A recent study proposed a hypothetical “first flower”: a prediction of what the flower of the shared ancestor of all flowering plants (or angiosperms) would have looked like.
This prediction was based on evolutionary trees constructed from molecular data from 792 modern angiosperms, coupled with physical floral traits for each species, such as number and arrangement of petals and sepals (collectively known as tepals), and whether there were separate male and female, or bisexual, flowers.
The resulting “ancestor of all flowers” is a beauty: it is a radially symmetrical flower with several whorls of tepals, each containing three tepals. These surround several whorls of stamens (pollen-bearing male organs), each containing three stamens. In the centre of the bisexual flower are more than five spirally arranged separate carpels (female organs). Quite when this hypothetical flower might have bloomed is a much thornier question. The authors of the study give a (really quite big) window of 140–250m years ago, which reflects the fact that dates for the origin of the angiosperms based on genetic sequence comparisons, employing “molecular clocks”, tend to be much, much earlier than any unequivocal evidence from fossils.
The fossil record for flowering plants, which make up the vast majority of plant species today, has been a hot topic since Darwin’s day. He famously described their rapid diversification as an ‘abominable mystery’, although he was less vexed by their origins than by how such rapid rates of evolution could be reconciled with his gradualistic model of natural selection. We now have much more fossil evidence to help us reconstruct the early story of the angiosperms, but the broad pattern which Darwin recognised still holds: there are lots of angiosperms from around 130m years ago (in the Early Cretaceous) onwards, and little evidence for them much before that.
The current state of fossil evidence for the earliest angiosperms was neatly summarised in a recent review by Patrick Herendeen and colleagues. Mesofossils (specimens on a scale of millimetres, which fill the gap between larger hand specimens, and microfossils such as pollen) have been particularly informative, and the Early Cretaceous rocks of Portugal have proved fruitful. Monetianthus closely resembles modern members of the water lily family (Nymphaeaceae), while fossil berries such as Canrightiopsis are close relatives of modern species of the Chloranthaceae. Another species, Kajanthus, is a member of the modern Ranunculales order. All of these groups are basal angiosperms: the flowering plants which molecular studies show us were the earliest to branch off from the mainstream, core angiosperm group. None are in rocks older than the Barremian subdivision of the Early Cretaceous, around the 130m-year-old mark. Further afield and of more typical size for a plant fossil, Archaeanthus from Kansas, USA and dated as late Albian, about 100m years ago, is considered to be a close relative of the modern Tulip Tree, Liriodendron. This places Archaeanthus within another notable basal angiosperm group, the Magnoliales.
Fossil pollen should be our friend in the search for the early angiosperms, since it is more widely dispersed and provides a less biased record than rarely preserved hand specimens. Interestingly, triaperturate pollen associated with modern eudicot angiosperms has not been found prior to the 130m year mark. Angiosperm-like monoaperturate pollen has been found in older rocks, but may have been produced by other seed-producing plant groups, so cannot be taken alone as unequivocal evidence of the presence of flowering plants.
Herendeen and colleagues discuss, and dismiss, many putative early angiosperm species where the material is too poorly preserved to be able to say with any certainty whether it is a flowering plant or not. They also list a number of cases where they think material has been wrongly assigned to the angiosperms, and more likely belongs with another seed plant group. Their conclusions: there are no pre-Cretaceous angiosperm fossils (so far) that bear close scrutiny. They also suggest that a careful search for fossils with unequivocal angiosperm stamens and carpels, in well-preserved plant assemblages from the latest Jurassic, would be the next sensible approach. Darwin’s abominable mystery remains an open and controversial case, for now …