CAM photosynthesis may have conferred an advantage during the Permian–Triassic mass extinction event

The end of the Palaeozoic Era approximately 252 million years ago (Ma) coincides with extensive volcanism from the Siberian Traps and was marked by global climate warming and environmental changes^1,2,3,4. This led to the Permian–Triassic mass extinction (PTME) where oceanic species extinction rates exceeded 81%, while terrestrial tetrapod genera experienced 89% losses¹. However, the nature of terrestrial vegetation response to this major environmental change is a matter of ongoing research and contrasting perspectives^5,6,7,8,9. This lack of consensus is partly due to the taphonomic influence on plant fossil preservation^9,10. Furthermore, precise dating of terrestrial sequences is difficult, making stratigraphic correlation of floras challenging; consequently, the PTME in terrestrial records is often discussed as the Permian–Triassic transition (PTT)^5,6,10. However, what is apparent is that the occurrence of a large-scale floral turnover at the PTT was followed by a distinct, low diversity and low abundance lycophyte-dominated community (Fig. 1)^5,6,11,12. Across a broad span of latitudes, from equatorial South China to high-latitude Siberia, the rise to dominance of the herbaceous lycophyte Tomiostrobus coincided with the extinction of the previously dominant Palaeozoic taxa, including Gigantopteris and Cordaites during the PTT (Fig. 1)^{5,6,12,13,14,15}.

For approximately 5 million years (Myr) after the PTME, the Earth experienced extreme warmth, with equatorial sea surface temperature over 35°C and equatorial land surface temperatures over 45°C (ref. ^3,16), linked to at least a fourfold increase in atmospheric CO₂ concentration to over 2,600 p.p.m. (refs. ^4,17,18). These conditions exceed both the photosynthesis optimal temperature threshold and pCO₂ saturation point for modern C₃ plants^19,20,21,22. The near-total dominance of herbaceous lycophytes in lowland settings in the post-extinction interval implies that they were perhaps uniquely adapted to these extreme earliest Mesozoic climates and environments^{8,11,12,23,24}. Understanding the specific traits that conferred survival advantages to these lycophytes is of critical importance to unravel the elusive killing mechanisms—and might provide insights for predicting future biosphere evolutionary trends under severe warming scenarios.

Our current understanding of these pioneer herbaceous lycophytes from the PTT is limited due to inconsistency in their taxonomy^{14,15,25,26,27,28}. These plants are structurally simple, and their stems, leaves and roots are typically indistinguishable from one another, but fortunately their sporophylls (fertile, sporangium-bearing leaves) are character rich (but also morphologically variable) thereby allowing distinct species and genera to be distinguished^13,14,15,28. But this combination of factors makes the identification of taxa from individual plant specimens problematic, leading to poorly resolved taxonomy and a limited understanding of their phylogeny. As an example, the widely used sporophyll genus Annalepis Fliche 1910 has been replaced taxonomically by Tomiostrobus Neuburg 1936 and Lepacyclotes Emmons 1856 in different studies^15,26, but whether these ‘taxa’ represent the same or multiple different taxa remains unresolved due to a lack of detailed analysis^25,28,29,30 (Supplementary Table 1). This poorly constrained taxonomy has hindered our understanding of their diversity, phylogenetic relationships, environmental importance and functionality.

Motivated by these questions, we collected data from 485 identifiable and measurable lycophyte sporophyll specimens from different regions and geological ages including living species; 285 come from late Permian to Middle Triassic strata of southwest China, and these were compared with 200 specimens recorded in the literature (Supplementary Information and Supplementary Data 1). Most of the specimens are isolated sporophylls, but for each genus there is at least one specimen representing a complete plant or a cone with sporophylls attached to the central axis (Supplementary Information). In this Article, we focus on lycophyte sporophylls because they are the most character-rich organs, show considerable phenotypic variation within and between taxa, and provide the best evidence on lycophyte diversity, phylogeny and functionality^{13,14,15,28,31}. To quantify the morphology of individual sporophylls, we scored them for 127 binary (present/absent) morphological character states (Ch-1 to Ch-127; Supplementary Information and Supplementary Data 1) in a morphometric database; these characters include the diagnostic features of each taxon and are also related to sporophyll function. By measuring multiple specimens of each ‘taxon’, we aim to characterize sporophyll heterophylly within single species and genera and plot ‘taxon’ morphospace using principal component analysis (PCA). Subsequently, the PCA data were used to identify a representative specimen for each ‘taxon’—selected solely for analytical purposes—to serve as an anchor point in the neighbourhood network analysis (NNA) for exploring morphological similarity and phylogenetic patterns. These specimens are not proposed as formal type specimens—they are solely used for computational purposes (Supplementary Figs. 2–4) to aid reproducibility. This taxonomic analysis allows us to place the lycophyte taxa from the PTT into a broader context by comparing them to their extinct and extant relatives (see Methods for detail).

To address the potential limitation of extant relatives’ traits not being inherited from their extinct ancestors, morphological/morphometric data are supplemented by carbon isotope data to infer variation in the photosynthetic pathway^24,32 of lycophytes from the PTT taxa. Carbon isotope data have been collected from individual sporophylls and the sediment surrounding them to ensure that we have sampled the fossil itself, rather than recovering a signal of dispersed carbon from the host sedimentary rock (Supplementary Fig. 6).

Then the latest version of the coupled Hadley Centre Earth System Model version 3 with a low-resolution ocean performed by the BRIDGE group (HadCM3BL) climate model is used to simulate both average and maximum daily land surface temperatures³. By integrating these climate simulations with the spatial and temporal distribution of fossil occurrences, we evaluate the physiological viability of these lycophytes under extreme greenhouse conditions.

