Focusing on 24 Asian evergreen oaks and stone oaks, we evaluate cold and drought tolerance of leaves sampled from living collections in common conditions. We pair this with natural history records to derive species’ climates of origin. We then use this dataset to test two hypotheses: i) drought and frost tolerance show greater evidence of adaptation to climate in evergreen species of Quercus than Lithocarpus; and ii) drought and frost tolerance traits show evidence of correlated evolution, indicating shared mechanisms of stress tolerance across environmental gradients.

Freezing injury of photosystem II to assess frost tolerance

We determined the level of frost injury based on at least 3 leaves per species from collections made between January 21^st and 26^th, 2023. Samples were collected mid-winter to allow for cold acclimation. Cut stems were placed in rose vials, and these were then stored overnight in the dark at room temperature. All freezing cycles were carried out within 48 hours of collection. Based on preliminary temperature-injury curves carried out in December 2022, we applied a -7°C cooling cycle for replication over the full set of 24 species, as it allowed us to detect the greatest amount of interspecific variation. Detached leaves were placed on moist filter paper on a cooling plate, in the dark. Leaves were cooled to -7°C at a rate of 15°C per hour and held at this minimum temperature for an hour, after which they were warmed back to 20°C. We measured the maximum quantum yield of photosystem II (Fv/Fm) of dark-adapted leaves using a closed FluorCam (Photon Systems Instruments, Czech Republic). Fv/Fm was recorded every 10 minutes throughout the cooling and minimum temperature portion of the cycle, and then additionally measured one, six, 12 and 24 hours after the cooling cycle finished. Fv/Fm was recorded as an image of the whole cooling plate, which contained multiple leaves. To go from pixel-level to whole leaf values of Fv/Fm, images were processed in ImageJ (Schneider et al., 2012). Individual leaves were traced, and mean Fv/Fm values were then recorded for each leaf. For each species, three leaves were exposed to the cooling cycle and an additional three leaves were exposed to a control treatment where temperature was not changed over the cooling cycle’s duration. In the control treatment, 19 of 24 species showed no change in Fv/Fm over this duration, and the remaining 5 species changed by less than 1%. 

Pressure-volume curves to assess drought tolerance

We determined the turgor loss point of at least 3 leaves per species from collections made between July 29^th and August 15^th, 2022, (n=83). Stems with recently fully expanded leaves were cut under water and placed in rose vials. Leaves were rehydrated by placing the stems in water overnight, in a dark room at ambient temperature. All pressure-volume curve measurements were then carried out within 48 hours of collection. We followed the bench drying method, using a Scholander pressure chamber (model 1505D-EXP; PMS Instrument Co., USA) and an analytical balance (AE16; Mettler-Toledo, USA), to measure full pressurevolume curves for each leaf. When a petiole was too short to fit through the gasket in the chamber, we artificially elongated the petiole, by cutting incisions parallel to the leaf’s midvein to extend the segment that fit through the gasket (Rodriguez-Dominguez et al., 2022). Leaves were scanned alongside a coin to calibrate scale, and leaf area was then measured from these using ImageJ (Schneider et al., 2012). Leaves were dried in an oven at 60°C for 48 hours, after which the leaf dry mass was measured. For each leaf, parameters were estimated using the spreadsheet provided by Sack, Pasquet-Kok and Bartlett (2010) by fitting a line through the linear section of the pressure-volume curve which included at least five observations after the turgor loss point.