One morning in 2009, I sat on a creaky bus winding its way up a mountainside in central Costa Rica, light-headed from diesel fumes as I clutched my many suitcases. They contained thousands of test tubes and sample vials, a toothbrush, a waterproof notebook and two changes of clothes. I was on my way to La Selva Biological Station, where I was to spend several months studying the wet, lowland rainforest’s response to increasingly common droughts. On either side of the narrow highway, trees bled into the mist like watercolours into paper, giving the impression of an infinite primeval forest bathed in clouds.