et al. 2019b). Biodiversity in soils is also a key factor in maintaining terrestrial ecosystems, as biodiversity loss in soils has been shown to have a negative impact on plant community diversity, carbon cycling, and nutrient dynamics (Wagg et al. 2014; de Graaff et al. 2015; Delgado-Baquerizo et al. 2020). Despite the importance of soil communities for biodiversity and ecosystem functions, relatively less is known about them compared to their aboveground counterparts, especially for microscopic organisms, where the majority of species likely remain undescribed (Nielsen et al. 2015; Geisen et al. 2019a). One such group is protists, which are the major consumers of soil microorganisms and represent the bulk of eukaryotic diversity (Adl et al. 2012; Geisen et al. 2018; Burki et al. 2021). Protists contribute to many important ecosystem processes in soil, including primary production, nutrient and carbon uptake and transfer through the food web, plant growth, and litter decomposition (Bonkowski and Clarholm 2012; Jassey et al. 2015; Geisen et al. 2016, 2021; Schmidt et al. 2016; Seppey et al. 2017; Singer et al. 2021). Their importance in ecosystem processes, along with their rapid response to environmental change and high abundance in soils, also makes them important bioindicators in both modern and paleoecological studies (e.g., Charman 2001; Payne 2013; Nielsen et al. 2015; Zhao et al. 2020; Fournier et al. 2022). Despite their importance, many species remain undescribed, including cryptic species that are similar morphologically and are only recognized through evaluating genetic or ecological differences (Kosakyan et al. 2012; Oliverio et al. 2020; De Luca et al. 2021). Therefore, the functional roles of many protists remain unknown.

One protist group that plays an important role in soil ecosystems and is an indicator of environmental change includes testate amoebae. This group of paraphyletic protists is defined by having an external shell or “test” that is formed from organic, siliceous, or carbonaceous materials (Ogden and Hedley 1980; Todorov and Bankov 2019). Their tests make them especially important for reconstructing environmental change in the past, as they are generally preserved in peat and sediment archives (Charman 2001). Within peatland ecosystems, testate amoebae have been demonstrated to have a strong response to hydrological conditions and nutrient gradients, and therefore various local to regional transfer function models have been developed to reconstruct environmental changes (e.g., Woodland et al. 1998; Booth 2008; Lamarre et al. 2013; Swindles et al. 2015; Amesbury et al. 2016, 2018; Beaulne et al. 2018; Šímová et al. 2022). However, morphological species that are rare tend to be omitted from these models and combined with undescribed species and cryptic diversity suggests that these models may be prone to losing key environmental information if species that are locally important are omitted (van Bellen et al. 2017; Marcisz et al. 2020). Therefore, continued work describing and understanding the environmental controls on different testate amoeba species is critical for improving their use within paleoenvironmental reconstructions.

Regions that have high peatland coverage are more reliant on peat core records than others to understand past environmental change. One such region is the Hudson and James Bay Lowlands in eastern Canada, which holds one of the largest continuous peatland complexes in the world and has >65% peatland coverage in approximately 400 000 km² (Tarnocai et al. 2011). Many paleoecological records from this region have relied on testate amoebae to infer hydrological changes on the landscape as a result of climatic, isostatic, and successional changes (e.g., Loisel and Garneau 2010; Bunbury et al. 2012; van Bellen et al. 2013; Holmquist et al. 2016; Ferguson and Pendea 2018; Bysouth and Finkelstein 2020; Zhang et al. 2020; Davies et al. 2021, 2023). In some paleoecological records in this region, a previously undescribed Pyxidicula Ehrenberg, 1838 species was encountered in sometimes high abundance in the paleo-archive (>50%; Zhang et al. 2020; Davies et al. 2021, 2023). These records relied on other species included in the North American transfer function of Amesbury et al. (2018) for hydrological reconstructions, as Pyxidicula were encountered in <1% of all samples in the modern dataset used to develop the model. However, their high abundance in the paleo-record suggests that it is critical to describe and characterize this new species’ environmental associations to strengthen ecohydrological reconstructions in this important peatland region.