Sporophyll morphological variability was encoded in the numerical character matrix (Supplementary Information and Supplementary Data 1) and visualized through two-dimensional PCA. Polygon areas within the PCA were used to determine the heterophylly of each taxon, resulted from the development stage or level of maturity growing position on the plant, or intraspecific phenotypic variation^13,25 (Fig. 2). Most direct size-related characters—such as Ch-28 to Ch-52 for sporophyll size (length, width, area) and Ch-92 to Ch-109 for sporangium size—contributed less than 0.1 to the top five principal components (character description is in the Supplementary Information, and the loading score is in Supplementary Table 3). This suggests that developmental or positional variation does not obscure taxonomic resolution within our character matrix. Visualization of the Phanerozoic lycopod sporophyll data from the Devonian to recent (Fig. 2b) reveals that Mesozoic taxa occupied a distinctly different morphospace to their Palaeozoic relatives (Fig. 2b)^14,33.

Within the Mesozoic lycopods, the herbaceous genera Annalepis, Tomisotrobus and Lepacyclotes were most common both spatially and temporally^{11,14,15,26,28}, especially in South China^5,6,29. The PCA analysis was initially focused on taxa from South China to avoid the possibility of convergent evolution of taxa from different regions occupying similar climate space. Following this, the analysis was conducted on the global dataset to test for a palaeo-phytogeography signal.

The PCA of lycophyte flora during the PTT to the Middle Triassic in South China (Fig. 2a) reveals important morphological overlaps and distinctions. Specifically, PCA of the Permian–Triassic transitional Annalepis share a substantial overlap in morphospace with Tomiostrobus radiatus Neuburg, 1936 (Fig. 2a), the type species of this genus from Russia. However, a distinct boundary is observed between these taxa and the Middle Triassic species of Annalepis, which includes the type species Annalepis zeilleri Fliche, 1910 and Lepacyclotes Emmons, 1856. This suggests that the herbaceous lycopods of the PTT belong to the genus Tomiostrobus (syn. Annalepis) Neuburg, 1936, while the Middle Triassic lycopods are better classified under the genus Lepacyclotes (syn. Annalepis) Emmons, 1856. This is further supported by global analyses of sporophyll clusters, as shown in Fig. 2b (ref. ^5,15,25,26).

Our PCA also reveals that Permian–Triassic transitional lycophytes from South China, within the Tomiostrobus (Annalepis) group, occupy two distinct morphospaces (Fig. 2a). Tomiostrobus brevicystis is confined to the right side (upper and lower quadrants), while the left side (upper and lower quadrants) includes Tomiostrobus zeilleri, Tomiostrobus angusta and two unidentified species (Fig. 2c). The observed overlap in PCA space suggests that T. zeilleri, T. angusta and the unidentified taxa may represent a single taxon (T. zeilleri comb. nov.), with T. brevicystis clearly distinct. Global distribution PCA of Tomiostrobus sporophylls identifies four clusters (Fig. 2d). Three clusters are situated along principal coordinate 2, with two low-latitude clusters—T. zeilleri comb. and T. brevicystis—and mid-to-high latitude clusters from Xinjiang (north-west China) and Russia. A fourth cluster, on the left, represents high-latitude taxa from Russia, Greenland and Australia. The overlapping morphospace of high-latitude taxa from the northern and southern hemispheres indicates possible convergent evolution driven by similar climatic conditions, and/or they are polar remnants of a previously widespread ancestor.

PCA of Middle Triassic Lepacyclotes from South China (Fig. 2e) reveals three distinct clusters: one combining Lepacyclotes angusta and Lepacyclotes latiloba within the morphospace of Lepacyclotes zeilleri (proposed as L. zeilleri comb.), a second cluster representing Lepacyclotes brevicystis and a third cluster for an undescribed Lepacyclotes sp. 2 with minor overlap with Lepacyclotes zeilleri comb. PCA of global Lepacyclotes occurrences (Fig. 2f) reveals six broad groupings that partially overlap one another but lack the clear separation between groups as seen in the Permian–Triassic transitional Tomiostrobus, indicating greater diversification of Tomiostrobus. The diversification of Tomiostrobus might indicate that the genus existed before the PTME when climatic conditions were more varied¹¹. Presumably they grew in isolated communities within stressed environments with poor preservation potential, for example, in and around mountain lakes³⁴. Or alternatively, it could indicate the rapid radiation of Tomiostrobus in the early stage of warming. By contrast, the similarity in morphology of later-evolved Lepacyclotes across different geological basins and latitudes might reflect the globally weakened latitudinal temperature gradients under the hothouse conditions following the PTME.

Our data reveal clear clustering of taxa in PCA space and highlight a degree of morphological variation within each taxon. To further explore the phylogenetic context for these observations, the specimen closest to the centroid of each PCA polygon (Supplementary Information) was taken as the most representative example of that taxon, and data from this individual was used to perform NNA; these taxa represent ‘voucher’ samples and are identified and illustrated in Supplementary Figs. 2–4. The NNA results are in line with data from our PCA identifying 12 distinct genera of lycopod sporophyll throughout the Phanerozoic. Within the Isoetales, Tomiostrobus, Lepacyclotes, Isoetities and Isoetes all belong to the family Isoetaceae, while Pleuromeia, Cyclostrobus and Lycostrobus belong to the family Pleuromeiaceae; the Permian–Triassic genus Tomiostrobus has the closest phylogenic similarity with recent Isoetes (Fig. 3).

The carbon isotope composition (δ¹³C) of extant and extinct plants has been successfully used to identify photosynthetic pathways and environmental stresses^24,32,35,36. However, carbon isotope fractionation within the same plant species can vary under different climatic and environmental conditions, particularly due to differences in water availability³⁷. Accordingly, we restricted our carbon-isotope analyses to latest Permian–Middle Triassic lycophytes preserved in coastal lowland deposits of South China (Supplementary Fig. 6). During the study interval, South China was in a low-latitude tropical region with limited temperature seasonality⁶, thus minimizing the influence of climatic factors such as temperature fluctuations and water-use efficiency on plant physiology and associated carbon isotope signatures. By plotting the sporophyll morphometrics (Supplementary Fig. 7) and δ¹³C values (Supplementary Fig. 8) against sedimentary facies for end-Permian to Middle Triassic South China plants, we find that neither taphonomy nor growth environment (for example, differences in salinity) is the primary control on carbon-isotope fractionation or taxonomic assignment. Rather, geologic age (reflecting background atmospheric CO₂) and genus exert stronger influences.

