Previous research (e.g., [26, 29, 38] and the present study support the idea that soil protist diversity increases at warm (and humid) sites. However, the novelty of our study is that we also provide evidence supporting the idea that this temperature-diversity relationship (or species-energy effect) might have an evolutionary origin rooted in the conservatism of ancestral regimes rather than on present-day thermal conditions. Here, we explored the phylogenetic signal for this thermal niche conservatism among distantly related soil protists by also assessing their elevational patterns in range-size distributions and their within-community phylogenetic structure over an elevational gradient.
In our study, average annual temperature was identified as the main driver of the decline in soil protist diversity with increasing elevation. This result was consistent with a species-energy effect, a climatically based mechanism proposed to explain the occurrence of diversity patterns in nature . The species-energy effect often proposes temperature as a proxy of ambient energy and as the main driver of biodiversity patterns . That prediction is true in places where temperature is at or below the tolerance range for life , such as on the way to mountaintops . In mountain slopes, temperature decreases ~0.6 °C for each 100-m increase in elevation , a trend that also applies to the elevational gradient studied herein. Since our lowest and warmest sampling site exhibits an average annual temperature of 10 °C , a temperature value that is already below the thermal optimum for most protists [30, 31, 40], it is obvious that environmental conditions, including temperature, become harsher for soil protists towards the top of the elevational gradient investigated despite the increase in precipitation along the gradient.
The abovementioned species-energy effect (as measured by the average annual temperature) does not prevent site-specific (local) abiotic factors to shape soil protist diversity. Indeed, soil protist diversity has been shown to vary in relation to nutrient availability, pH and conductivity, among others local abiotic factors . However, these factors seem to be more relevant at fine spatial scales, or microbial cultures [41, 42]. By contrast, climatic conditions seem to explain better the diversity patterns at broad-spatial scales, which are relevant for soil protist biogeography and macroecology [11, 17, 26, 29]. In fact, local abiotic factors often lose power to predict diversity patterns at increasing spatial scales . This is mainly because the factors influencing diversity gradients are scale dependent, and thus, explanations vary with the spatial scope of the analysis (extent) and with sample resolution (grain) . Also, climatic conditions ultimately regulate local abiotic conditions, explaining why climate often arises as the main driver of biogeographical and macroecological patterns of eukaryotic organisms , including soil protists [17, 29]. Nevertheless, some studies also highlight the joint role of local abiotic factors and climate in shaping elevational gradients in soil protist diversity (e.g., [44, 45]. However, as these studies did not control for confounding effects of local abiotic conditions, the relative contribution of local and climatic drivers in shaping soil protist diversity could not be determined.
In this study, we did control for confounding effects of local abiotic conditions since we always sampled along an elevational gradient of beech-dominated forests. Beech and other dominant trees homogenize soil properties and change their physicochemical characteristics (pH, conductivity, organic matter, etc.) [46, 47]. This contributes to reducing the effect of site-specific abiotic factors in driving soil protist diversity along elevational gradients [38, 48]. This could explain why local abiotic factors had little, or no predictive power as compared to temperature (species-energy effect) to explain protist diversity patterns.
Water, either alone  or in conjunction with temperature , is another driver of soil protist diversity across broad spatial scales. However, the role of water as a predictor of biodiversity varies spatially, becoming more important in sites where it is scarce . In fact, water does not predict soil protist diversity in wet environments . We did not record any relationship between humidity and soil protist diversity. So, as we expected, water was not a limiting factor at the study site. Indeed, precipitation (another proxy of water availability) increases along the elevation gradient investigated , while our results show that protist diversity tends to decrease. So, while temperature is really putting the fitness of protists to the test, water availability seems to be high enough (at least on average despite the regular occurrence of drought periods in summer) to not limit their physiological functions in the study site.
The assessment of the elevational patterns in soil protist range-size distributions suggested that the species-energy effect observed might be ultimately driven by a thermal niche conservatism. Indeed, all soil protist taxa investigated exhibited a Rapoport effect depicted by a progressive increase in their distribution ranges from lower and warmer sites to higher and colder sites. The occurrence of a Rapoport effect has been related to the existence of thermal evolutionary constraints that prevent multicellular organisms [10, 51] and soil protists  from adapting to and eventually colonizing areas with severe, often exceedingly cold (or hot) climates. So, the observation of a Rapoport effect suggests that, in general, soil protists lack ecophysiological traits to overcome the low temperatures that predominate at higher elevational sites over the year. Indeed, the low diversity recorded at higher and colder sites, coupled with the occurrence of a Rapoport effect supports the idea that few soil protist taxa can physiologically overcome the lower temperatures of higher sites. Probably at higher and colder elevations, habitat filtering processes select comparatively few eurythermal protists, which, in turn, represent cases of recent adaptation to exceedingly cold temperatures.
In addition to assessing the elevational patterns in soil protist range-sizes distributions, we also investigated their within-community phylogenetic structure to track the existence of a thermal niche conservatism. The soil protist group that combines all the taxa investigated exhibited both phylogenetically clustered and phylogenetically over-dispersed communities over the elevational gradient. This suggests that, overall, the soil protist group is represented both by taxa with broad and narrow thermal tolerances. The analyses performed on individual taxonomic groups, confirmed the above-mentioned outcome. Bacillariophyta, the only phototrophic taxon included in our study, exhibited phylogenetic overdispersion among co-occurring representatives and, therefore, communities structured by biotic processes [13, 37]. This outcome suggests that Bacillariophyta have broader thermal tolerances than the heterotrophic soil protist taxa investigated. Indeed, phototrophic protists exhibit higher reproductive rates than heterotrophic protists at cold temperatures [30, 31, 40]. Given that the temperature range recorded at our study sites seems not to play an important role on the structuring of Bacillariophyta communities, we hypothesize that habitat filtering (as measured by the average annual temperature) possibly plays a pivotal role in the assembly of Bacillariophyta communities at higher and colder elevations than those surveyed at our study site (and thus not in beech forests). Probably, at our study site, competition for sunlight is the most important biotic process in the assembly of their local communities, since spotlights are scarce and unevenly distributed in forests. Future research performed over broader elevational ranges are needed to test our hypothesis. By contrast, Cercomonadida, Ciliophora, Euglyphida and Kinetoplastida (all of them heterotrophic taxa), exhibited phylogenetic clustering among their co-occurring representatives. This result confirmed that these taxa exhibit local communities structured by habitat filtering over the elevational gradient investigated [13, 37]. Therefore, heterotrophic soil protists exhibit a narrower thermal tolerance than Bacillariophyta to the monotonic decrease in temperature observed in the study site.