Overall, this study had four main objectives: (1) formally describe a new species that was encountered in paleoecological records from the Hudson and James Bay region, (2) provide morphometric measurements of the new species and others in the Pyxidicula genus for comparison, (3) investigate the environmental associations and tolerances of the new species and others in the genus to demonstrate differences in their associated habitats, and (4) compare modern and paleo-assemblages to investigate the roles of taphonomy in their relative importance for reconstructing past environmental changes.

The lowland areas surrounding the Hudson and James Bays of Canada are characterized by a cool and humid snow climate (Dfc; Köppen–Geiger climate classification; Kottek et al. 2006). The study region lies within the Hudson Plains and Taiga Shield ecozones (Fig. 1; Ecological Stratification Working Group 1995), which have mean annual air temperatures of −2 and −5 °C, respectively (1961–1990; Marshall et al. 1999). Annual precipitation total is 566 mm in the Hudson Plain and 656 mm in the Taiga Shield and total snowfall is 239 and 257 cm, respectively (1961–1990; Marshall et al. 1999). Mean summer temperatures range from 12 to 16 °C (June–July–August) and mean winter temperatures range from −14 to −22 °C (November–December–January; 1981–2010; Government of Canada 2024). Mean maximum temperature is the highest in July at 23 °C for both ecozones and the lowest mean minimum temperature occurs in January and is −31 and −35 °C for the Hudson Plain and Taiga Shield, respectively (1961–1990; Marshall et al. 1999).

Specimen measurement was undertaken on subfossil specimens from peat cores collected for paleoecological analysis in Davies et al. (2021) and Davies et al. (2023) (HRST-1301 and ROFTGH27; Fig. 1). Specimens from the newly identified species (Pyxidicula muskegii sp. nov.; N = 42) and Pyxidicula cymbalum Penard, 1902 (N = 41) were measured and were previously used to support trait analysis for Davies et al. (2023). Further, P. cymbalum is the described species that has the most similar test morphology that was also encountered in the region. Peat sample preparation methods for these samples are outlined in Davies et al. (2021, 2023). Specimens were selected from multiple depths in the peat cores where they were abundant and were measured using photographs at 400× magnification with a Zeiss Axio Cam MRC camera attached to a Zeiss Axio Imager A1 compound microscope. Images were measured in the program Zeiss Zen 2.3 (Blue Edition) with the measurement tool. Repeated measurements of test diameter yielded a relative standard deviation for measurements of <1%. Morphometric measurements of each specimen included aperture diameter, test diameter, and test height (Fig. 2). Differences between the test diameter of P. muskegii and P. cymbalum were compared using a Welch’s t test in the R (R Core Team 2021).

Scanning electron microscope (SEM) images were taken to describe the surface pattern of the test of P. muskegii. Specimens were isolated, placed on a 0.8 µm Millipore^TM isopore membrane filter, mounted on a stub with carbon tape, and coated with 3 nm of platinum using a Leica SDC500 metal coater. SEM imaging was performed on a Zeiss Supra VP55 at 10 kV at the University of Toronto, Canada.

Modern samples from the western Hudson Bay Lowland (HBL) margin were collected to assess the environmental conditions that are associated with different species within the genus Pyxidicula (Fig. 1). These species included the previously undescribed species, as well as P. cymbalum, and Pyxidicula patens Claparède and Lachmann, 1858. Specimens from these modern samples were not included in the morphometric analysis as P. muskegii was generally rare (<3%) and these samples were collected post-measurement of the subfossil specimens from the peat core locations (HRST1301 and ROFTGH27; Fig. 1; Davies et al. 2021, 2023). Environmental variables included in the analysis were temperature, pH, conductivity, water table depth, percent vegetation cover, and peat properties (bulk density, C:N ratio, and water content). A total of 39 monoliths of 10 × 10 × 10 cm were taken at 15 peatland sites representing a range of microtopographic gradients and peatland types (Fig. 1). At each sample location, water pH, temperature, and conductivity were measured using an Oakton meter (Eutech Instruments) at the time of monolith sampling. Water table depth was measured after the water level equilibrated in the hole left from sampling (∼1 h). A positive water table depth indicates that the water is below the surface. An estimate of percent vegetation cover at each sampling location was performed in a 1 × 1 m plot surrounding the sampling location prior to monolith sampling. Plant species were identified to the highest taxonomic resolution possible using plant nomenclature after Flora of North America Editorial Committee (1993+). Sphagnum species within the uppermost portion of the monolith were subsequently identified in the laboratory using Bastien and Garneau (1997). Each plant species or group cover was categorized into rare (<10%; 1), occasional (10%–25%; 2), frequent (25%–50%; 3), abundant (50%–75%; 4), and dominant (>75%; 5) for analysis. For microtopography, sites were classified as a hummock (high point), lawn, or carpet. Lawns and carpets are both considered low points and were distinguished based on whether the peat was firm or loose, respectively (Rydin and Jeglum 2013).