In modern trees, intra-organ variation in carbon isotope values—for example, from the mid-vein to the leaf margin—can reach up to 3 (ref. ³⁸). To reduce such internal variability in fossil samples, we collected material from the entire organ whenever possible. This approach is essential because Permian–Triassic plants are generally small, and even a single fossil specimen often does not yield sufficient organic material for δ¹³C analysis (see specimen pictures and scale in the Supplementary Information). Furthermore, to avoid contamination from host sediment, only the exposed surface of each fossil was sampled. For extremely small or thin-cuticle plants—such as the PTT seed fern Germaropteris and Triassic lycophytes—specimens of the same species, from the same locality and stratigraphic layer, were pooled to obtain sufficient carbon for analysis. This resulted in fewer but higher-quality data points (Fig. 4).

Similar δ¹³C values of organic matter in the matrix associated with each plant fossil in near-shore sedimentary facies from different locations suggests that the plant fossils are likely penecontemporaneous (see Supplementary Figs. 6a and 6d for details). A few literature-derived plant δ¹³C values lacking corresponding sedimentary background data were considered unreliable and therefore excluded from our main analysis, though they are listed in the Supplementary Information for reference and comparison.

The result shows the late Permian pre-extinction arborescent lycopod Lepidodendron, the conifer Anshuncladus and other plants all share a similar δ¹³C value of about −24.6‰ (Fig. 4), indicating a similar physiology in carbon isotope fractionation, likely C₃ (refs. ^39,40). After extinction, the mean δ¹³C values of the non-lycophyte flora are more negative (approximately −30.5 ± 1.0‰), tracking the isotopic shift in global atmospheric δ¹³C (CO₂)^1,2,4,16. By contrast, the mean δ¹³C composition of the Permian–Triassic transitional lycophyte Tomiostrobus flora is ~3.4 ± 0.61‰ (~1.2‰ to ~6.5‰) higher than the contemporaneous non-lycophyte flora (Fig. 4). The extreme environmental conditions after the PTME led to a 5 Myr coal gap and scarcity of Early Triassic terrestrial plant fossils⁴¹, leading to a relatively sparse dataset for this part of the study. Therefore, we were unable to conduct carbon isotope comparisons at the family level. Rather, we performed broader comparisons between lycophyte and non-lycophyte taxa, which may introduce potential broader uncertainties linked to differences in phylogeny, growth environments and post-depositional processes. The Middle Triassic Lepacyclotes–Pleuromeia lycophyte flora have median δ¹³C values 0.73 ± 0.41‰ higher than contemporaneous non-lycophytes (including Neocalamites, Voltzia, megaphyllous leaf with Spirorbis, indeterminate conifer and indeterminate seeds) (Fig. 4 and Supplementary Fig. 6). Considering the ~4-fold increase in pCO₂ and substantial temperature increase in the Early Triassic^4,16, the relative carbon isotope stability of these herbaceous lycophytes is remarkable. The fossil material selected for carbon isotope analysis was confirmed to be well-preserved cuticle, based on fluorescence microscopy observations that revealed epidermis-like cellular structures (Supplementary Fig. 34), supporting the interpretation that the δ¹³C values reflect original plant tissue rather than recalcitrant diagenetic residues.

Nevertheless, the multiple factors influencing carbon isotope fractionation—particularly the large natural variability in the isotopic composition of source materials including atmospheric CO₂, CO₂ derived from sediment organic matter decomposition and dissolved inorganic carbon in water—introduce considerable uncertainty, thereby limiting the extent to which our isotope data can be used to calculate crassulacean acid metabolism (CAM) productivity directly. To place our morphological and isotopic results to a broader context, we used the Earth system model HadCM3BL to simulate palaeo-climate conditions—particularly changes in land surface temperature—across the PTT³. By coupling these simulations with the known fossil occurrences of Triassic lycophytes, we aim to more broadly evaluate whether extreme thermal conditions could have necessitated the use of CAM photosynthesis for survival.

The Earth system model HadCM3BL is capable of simulating robust climate conditions for the Permian–Triassic interval, consistent with multiple climatic and environmental proxy records³. Using this model, we generated maps of both average and absolute maximum daily land surface temperatures for three key intervals: the end-Permian Changhsingian (pre-PTME; Fig. 5g,h), the PTT (syn-PTME; Figs. 5d,e) and the Early Triassic Induan (post-PTME; Fig. 5a,b), under reconstructed atmospheric CO₂ concentrations, sea surface temperature proxies and climatic facies and mineralogical data (see detailed explanation in ‘HadCM3BL climate simulation’ in Methods).

By overlaying palaeogeographically corrected macrofossil and microfossil records of Triassic lycophytes onto these palaeogeographic maps (Fig. 5c,f,i), we determined the modelled average and maximum land surface temperatures at each fossil locality. Fossil evidence shows that lycophytes were most widespread during the PTME, with many occurrences located between 45° N and 80° S where average maximum daily land surface temperatures exceeded 40 °C (Fig. 5).

