Darwin of course was neither immune to the fashions of society nor to the lure of exotic flowers, and he had kept and admired Orchids himself. As his health suffered he would increasingly turn to experiments with plants to satisfy his curious mind, as these could be conducted from the safety of his own home in Downe, Kent. Darwin therefore began a series of experiments to work out how Orchids are fertilised and why they have such elaborate flowers.
Darwin would cover some orchids with glass bells, others he left exposed. After flowering he would collect the seeds and try to germinate them, he found that only the orchids left exposed produced fertile seeds. He was able to pinpoint the important difference to the plants pollinia: structures, like little bags, full of pollen. These pollinia would disappear from the uncovered plants, but not the covered ones. Darwin used his network of collaborators across the world to collect facts about how these pollinia are distributed by insects. He found that many species of moths and butterflies have been found with pollinia on their proboscis, which is like a long tongue used to drink the flower’s nectar.
Further work demonstrated that each pollinia was positioned and shaped just right so that the butterflies and moths which most often visit that species pick them up. They then fly off to the next plant and deposit one set of pollinia, which fertilise the plant, and pick up another set. The flowers and pollinia of orchids were shaped by natural selection to attract and exploit insects to cross-fertilise different plants.
This work was published as On the Various Contrivances by which British and Foreign Orchids are Fertilised by Insects in 1862 and provided evidence of how natural selection can produce complex and beautiful products. It also confirmed to Darwin that plants and animals benefit by avoiding self-fertilisation; orchids went to great lengths to avoid it. Darwin thought this was important for his theory as it creates variation for natural selection to act on.
He later expanded on these ideas, again with plants, in two books The Effects of Cross and Self Fertilisation in the Vegetable Kingdom (1876) and Different Forms of Flowers on Plants of the Same Species (1877). These together provide more evidence that plants are adapted by selection to favour cross-fertilisation rather than self-fertilisation, and that the seeds of self-fertilised plants were generally less successful than those of cross-fertilised plants.