Peat properties were assessed using a 2.5 × 2 × 5 cm subsample taken from the uppermost portion of the monolith. Samples were dried overnight (>16 h) to a constant weight at 60 °C. Dry bulk density was calculated as the dry mass of the sample divided by the wet volume. Soil water content (%) was determined as the difference between the wet and dry sample divided by the wet sample weight. Samples were ground to a fine powder using a MM200 Retch ball mill and approximately 250 mg was analyzed for C and N content using an Elementar Vario MAX Cube CN Macro Elemental Analyzer at the Ontario Forestry Research Institute, Sault Ste. Marie, Canada. Standards analyzed were within 10% of reported averages and had an analytical reproducibility of <2% and <3% for carbon and nitrogen, respectively. Replicates were performed every 10 samples (N = 4) and had relative standard deviations of <1% for carbon and <2% for nitrogen.

The testate amoeba assemblage was quantified using a 5 cm³ subsample taken from the upper 5 cm of each monolith. Samples were processed using a modified standard method from Booth et al. (2010), where samples were soaked for >16 h and then sieved to retain the 300–15 µm fraction. Samples were mounted in glycerol and counted at 400× magnification using a Zeiss Axio Imager A1 compound microscope. A total of 100 tests were counted in each sample to be representative of the major taxa, with the exception of samples that had very low concentrations (<2500 tests cm⁻³; N = 2), where 50 tests were counted to meet the minimum requirement for a representative sample (Payne and Mitchell 2009). Test concentration was calculated by adding 2.5 mL of palynospheres at the start of processing (SG06 Special Blend Lot SG607: N = 2.4 ± 0.9 × 10⁴ mL⁻¹ for the 20.3 µm spheres and N = 7.76 ± 0.46 × 10³ mL⁻¹ for the 46.3 µm spheres; Lot SG610: N = 2.37 ± 0.18 × 10⁴ mL⁻¹ for the 20.3 µm spheres and N = 1.14 ± 0.14 x 10⁴ mL⁻¹ for the 38 µm spheres). Taxonomic classification was performed to the highest resolution possible, referencing Ogden and Hedley (1980), Charman et al. (2000), Booth (2008), Mazei and Warren (2015), and Todorov and Bankov (2019). Taxonomic nomenclature generally follows Amesbury et al. (2018) with updated taxonomy based on Kosakyan et al. (2016), Gomaa et al. (2017), Duckert et al. (2018), and González-Miguéns et al. (2022a, 2022b).

Non-metric multidimensional scaling (NMDS) was performed to assess the linkages between environmental parameters, peat properties, and vegetation assemblages to testate amoeba communities using the western HBL samples (N = 39). Testate amoeba abundance data were standardized to proportions prior to analysis. The NMDS was run using the “metaMDS” function in the vegan package in R based on Bray–Curtis dissimilarity (k = 2; autotransform = TRUE; Oksanen et al. 2018). Environmental surfaces for the environmental variables, peat properties, and vegetation assemblages were fit to the NMDS post hoc and significance tested using the “envfit” function in the vegan package in R (N = 999 permutations; Oksanen et al. 2018).