Extant C₃ plants have an optimal growing temperature of 10–35 °C and are unable to survive at higher temperatures due to physiological constraints such as water limitation, Rubisco enzyme deactivation and elevated photorespiration^19,22,42,43. By contrast, these Triassic lycophytes were able to persist in regions such as South China, North China, Xinjiang, Europe, Australia, India and Argentina, where the modelled average maximum daily temperature exceeded 40 °C and the absolute maximum daily temperatures ranged from 45 °C to 65 °C (Fig. 5). One potential photosynthetic pathway that could accommodate such high daily temperatures is C₄ given that plants using this pathway are known for their drought and heat tolerance^19,44. C₄ plants, however, are restricted to the angiosperm clade, with the earliest records dating to the Oligocene, and did not exist during the Permian–Triassic⁴⁴. Alternatively, CAM photosynthesis has been previously hypothesized in deeper time^{11,32,36,45,46,47}. CAM plants—dominant in recent hot, semi-arid to arid regions worldwide, including deserts—can persist under conditions with surface temperatures approaching 70 °C (refs. ^21,22,48). The survival of Triassic lycophytes under comparable extreme heat is therefore more consistent with the possibility of an alternative photosynthetic pathway, such as CAM, rather than solely the C₃ type.

The PCA and NNA results indicate a close morphological linkage and thus a close phylogenic relationship between the Permian–Triassic transitional lycophytes and recent Isoetes. This is especially clear when comparisons are made to Tomiostrobus, a stratigraphically confined Permian–Triassic transitional species (Figs. 2 and 3). The morphological similarity between the two taxa is driven by a number of key structures shared between these temporally disparate sporophylls, including a herbaceous growth form (Ch-3), sporophylls arranged in compact clusters along the cone axis (Ch-11), a long and slender leaf apex (Ch-58, Ch-59), a wide side angle (Ch-73), a hastate (spearhead-shaped) leaf base that indicates a tight attachment to the central axis (Ch-80), the presence of a prominent longitudinal vein in the leaf apex (Ch-82), a clavate (club-shaped) sporangium (Ch-88) and relatively small sporangium size (Ch-99) (see detailed character explanation in the Supplementary Information with annotation figures and the biplot in Supplementary Information)^14,15. In extant Isoetes, these features are associated with increased buoyancy facilitating sporophyll transportation and dispersal through water^14,29. It is likely that Tomiostrobus had the same traits as Isoetes which permitted its widespread spatial distribution along continental margins (Fig. 5c,f)^{14,15,28,29,30}.

The close phylogenetic relationship between extant Isoetes and the Permian–Triassic transitional lycophyte flora allows us to hypothesize about the factors that favoured their proliferation during the PTT. Extant Isoetes are mostly semi-aquatic to aquatic and are renowned for ecophysiological flexibility regarding their photosynthetic pathway (facultative CAM) and their capacity to absorb CO₂ from sediments and the water through their roots (passive diffusion)^46,49,50,51. The pre-extinction arborescent lycophytes, such as Lepidodendron, similar to extant Isoetes of the same class, had abundant aerenchyma tissues inside their trunk which in extant Isoetes enables the transportation and storage of CO₂ as malic acid for allowing CAM photosynthesis, indicating the potential of CAM within the class Lycopiosida²³. CAM has also been inferred in the Late Triassic Mesenteriophyllum, a Pleuromiacea from polar regions that lacks stomata and thus has been assumed to have relied on CO₂ absorbed through its roots and CAM photosynthesis¹¹. Together, these morphological, physiological and ecological parallels indicate strong evolutionary conservatism within the lycophyte clade³³, supporting the hypothesis that CAM capability could have persisted over geological timescales⁴⁶.

In extant facultative CAM species, such as Isoetes, the proportion of photosynthate derived from the CAM pathway increases with stress, whereas under low-stress conditions they function primarily as C₃ plants^36,46. During CAM, stomata remain closed during the hot period of the day to reduce water loss during respiration and photosynthesis⁴⁸. At night, when temperatures drop, they open their stomata to absorb CO₂, storing it as malic acid in vacuoles, which is used for photosynthesis during the day^49,50,51,52. This adaptation helps these facultative CAM plants survive in hot, arid conditions and reduces photorespiration by concentrating CO₂ (refs. ^{21,22,45,48,53}).

Theoretically, switching between photosynthetic pathways impacts the carbon isotopic signature of plant tissues, with facultative plants from more equable environments having a typical C₃ isotopic signature and stressed plants having a less negative (more enriched) δ¹³C value due to rising CAM contribution⁴⁵. However, Isoetes absorbs a portion of its CO₂ from sediment-derived sources with typically more negative δ¹³C values, which can offset the expected positive shift in Isoetes δ¹³C due to CAM^34,45. As a result, Isoetes tends to show δ¹³C values comparable to those of C₃ plants^24,53,54. For example, the extant aquatic species Isoetes howellii, found in standing lakes, has a δ¹³C_org of ~29‰ (±0.9‰)⁵³. Conversely, Isoetes from more water-stressed environments such as the seasonally drought-tolerant Isoetes (Stylites) andicola has a δ¹³C_org value of −22.5‰ (ref. ⁵⁵), potentially reflecting a higher proportion of photosynthesis via stress-induced CAM, although the nature of this stress fractionation response is yet to be fully characterized^{24,34,45,46,53,55}.

Atmospheric CO₂ concentration potentially increased from fourfold to sixfold during the PTT⁴, accompanied by a notable negative C isotope excursion. In South China, there is a ~6.5‰ negative shift in the bulk organic δ¹³C values¹⁶. Similar trends are observed globally, including a general 4‰ to 8‰ negative shift in total organic carbon δ¹³C values in terrestrial sediments and plant tissues (down to −32‰), and a ~3.5‰ decrease in marine carbonate δ¹³C values (to −1‰)^1,16,18. During this transition, the lycophyte δ¹³C_org values from this study (lycophyte δ¹³C_org −27.2 ± 1.2‰) are notably less negative than those of non-lycophyte vegetation (non-lycophyte δ¹³C_org −30.5 ± 1.0‰) and are closer to the δ¹³C_org values of associated sediments (−27.6 ± 1.3‰) (Supplementary Fig. 6). The pronounced negative shift in δ¹³C_org values of non-lycophyte plants and associated sediments is consistent with previous records, reflecting a major disturbance in the global carbon cycle. The consistent ~1‰ difference between each Triassic lycophyte specimen (black dots, Supplementary Fig. 6) and the surrounding matrix (red dots, Supplementary Fig. 6) confirms that these values represent primary plant material rather than diagenetic alteration (Supplementary Fig. 6).

