Greenhouse experiment
Between July and November 2019, seeds were germinated in germination trays containing a 2:1 mixture of all-purpose soil and cocopeat at the greenhouse at the National Centre of Biological Sciences (Bengaluru, India). Because species fruited at different times, had recalcitrant seeds, and germination times varied among species, we planted species for germination whenever seeds became available. Seeds were watered every alternate day and monitored for germination. Once germinants lost their cotyledon leaves and developed at least four mature leaves they were transplanted into PVC pipes of diameter 15 cm and depth of 45 cm. The PVC columns were filled with a prepared soil mixture (all-purpose soil: sand: cocopeat: compost = 1:1:1:0.1) and randomly assigned to six blocks within the greenhouse. Seedlings within a block were randomly assigned to drought and control treatments. Each treatment per block had an equal number of individuals of a species, but numbers differed among species depending on seedling availability. To maintain a consistent post-germination age at which seedlings were subject to drought, species were included in the experiment as their seedlings became available. All new seedlings were allowed to acclimate in their pots for two weeks and watered every two days to avoid transplant shock. Seedlings that perished during this period were replaced and included after acclimation. In total, we had 434 seedlings in the experiment with 1-3 individuals per species per treatment per block.
After the acclimation period, we stopped watering the seedlings in the drought treatment for each block. Seedlings in the control treatment were given 75-100 ml of water twice weekly to ensure a consistent level of soil moisture. The greenhouse was maintained at approximately 65% relative humidity and 26°C temperature. Initially, we planned the drought treatment to last 16 weeks to simulate the post-monsoon seasonal drought that occurs from January through April in the Western Ghats. However, monitoring and measurements were disrupted due to the COVID-19 lockdown. For 12 of 16 species, we implemented the experiment for 16 weeks. For the remaining four species, control plants were watered by the greenhouse staff but data collection was not possible due to restricted access to the facility. We harvested these plants in the 12th week of their drought experiment when we got permission to access the facility. These 12 weeks still give us a reasonable idea of interspecific variation in response to seasonal drought conditions.
For growth, stem height, leaf length and width (for the four largest leaves) and survival was recorded weekly for the first 8 weeks and then once a month till the plants were harvested. We calculated the growth rate at the end of 12 weeks for all individuals as follows: (final height – initial height)/time in weeks. Seedling final height was strongly correlated with initial height (r = 0.7, t = 20.8,p < 0.001), hence we calculated relative growth rates, standardized by initial height as follows: (final height – initial height)/(time in weeks*initial height). Photosynthesis was measured monthly using a LiCOR 6400XT on seedlings of all species from control and drought treatments. Photosynthesis rates were measured with a 2x3 cm2 leaf chamber. The chamber conditions were maintained at PAR 1500 µmolm-2s-1, 65% humidity, 400 µmols-1 flow rate, fan speed at 10,000 RPM. Measurements were taken between 0700 - 1200 hours.
Although we had aimed to measure above and below-ground biomass allocation for all individuals in both drought and control, we lost many samples in an oven fire. Moreover, due to COVID-19 related restrictions on accessing the facility, ~20% of individuals could not be harvested on time. Hence, we were unable to assess relative differences in above- vs. below-ground allocations in relation to water availability.