The modern samples from the western HBL (N = 39) were combined with the uppermost samples from the paleoecological records of Piilo et al. (2019) and Zhang et al. (2020) from the eastern James Bay region (N = 12) to assess the water table depth and C:N ratio optimum and tolerance for each identified morphospecies (Fig. 1). Water table depth and C:N ratio were selected as they were the environmental parameters common to the two datasets that also had the strongest relationship to the NMDS (see Table S2). C:N ratio is interpreted to represent the presence of Sphagnum mosses, where a high ratio is related to high Sphagnum abundance (Davies et al. 2023). The testate amoeba assemblage and C:N ratio at sample depths 1,3, and 5 cm from each site in the Zhang et al. (2020) and Piilo et al. (2019) datasets, respectively, were averaged to be comparable to the western HBL samples. Weighted average optimum (U) and tolerance (I) values were calculated using the following equations (Birks et al. 1990):

where Y is the relative abundance of species k and X is the environmental parameter in each sample (N = 51 total). Only species that occurred in five or more samples were included in the optima and tolerance calculations (see Table S3).

To compare the abundances of Pyxidicula from modern samples to paleoecological records, core data were compiled from the Hudson and James Bay Lowlands regions where the new morphospecies and P. cymbalum had been reported. A total of 15 cores were compiled, 12 cores from Zhang et al. (2020), one from Davies et al. (2021), and two from Davies et al. (2023). Only the upper 50 cm were analyzed in this study to match the maximum depth of the records from Zhang et al. (2020) and to assess changes in relative abundance down core with no major changes in peatland type (i.e., rich fen vs. poor fen or bog).

The new species is given the name Pyxidicula muskegii sp. nov. after the word “muskeg” that originates from the Cree word “maskek” for a boreal peatland (Alberta Elders Cree Dictionary 2024; Moose Factory’s Community Language Project 2024). Additionally, the word “maskek” is associated traditionally with the Cree word for “medicine” linking the land with healing (L. Martin (personal communication, 2023)). The name was given as the type location is in a peatland on the traditional lands of the Omushkego Cree Nations. This new species has the following classification (following Adl et al. 2019 and González-Miguéns et al. 2022b):

Domain

Eukaryota

Phylum

Amoebozoa Lühe, 1913, emend. Cavalier-Smith, 1998

Class

Tubulinea Smirnov et al., 2005

Clade

Elardia Kang et al., 2017

Order

Arcellinida Kent, 1880

Suborder

Organoconcha Lahr et al., 2019

Family

Microchlamyiidae Ogden, 1985

Genus

Pyxidicula Ehrenberg, 1838

Species

Muskegii Davies and Finkelstein, 2024

This publication and nomenclature act have been registered in Zoobank (https://ww.zoobank.org; Species: urn:lsid:zoobank.org:act:718A4F1E-F2F5-413 A-8110-C0CB74B5BC3; Publication: urn:lsid:zoobank.org:pub:723F6FDE-437 A-4BD5-B299-1BE838D86078).

The test of P. muskegii is organic, colourless to brown, nearly hemispherical in shape, and thickens toward the aperture (Fig. 3 and Table 1). Mean test diameter is 64.4 ± 7.1 µm (Table 1). The test surface has an irregular areolar pattern (Fig. 3). The aperture is central, circular, and spans over 80% of the test diameter (Table 1). Some specimens have an extension of the test wall or lip surrounding the aperture (Fig. 3). This species most resembles P. cymbalumPenard, 1902. However, P. cymbalum has a significantly larger test diameter (87.7 ± 5.2 µm) and is saucer-shaped instead of nearly hemispheric (Table 2 and Fig. 4; t(74.885) = 16.979; p << 0.001).