Extant Isoetes, using the facultative CAM pathway, partially use sediment-derived CO₂, which is typically ¹³C-depleted compared to the air, as a substrate for carbon assimilation^34,45,53. The more negative δ¹³C of sediment CO₂ offsets the ¹³C-enrichment associated with the CAM pathway, resulting in a δ¹³C composition of Isoetes that can overlap with those of C₃ plants relying on atmospheric CO₂ for carbon assimilation^34,45,53. If the Triassic lycophytes used purely the C₃ photosynthetic pathway and assimilated only sediment-derived CO₂, then their δ¹³C values would be more negative than that of the sediments. Conversely, if they used CAM photosynthesis with exclusively atmospheric CO₂, their δ¹³C values would be expected to exceed those of both contemporaneous non-lycophytes (including the Permian–Triassic transitional Germaropteris leaf, Middle Triassic Neocalamites, Voltzia, megaphyllous leaf with Spirorbis, indeterminate conifer and indeterminate seeds) and the surrounding sediments. Therefore, the observed δ¹³C_org values of the Triassic lycophytes—relatively enriched compared to non-lycophytes, but similar to those of associated sediments—suggests a distinct carbon isotope fractionation pattern associated with CAM photosynthesis involving partial uptake of sediment CO₂ or a higher proportion of C₃ relative to CAM photosynthesis (see Supplementary Fig. 9 for detailed analysis).

Although certain identification of present-day CAM photosynthesis in plants is linked to nighttime malic acid accumulation, this cannot currently be tested for in fossil plants. Our carbon isotope data, however, when combined with our phylogenetic analysis and climate modelling, is most parsimoniously interpreted as evidence of Permian–Triassic transitional lycophytes using CAM as an adaptive mechanism to cope with harsh earliest Triassic climate^1,3,4,16,56. The Permian–Triassic transitional herbaceous lycophytes that dominated coastal habitats have elevated δ¹³C compositions relative to contemporaneous non-lycophyte plants (Fig. 4). The difference in carbon isotope compositions between the contemporaneous floras (lycophyte compared to non-lycophyte) is at its greatest in the Tomiostrobus Permian–Triassic transitional flora and declines through the earliest Triassic (Fig. 4). We suggest this isotopic shift records a gradual transition to a less stressful climate³ and a reduction in the utilization of CAM by plants which can operate both C₃ and CAM photosynthesis facultatively. However, the scarcity and poor preservation of plants through this time interval results in a very limited fossil record, so this assertion cannot be fully tested at present.

Extant Isoetes provide insights into how post-PTME lycophytes such as Tomiostrobus may have thrived. Some species (for example, Isoetes piedmontana) switch between C₃ and CAM photosynthesis depending on seasonal stress intensity and retreat to a corm under extreme drought or heat exceeding their highest tolerance^34,50,51. Others (for example, Isoetes sinensis) use antioxidant enzyme systems to withstand desiccation and heavy-metal stress^{51,52,57,58,59,60,61}. Some Triassic lycophytes, such as Tomiostrobus from South China, are interpreted as inhabiting paralic settings, akin to modern tidal-shore relatives (for example, Isoetes riparia)⁶², where emergent and submerged forms would have benefited from the thermal buffering of water. Together, these traits—including CAM flexibility¹¹, dormancy⁵¹, aquatic habits^34,49,57,62 and antioxidant defenses^{52,57,58,59,60,61}—likely contributed to the resilience of Triassic lycophytes and highlight continuity with the survival strategies of modern Isoetes^{5,8,11,12,14,15,30}.

The PTT was highly anomalous: established, geographically widespread, diverse lowland arboreal forest ecosystems^5,6,25 were rapidly replaced by low-diversity, herbaceous, lycophyte-dominated communities across the transition^{5,6,8,11,12,30}. This switch marks a change in plant body size and a reduced biomass^8,49,63. Furthermore, our phylogenetic and isotopic analyses suggest that the PTT lycophytes were able to use the facultative CAM photosynthetic pathway, and HadCM3BL climate model simulations suggest that these lycophytes managed to survive in an area with surface temperature higher than the highest tolerance of extant C₃ plants. A terrestrial lowland biosphere dominated by CAM plants is greatly different from one dominated by C₃ photosynthesis. As an example, while CAM plants have a higher CO₂ fixation efficiency, the storage of CO₂ as acids results in their relatively lower carboxylation efficiency which feeds through to lower productivity and less growth^49,50,53,64. Even though increasing CO₂ after the PTME may have helped carbon assimilation efficiency of CAM plants⁶³, the overall productivity of these herbaceous lycopods, resembling present-day CAM plants under chamber CO₂ experiments^53,63, would have been much lower than the pre-extinction ever-wet arborescent forests of the late Permian^45,49,53,65.

Consequently, the dominance of Triassic dwarf lycophytes capable of flexibly operating CAM photosynthesis would have reduced terrestrial organic carbon burial via photosynthesis and bio-weathering⁸, as well as lowered nutrient fluxes to the ocean^66,67—a feedback that would have amplified the post-PTME warming trend⁶⁸. However, plant macrofossils alone provide only a partial view of vegetation composition across environments^5,8,10. To capture this more realistically, vegetation models need to incorporate the CAM functional type, at least from the PTME onward, to better simulate terrestrial biomes and productivity. Such improvements are critical for robust carbon-isotope mass-balance modelling and for evaluating the broader environmental consequences within an Earth system framework.

At the same time, the persistence of CAM lycophytes can be viewed as a critical survival strategy under the extreme precipitation variability, prolonged droughts and warmth characteristic of the ‘mega-El Niño’ world of the Early Triassic³. This resilience ensured that some lowland terrestrial vegetation cover was maintained, which may have prevented an even more profound collapse of terrestrial ecosystems and a shift to extreme greenhouse conditions well beyond the ~5 Myr recovery interval^{8,21,22,48,68}.