Pyxidicula muskegii has been found in bogs and fens within the Hudson Plains and Taiga Shield ecozones of eastern Canada (Fig. 1; Zhang et al. 2020; Davies et al. 2021, 2023). In modern samples, P. muskegii has been found at a relative abundance of <3% when present (Table 3). Specimens were also only encountered in lawns (Table 3). No living specimens or cytoplasm were found; therefore, molecular analysis was not possible within the materials of this study. This species was found at higher percent abundance with depth in paleoecological records, reaching abundances of over 20% below 10 cm at multiple sites (N = 7; Fig. 5). In modern samples from this study, P. muskegii was commonly associated with Archerella flavum, Hyalosphenia papilio, Hyalosphenia elegans, and Nebela tincta, which were in over 70% of the samples that had P. muskegii present. Species that were found in over 50% of the samples containing P. muskegii include Alabasta militaris, Euglypha tuberculata, Galeripora catinus, Heleopera sphagni, Nebela collaris, Phryganella acropodia, and Physochila griseola (see also Figs. 7 and S2).

The type location for P. muskegii is an open bog site in the southern James Bay region approximately 160 km north-northeast of Hearst, Ontario, Canada (50.7656°N, 82.7794°W; 168 m asl; Davies et al. 2021). The specimen was found within Sphagnum moss and in the underlying Sphagnum-dominated peat. Peat samples containing specimens are stored at the University of Toronto, Canada. Permanent slides of holotype and paratype specimens are archived at the Canadian Museum of Nature, Ottawa, Canada (CMNI 2024-0003; CNMI 2024-0004; CNMI 2024-0005; CMNI 2024-0006; images of specimens found in Fig. S3).

Modern testate amoeba assemblages in the western HBL margin were associated with two major environmental factors: nutrient status and hydrological conditions (Fig. 6). Nutrient status was related to the peatland type, where higher pH, bulk density, conductivity, and non-Sphagnum moss abundance were found in swamp and rich fen sites versus poor fens and bogs (Fig. 6). Hydrological conditions were related to microtopographic gradients, where higher soil moisture content and sedge abundance occurred in lawn and carpet versus hummock samples (Fig. 6). Water table depth and C:N ratio had the best fit with the NMDS axes of the environmental parameters and Sphagnum moss and ericaceous shrub abundance had the best fit of the vegetation assemblages (Table S2). C:N ratio and Sphagnum moss abundance were associated with NMDS1 and water table depth and ericaceous shrub abundance were associated with NMDS2 (Fig. 6).

Pyxidicula muskegii was distinct from other identified species within the genus in terms of its environmental associations and water table depth and C:N ratio optima (Figs. 6 and 7). Although associated with similar hydrological conditions, P. muskegii was associated with samples that have higher Sphagnum moss abundance and C:N ratio (Figs. 6 and 7). It also had a wider range of tolerance for C:N ratio and the presence of Sphagnum than both P. cymbalum and P. patens (Fig. 7). The new species also had a drier water table optimum than the other recorded Pyxidicula species, but still was found at water table depths less than 10 cm (Fig. 7).

A distinct morphology from other described Pyxidicula taxa, including its hemispheric shape, irregular areolar pattern over the test surface, and average test width of 64 µm, supports that P. muskegii is a separate species (Table 4). Like all other species in the genus, P. muskegii has a round, organic test with an alveolar pattern or fine punctuations on the surface and an aperture that is over 75% of the test diameter (Grospietsch 1958; Jax 1985; Tsyganov et al. 2016). Although the most similar to P. muskegii due to its similar test surface pattern, P. cymbalum is generally larger and has a saucer instead of a hemispheric shape (Penard 1902; Figs. 3 and 4; Tables 1 and 2). Pyxidicula gibbosa Schönborn, 1966, Pyxidicula invisitata Awerinzew, 1906, and P. patens all overlap in test diameter, but all have test features that are absent in P. muskegii, including test surface undulations, a hollow and divided test border, and a distinct collar (Table 4). Other species are either smaller or larger than the range of test diameter demonstrated in this study (Pyxidicula husvikensis (Beyens and Chardez), 1997, Pyxidicula operculata (Agardh), 1827, and Pyxidicula ornata Bartoš, 1954; Table 4). Therefore P. muskegii can be consistently identified based on distinct morphological differences.

Pyxidicula muskegii occurs in drier and more oligotrophic conditions in comparison to other species in the Pyxidicula genus, including the other commonly encountered P. cymbalum in this study, which further supports its description as a separate taxon (0–27 cm and pH range 3.22–5.81; Table 3; Jax 1985; Lamentowicz et al. 2013a; Karpińska-Kołaczek et al. 2022). Their similar test surface pattern but size differences in different habitats could alternatively suggest that P. muskegii may represent intraspecific variation within P. cymbalum (Bobrov and Mazei 2004). Intraspecific variation may be especially important in a region like the HBL where there is a mosaic of peatland types within close proximity that could allow for local adaptation of a population (Riley 2011; Macumber et al. 2020). However, minimal size and occurrence overlap and test morphology variability within 20% for each species suggest that morphological differences are beyond phenotypic variation, as seen in other testate amoebae groups (Tables 1 and 2; Bobrov and Mazei 2004). Therefore, despite not encountering live specimens or cytoplasm in the samples studied, both morphological and ecological preferences of P. muskegii support its formal description.

The occurrence of P. muskegii within a wide variety of peatland types supports that the genus can be an important component of community assemblages outside of rich fen conditions where the majority of studies have reported this genus. Other Pyxidicula have also been encountered in ombrotrophic conditions, including P. ornata, further supporting that this genus can be found in a wide variety of habitats (Hájková et al. 2012). Although P. muskegii was encountered in bogs and fens within the Hudson and James Bay region, it was still restricted to relatively wet microforms, suggesting that it is still a wet indicator like other species in the genus (Table 3; Jax 1985; Lamentowicz et al. 2013a; Karpińska-Kołaczek et al. 2022). This also fits well with trait analysis where larger apertures have been associated with wetter conditions (Fournier et al. 2015; Marcisz et al. 2020). However, this species has only been described formally in one geographic location and further work is needed to understand the biogeography of this species outside of Hudson and James Bay peatlands. This also should be extended to the genus in general, since testate amoebae communities in fens remain relatively understudied (Opravilová and Hájek 2006; Lamentowicz et al. 2011, 2013a; Payne 2011; Lizoňová et al. 2019; Šímová et al. 2022; Davies et al. 2023). Furthermore, Pyxidicula are often overlooked in routine analysis due to their small, indistinct, and transparent shells, thus limiting our knowledge on their ecology (Ogden 1987; Lamentowicz et al. 2013a; Šímová et al. 2022).

Both P. muskegii and P. cymbalum increase in relative abundance downcore with limited shifts in environmental conditions in the Hudson and James Bay sites, suggesting an influence of taphonomy on the assemblages (Fig. 5). Taphonomic loss of idiosomic tests, for example, has been demonstrated to occur in both ombrotrophic and minerotrophic settings (Mitchell et al. 2008; Swindles et al. 2020; Sim et al. 2021). Despite loss of ecological information due to taphonomic processes, previous work has also demonstrated that transfer function models are fairly robust to species loss and can rely on species that remain preserved in ombrotrophic settings (Mitchell et al. 2008; Swindles et al. 2020; Sim et al. 2021). In minerotrophic settings, however, idiosomic tests can be a significant portion of the assemblage and as a result, fen reconstructions can differ significantly when these species are removed (e.g., Euglypha, Corythion, Placocista, Tracheuglypha, and Trinema; Mitchell et al. 2008; Swindles et al. 2020). Organic tests, such as Pyxidicula, tend to preserve well and can even be found in pollen slides, making them ideal candidates for characterizing environmental conditions in the past (Mitchell et al. 2008; Magnan et al. 2020). Despite this, the low abundance of Pyxidicula in modern samples has led to their exclusion from models or only a limited number of species being included, potentially further biasing paleoecological records from minerotrophic systems (Amesbury et al. 2018; Šímová et al. 2022). Therefore, it is critical to include Pyxidicula in future training datasets, and the dataset presented here provides a basis for this work.