Lycopods sporophyll character identification and measurement

The characters used to differentiate lycopods include root structure, overall plant morphology, cone (strobilus) structure, spore type and number, sporophyll characteristics and sporangium features, incorporating both terminological organ descriptions and topological measurements³¹. Our study encompasses 127 characters for Isoetales and Lepidodendrales lycopods, with a primary focus on reproductive organs—particularly sporophylls and sporangia—which are more commonly preserved in the fossil record. Although spores are widely used in lycophyte taxonomy, most are found as dispersed specimens rather than in situ, making it difficult to confidently associate them with specific plant macrofossils and resulting in substantial missing data. As the morphology of sporophylls and sporangia is already sufficient to distinguish among taxa in our dataset, we do not emphasize spore data in depth in this study with only simple classification. Future research integrating spore ultrastructures can further refine lycophyte phylogenetic relationships.

Key distinguishing characters include overall plant growth habit (Ch-3), sporophyll phylotaxy (Ch-11), presence or absence of isophylly/heterophylly (Ch-17), apex shape and presence (Ch-58, Ch-59), base shape related to sporophyll attachment (Ch-80), sporangium shape (Ch-88, Ch-89) and sporangium surface ornamentation or structure (Ch-110, Ch-111) (see the loading value of each character in Supplementary Fig. 5). Detailed explanations and figure annotations for these characters are provided in the Supplementary Information. Each specimen of every taxon is coded in a character matrix (Supplementary Data 1), with images and sketches of lycopod sporophyll fossils available in the Supplementary Information.

When selecting characters to distinguish between species, having more characters does not always improve the outcome. Speciation is influenced by isolation and adaptation to different environments. Each species comprises individuals that have evolved under similar environmental conditions, leading to the development of new morphological characters derived from ancestral traits. Therefore, selecting morphological characters with functional role is crucial, especially for studies related to plant physiology. Including too many non-functional characters can dilute the results and reduce their reliability. Characters inherited from common ancestors should be excluded when performing clustering within the same family or order. In addition, random characters lacking functional roles—potentially arising from genetic mutations or preservation biases rather than natural selection—should also be excluded.

In animal phylogenetics, characters are categorized and weighted based on their functional roles⁶⁹. Similarly, in this study, we have reviewed and discussed the potential functions of the characters used to inform subsequent phylogenetic and ecological analyses. Many characters, such as sporophyll shape and sporangium position, are related to water transport capabilities, while sporophyll base shape affects the attachment and transport of sporophylls on the central axis^14,29. Detailed functional inferences for most characters are provided in the discussion section of the Supplementary Information. However, some characters in our matrix lack clearly defined functions, a challenge exacerbated by the limited availability of close extant relatives and the recent extinction of many genera^70,71,72. Given the existing gap between plant morphology and function, each character in our matrix is considered equally important^71,72.

PCA

Two-dimensional PCA was conducted on the presence/absence of data for lycopod characters, using Euclidean distances in PAST (v4.02)^10,71. The method effectively reveals both gradual and distinct variations in sporophyll morphology. Gradual variations are considered within-species diversity, while distinct variations are interpreted as representing different species or subspecies. To capture as much morphological variation (heterophylly) as possible within each taxon, all available plant fossil samples were included in the PCA. In cases where fossils were incomplete but identifiable, missing portions were inferred by comparison with better-preserved specimens of the same species. Fossils that were poorly preserved with unpredictable missing parts or lacking critical information were excluded from the analysis. Consequently, some lycopods of interest may be absent from the dataset. Researchers are encouraged to follow the protocols outlined in the Supplementary Information for incorporating their own fossil collections to enhance the database.

We used original taxonomic names rather than combined or revised names to avoid conflating data and introducing potential biases. For instance, ref. ²⁵ proposed synonymizing dispersed sporophylls previously classified as four species by ref. ³⁰ into a single species, Tomiostrobus sinensis, which was excluded from our analyses.

In the PCA, each character represents an independent dimension, with data point locations determined by their Euclidean distances across these dimensions. Taxa are grouped based on all data points corresponding to a specific species or genus. For visualization, the high-dimensional taxon volumes were projected into a two-dimensional space that captures the maximum amount of character information. The summary scores for each principal component (PC), representing the percentage of variance explained, are listed in Supplementary Table 2. The top three principal components are used for generating the two-dimensional morphospace plots, with the highest score PC1–PC2 shown in Fig. 2 and additional PC1–PC3 plots in Supplementary Fig. 1. Note that only a subset of character information is included in the PCA analysis.

In the PCA plots, polygons of different colours represent clusters of sporophyll characters corresponding to individual species groups. The area of these polygons reflects the range of morphological variation within each taxon, with larger areas indicating greater variation⁷¹. Proximity between polygons suggests potential close relationships that warrant further NNA. Overlapping morphospaces are interpreted as potentially representing subspecies. In the PCA analyses shown in Fig. 2, certain characters present or absent in all selected taxa were excluded to prevent data dilution (highlighted as red in Supplementary Data 1). All the fossils are preserved in Room 014B, Main Building, China University of Geosciences (Wuhan).

NNA of cladistic matrix

Neighbourhood network (NNA) is a clustering method that incorporates all characters and is widely used in phylogenetic analysis. It is particularly useful for phylogenetically unsorted taxa, such as most plant fossils, where homoplastic (incompatible) signals can overshadow phylogenetic signals, potentially leading to incorrect tree inferences⁷². Unlike dichotomous tree methods, neighbourhood network effectively handles non-tree and incompatible signals by representing them as a network, thus providing a more accurate depiction of ancestral–descendant relationships⁷².

For phylogenetic NNA analyses, we selected one ‘best-preserved’ specimen per taxon to represent the taxon. However, given the heterophylly within taxa as illustrated by the polygon areas in our PCA results, defining the best-preserved fossil can be ambiguous. To minimize subjective bias and mitigate the influence of incompatible data, we selected only one specimen per species that was closest to the centroid of the polygons in the PCA results, reflecting the morphological variation of sporophylls within each lycopod taxon. All taxa and character matrix data were stored in Mesquite (v3.70) and uploaded into PAUP (v4) for distance matrix calculations. The resulting distance matrix was then used to generate the neighbourhood network in SplitTree (v4.18.3). For detailed procedural instructions, refer to ref. ⁷².

The distance between each tip in the NeighborNet represents the morphological distance between samples, with a 0.1 scale bar indicating the distance in pixels.

We compared the results of the NNA with the PCA results to ensure consistency in phylogenetic information. Both results indicate 12 independent genera of lycophyte sporophyll: Palaeozoic Cyclostigma, Achlamydocarpon, Lepidostrobophyllum, Mazocarpon, Lepidostrobus and Moscovstrobus, and Mesozoic to recent Lycostrobus, Cyclostrobus, Pleuromeia (Pleuromeialean, Lycomeia), Isoetes (Isoetites), Tomiostrobus and Lepacyclotes (Fig. 3). There are clear transitional taxa between each genus in the families Isoetaceae and Pleuromeiaceae. For example, Tomiostrobus (Skilliostrobus) australis (number 310) occurs between Tomiostrobus and Isoetites, while the Lepacyclotes found in North China during the Middle Triassic have the highest similarity with the Permian–Triassic transitional Tomiostrobus angusta. The latter is included within Tomiostrobus zeilleri, and Pleuromeia shaolinii is associated with Pleuromeia and Cyclostrobus (Fig. 3). Based on this comparison, we are able to revise the taxonomy of Triassic Isoetales lycopod sporophylls to robustly distinguish genera, species and subspecies based on our presence–absence data and our morphometric analysis (Supplementary Table 2). Our result suggests there are 26 species including 44 subspecies on a global scale, rather than the 73 species suggested by the existing taxonomy (Supplementary Table 4). Our dataset contains recent Isoetes species and comparable fossil lycopod species, providing a window that links fossil plants to their living descendants. This allows for an exploration of the linkage between morphology, genetics and phylogeny. In Supplementary Table 4, the red and bold taxa are distinct extant species with species designation via either morphological and/or genetic information; thus, these occurrences represent valid taxa and should not be synonymized with taxa in the same branch of the NNA tree. Comparisons between our revised taxonomic groupings based solely on morphology and the current genetically based phylogenetic species lists of Isoetes and Isoetites suggests that our dataset and data processing methods (PCA and NNA) might have artificially reduced the diversity. This is due to factors including (1) morphological characters from other parts of the plant aside from sporophylls distinguishing living species, (2) loss of morphological information during fossilization and (3) the increasing primacy of genomic information in systematics of living species. For example, in our morphological analysis the extant species Isoetes cangae and Isoetes serracarajensis resolve as a single species, whereas molecular analysis identifies them as distinct species⁷³. Overall, these results suggest that our morphology-based phylogeny is, predictably, of lower resolution than a genetic-based taxonomic system, especially in closely related species with similar sporophyll organization. However, this integration of extant and extinct plants into a single phylogenetic framework allows us to pose new questions about the ecophysiology of these extant floras.

Carbon isotopes

Carbon isotope ratios reflect the balance of physiological processes in plants, such as photosynthesis, respiration and transpiration over the lifetime of that tissue. These processes are influenced by atmospheric CO₂ pressure, temperature, and local environmental factors such as water availability and salinity. To accurately separate physiological differences in palaeo-plants from environmental influences, sedimentary facies analysis is crucial before selecting plant samples for carbon isotope testing.

Over the past decade, we have conducted sedimentary surveys in South China with the assistance of numerous collaborators. We collected plant fossils from various sedimentary facies and reconstructed plant habitats based on fossil preservation conditions and sedimentary facies⁵. Our study covers floras collected from terrestrial, paralic and deep-sea facies⁵. To assess the impact of atmospheric CO₂ pressure, we sampled plant fossils with carbon films or cuticles from the Late Permian to the Middle Triassic, alongside proxy-based atmospheric CO₂ content reconstructions for each substage. The age, facies and palaeoenvironments of each flora are detailed in ref. ⁵. The specific parts of the plant fossils that were sampled are detailed in the Supplementary Information.

To ensure that the carbon isotope samples are derived from plant fossils, we analysed both the organic matter from the plant fossils and the surrounding rock matrix. An ~1‰ difference in δ¹³C_org between the plant fossils and the surrounding matrix confirms the reliability of the samples⁵⁴. Matrix samples were cleaned with compressed air to prevent cross-contamination and are documented in the Supplementary Information. Only identifiable plant fossils were sampled. Samples were extracted using an alloy scalpel, with a minimum of 20 mg per plant body fossil and 5 g for surrounding rock. To avoid contamination by surrounding matrix to the plant fossil samples, we systematically scratched as thin layers as possible. To get enough sampling amount for small plants, for example, the Triassic lycopods, some samples of the same species and specimen were gathered as one sample, resulting in fewer samples but higher accuracy of each datapoint. Considering each part of the plants may bear slightly different carbon isotope value, all parts of each plant fossil were sampled, including leaves (vegetative/sporophylls), branches, seeds, petioles and veins.