The modern dataset in this study demonstrates that when assessed along a gradient of minerotrophic and ombrotrophic systems, nutrient status and pH are the primary control on testate amoeba assemblages in the western Hudson Bay region, suggesting that these communities could be used to reconstruct changing nutrient status (NMDS1; Fig. 6). Nutrient status as a primary control on testate amoeba assemblages has also been recorded in northern peatlands on the Alaska slope (Taylor et al. 2019) and east central Europe (Lamentowicz et al. 2013b; Šímová et al. 2022). As in these studies, the linkages between vegetation communities and nutrient status with testate amoeba communities are likely both direct and indirect, where moss types have differing environments that they provide for the testate amoeba community and are in turn controlled by the pH and redox potential of the peatland (Lizoňová et al. 2019; Šímová et al. 2022). Within the environmental variables considered in this study, C:N ratio had the second strongest relationship with NMDS axes 1 and 2, suggesting that moss type has a strong control on testate amoeba community assemblage, as seen in other studies (Lamentowicz et al. 2010; Lizoňová et al. 2019). Therefore, characterizing the vegetation and testate amoeba assemblages concurrently can help to support paleoecological interpretations, especially when taphonomic and taxonomic issues within models can limit their interpretations.

Species within the genus Pyxidicula are often not included in regional models for reconstructions of hydrology in peatlands due to their relatively low occurrence and abundance, their small and transparent tests that can be difficult to identify, and their general occurrence in rich fens that are underrepresented compared to other peatland types concerning testate amoeba (paleo)ecology in the literature (Lamentowicz et al. 2013a; Šímová et al. 2022). However, in Hudson and James Bay regions of Canada Pyxidicula can often be a significant portion of the subfossil assemblage in peatlands, including a species that has been previously undescribed (Zhang et al. 2020; Davies et al. 2021, 2023). We describe this new species, P. muskegii, and demonstrate that it has a distinct morphology compared to the species it most resembles, P. cymbalum. Pyxidicula muskegii also was found to have a higher tolerance to drier conditions and was found associated with Sphagnum moss on lawns, which is different than other Pyxidicula species found in modern samples both within the region and in other northern peatlands (Hájková et al. 2012; Lamentowicz et al. 2013a; Karpińska-Kołaczek et al. 2022; Šímová et al. 2022). The data in this study provide detailed morphometric measurements for the new species and P. cymbalum, which can be used in functional trait analysis (Davies et al. 2023) and the modern community assemblages that include Pyxidicula taxa can be used to improve paleohydrological reconstructions, especially for fens where other species are more prone to taphonomic loss (Mitchell et al. 2008).

We thank the 2018 Ontario Ministry of Natural Resources and Forestry field crew for field assistance and C. Barreto and S. Bowman for assistance with C and N analysis. We thank L. Martin, Cheechoo, and C. Lincez of the Mushkegowuk Council for supporting our research in the Hudson Bay Lowlands, sharing their knowledge of the peatlands on their traditional lands, and for contributing Cree knowledge to the naming of Pyxidicula muskegii.

Modern testate amoeba assemblages and their associated environmental, soil property, and vegetation data are found in the Supplementary Materials (cjes-2024-0039-suppb) and will be archived in the Neotoma Database (Williams et al. 2018; www.neotomadb.org). Morphometric measurements of P. cymbalum and P. muskegii are also found in the Supplementary Materials (cjes-2024-0039-suppb).

Conceptualization: MAD, JB, JWM, SAF

Data curation: MAD

Formal analysis: MAD

Funding acquisition: SAF

Investigation: MAD, JB, MJA, SRP, HZ, MG, GTS

Resources: JB, JWM

Supervision: JB, JWM, SAF

Visualization: MAD

Writing – original draft: MAD

Writing – review & editing: MAD, JB, MJA, SRP, HZ, MG, GTS, JWM, SAF

Funding for this project was provided through Natural Sciences and Engineering Research Council of Canada (NSERC) grants to SAF, a NSERC-Post Graduate Scholarship and Canadian Quaternary Association (CANQUA) Alexis Dreimanis Award to MAD, and the Government of Ontario.

Supplementary data are available with the article at doi:10.1139/cjes-2024-0039.