To eliminate the influence of inorganic carbon on the carbon isotope signal, all samples for organic carbon isotope testing—including plant carbon such as cuticles and surrounding rock—were treated with 15% HCl acid then repeatedly rinsed with deionized water before drying at 45 °C and subsequent crushing. The description of each sample and the carbon isotope data are presented in the Supplementary Information. The prepared samples were analysed for organic carbon isotope ratios using a Mat253 Plus (Thermo Fisher, MAT 253 Plus Isotope Ratio Mass Spectrometer) and a Delta V advantage (Thermo Fisher) at the State Key Laboratory of Biogeology and Environmental Geology, China University of Geosciences (Wuhan), and an EA-IRMS system (Elemental Analyzer–Isotope Ratio Mass Spectrometry) at the Stable Isotope Facility, Department of Plant Sciences, University of California, Davis. For the Mat253 Plus, calibration was based on GBW (Guobiao Wuzhi, Chinese National Standard Reference Materials) standards (GBW04407, −22.43; GBW04408, −36.91‰) with ACET (acetanilide) (−26.33‰) as the internal standard. For the Delta V Advantage, reference materials included USGS40 (−26.39‰), USGS24 (−16.05‰), and IVA33802174 Urea (−37.32‰). For the EA-IRMS system, multiple laboratory reference materials were used for scale normalization and quality control, including caffeine (δ¹³C −34.90 ± 0.09‰; δ¹⁵N −2.74 ± 0.10‰), glutamic acid (δ¹³C −10.98 ± 0.10‰; δ¹⁵N −8.54 ± 0.08‰), glutathione (δ¹³C −18.27 ± 0.07‰; δ¹⁵N −5.00 ± 0.04‰), scallop (δ¹³C −16.74 ± 0.10‰; δ¹⁵N 9.37 ± 0.06‰) and nylon powder (δ¹³C −24.90 ± 0.05‰; δ¹⁵N −1.12 ± 0.16‰), among others. Analytical precision was better than ±0.1‰ (1σ) for standards and typically within ±0.2‰ for samples, with a maximum uncertainty of ±0.5‰ in cases of low signal intensity or abnormal matrices (for example, high halogen or sulfur contents). Replicate analyses of samples yielded reproducibility better than ±0.2‰ (Supplementary Table 5). All remaining samples and plant fossils are stored in Room 014B, Main Building, China University of Geosciences (Wuhan) and University of Leeds.

HadCM3BL model simulations

HadCM3BL is an Earth system model that incorporates atmosphere, ocean, land and biosphere, developed by the UK Metoffice and University of Bristol³. Specifically, we use HadCM3LB-M2.1aD with a grid resolution of 3.75° × 2.5° in longitude × latitude in both the atmosphere (19 vertical levels) and ocean (20 vertical levels), using the Arakawa B-grid scheme. The model uses a dynamic vegetation scheme, which is crucial for such studies: the Top-Down Representation of Interactive Foliage and Flora Including Dynamics with the MOESE 2.1 land surface scheme. Desert soil albedo is interactively updated on the basis of the soil carbon content, where low soil carbon concentrations result in a modified soil albedo of 0.32 (average modern-day Saharan albedo).

Typically, the ozone distribution is prescribed as a static latitude–pressure–time distribution in many climate models. However, as the climate warms, the tropopause rises, meaning that stratospheric ozone penetrates into the troposphere, which is unphysical if a pre-industrial tropopause height is prescribed for warm time periods. Instead, the ozone distribution is prescribed using a dynamic approach in which ozone is dynamically coupled to the model tropopause height with constant values for the troposphere (0.02 p.p.m.), tropopause (0.2 p.p.m.) and stratosphere (5.5 p.p.m.). This change makes a negligible difference to the global mean surface temperature but does have a small impact on the stratospheric temperature and winds.

A range of boundary conditions are required to configure the model for Permo-Triassic conditions. The Getech Plc. palaeogeography (land–sea distribution, bathymetry, topography) is used as well as time-specific atmospheric pCO₂ (detailed below) and solar luminosity. Each simulation was fully equilibrated in both the atmosphere and deep ocean following a three-stage spin-up protocol so that each simulation is fully equilibrated: (1) the globally and volume-integrated annual mean ocean temperature trend is less than 1 °C per 1,000 years, (2) trends in surface air temperature are less than 0.3 °C per 1,000 years, and (3) net energy balance at the top of the atmosphere, averaged over a 100 year period at the end of the simulation, is less than 0.25 W m⁻². These simulations have generally been run for over 10,000 model years to ensure complete Earth system equilibrium. Climate means were then produced from the last 100 years of the simulation.

Using systematic proxy data, including sea surface temperature, atmospheric CO₂ and sedimentary observations such as climatically sensitive minerals/facies, HadCM3BL successfully established robust simulations across the PTME interval that shows a mega-El Niño and stronger temperature fluctuations both on land and in the ocean due to the collapse of meridional overturning circulation and a contracted Hadley cell³.

In this work, we ran the end-Permian Changhsingian, PTT and Early Triassic Induan scenarios using HadCM3BL with atmospheric CO₂ concentrations of 412 p.p.m., 2,568 p.p.m. and 4,000 p.p.m., respectively, derived from boundary values reconstructed using plant stomatal, palaeosol and plant carbon isotope fractionation proxies^4,17,18. All simulations can be found on the Bristol BRIDGE website (https://www.bristol.ac.uk/geography/research/bridge/). After stabilization of the atmosphere–ocean–vegetation coupling, we outputted the average and absolute maximum daily land surface temperature.

The global average maximum daily land surface temperature is the average of each day’s highest temperature over a year, describing the overall thermal intensity experienced by the land surface⁷⁴. This metric is essential in capturing the cumulative effect of heat extremes, which are critical for assessing the habitability of terrestrial environments, especially for vegetation. Unlike mean annual temperature, this index reflects both the frequency and intensity of high-temperature events, providing insights into seasonal thermal stress and potential physiological thresholds for plant survival and function⁷⁴.

The global absolute maximum daily land surface temperature, by contrast, captures the single highest temperature recorded in each grid cell of each scenario. This metric reflects the most extreme thermal event experienced at each location, providing crucial information on the upper thermal limits of the environment. It is particularly valuable for evaluating the survivability of organisms under short-term extreme heat stress, which may exceed physiological thresholds even if average conditions are tolerable. This parameter helps identify thermal hotspots and assess the risks of episodic temperature extremes that can drive ecological collapse or restrict species distributions.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.