[MUSIC] Hello. My name is Gitte Petersen and I'm a botanist at the Natural History Museum of Denmark. I am standing in the botanical garden in the month of May and we can see we are surrounded by flowering plants in their full bloom. The diversity of plants in the garden is obviously far greater than the the diversity in most natural habitats. But still most terrestrial ecosystems on Earth are like this garden, dominated by flowering plants. But 150 million years ago, the world looked different. There were plenty of plants, but no flowers. So the natural question to ask is, how did flowering plants evolve? What did they evolve from? Or phrased in a different way, what are their closest relatives? And when did they evolve? And how did the first flowering plants look? These are some of the questions we will try to answer. But first of all, we need to define what we actually mean by flowering plants. Basically, we must define what a flower is. You can see a perfect or complete flower in this diagram. From outside, the flower is composed of sepals, which are usually green and of petals, which are usually colored. Then come the male reproductive organs, the stamens, with their anthers including the pollen grains enclosed in pollen sacks. In the center of the flower, you see the female part, the ovary composed of carpels enclosing the seeds. Let me show you the same thing in a real flower over here. This is a paeony with large flowers ideal for demonstration. If we look at this flower, we can see from below the green sepals and then followed by the bright red petals. Inside the flower there are lots of stamens. And in the center of the flower, in this one there are two carpels. If I pick a fruit that is a bit more ripe, like this one. I can open the carpel and inside we can see a lot of small pale seeds. Some of these structures are unique to flowers and flowering plants. New inventions, so to speak. The sepals and the petals are new, and so are the stamens and the carpels. None of these structures are found anywhere else in the plant kingdom. But other parts are not new. The seeds and the pollen themselves are not exclusive to flowering plants. We find both seeds and pollen in all seed plants. So the presence of seeds and pollen tells us that the flowering plants must have originated within the seed plants. So let's go and take a look at some of them. Apart from the flowering plants, there are four groups of extant seed plants. We can see the largest group, the conifers, all here in the background and another group, a very small group, are the gnetophytes, represented by this, Gnetum, where we can see a lot of naked seeds here in this, structure. Then, there is a group composed of just a single species, the ginkgo. I just have this branch here. The final group of seed plants are the Cycads. We will go and see them in a little while. Each of these groups of seed plants has it's own special morphology of the reproductive organs. And if we knew how the seed plants were related, it would help us to understand how the flower had evolved. But unfortunately, we don't know that. Even though many attempts to resolve the phylogeny of the seed plants have been made, the results are very different. So even after 30 years of molecular phylogenetics, we still have no clear answer. It's also important to keep in mind that several groups of seed plants are now extinct. So for all those, we only have the morphological evidence that must be interpreted from more of less well-preserved fossils. So largely due to this uncertainty about the relationships of the seed plants and the precise origin of the flowering plants, there are two competing hypotheses about how the flower was developed. The two theories are the euanthial theory and the pseudanthial theory. According to both, the petals and the sepals of modern flowers are developed from leaves. The main difference between the theories is that the euanthial theory explains the flower as developed by compression of one unbranched terminal shoot, whereas the pseudanthial theory explains it as compression of a branched shoot. So according to the euanthial theory, stamens are leaf structures developed from special leaves, which are called microsporophylls, which have pollen sacks along their margin. In contrast, the pseudanthial theory explains the stamens as stem structures with terminal pollen sacks. When it comes to the carpels, both theories explain them as leaves, but according to the euanthial theory, the carpels are seed-baring leaves, the macrosporophylls, which are fused along their margins and thereby enclosing the seeds. The pseudanthial theory says that the carpels are leaves originally having subtended branches with terminally placed seeds. Even though it's still not clear which theory is correct, most evidence point towards the euanthial theory, which explained all the parts of the flower as leaf structures. But in any case, the ovary and eventually the ripe fruit is the leaf structure enclosing the seeds. And this explains the scientific name for the flowering plants. The angiosperms, which translates into something like plants with covered seeds. To demonstrate the possibility of developing a closed ovary from a leaf with marginal seeds, let's go and take a look at a Cycad. This is a species of Cycas. As you see, it's a small tree with a rosette of leaves at the top, and in the middle it has some brown special seed-bearing leaves, the macrosporophylls. I have cut one of them off, and in the margin you can see the large naked seeds. Only the large green ones are developing. The rest has aborted maybe because pollination doesn't work well here inside the greenhouse. The leaf is very tough, but imagine that I could fold it along the middle and if the two margins fused, then we would end up with an ovary composed of a single carpel, with all the seeds placed in one side. And this is actually exactly as we already saw it in the paeony. Each carpel is closed individually and all the seeds sit in the same side of the closed carpel. But when I open it it becomes clear that the seeds are actually sitting along both margins of the carpel. So now we have developed the flower, but when did this development take place? We have two sources of evidence. The fossil record and dating based on molecular phylogenetics. The oldest fossil flowers so far discovered are some 125 million years old. But the oldest remains are just pollen grains, which can be dated back to some 136 million years to the Early Cretaceous. Molecular dating tends to suggest that the flowering plants are older and that may be true because the fossil record can only provide a minimum age. But molecular dating is also full of pitfalls, and even some recent estimates varies from an age around 140 million years, which fits the fossil record nicely, to some 200 million years. The fossil record is not only a source for dating the flowering plant, but it also tells us about how they looked. But the fossil record of early flowers is scarce. And for the first 10 million years or more, we only know how the pollen looked. There's also a limit to how many details we can get from fossils. And the structures that we see are often open to interpretation. Just take this very famous fossil of a plant named Archaefructus sinensis, discovered in China, and described in 2002. Despite that it's an extremely well-preserved complete plant, it is still debated whether it's a plant with an inflorescence composed of individual female and male flowers, or whether the plant has an elongated flower with stamens in the lower part, and carpels in the upper part. Luckily, we have another source of evidence about the morphology of early flowers, that is, modern flowers. Just as we can use the phylogeny for dating, we can use it for tracing particular morphological traits back in time. Just imagine that this simple phylogenetic tree shows a relationship of seven species with either red or yellow flowers. If two sister species both have the same color we must assume that their ancestor had this color too. And so we can trace colors back in time, pinpoint the origin of the change in colors, and in this case determine that the ancestral flower was red. However, tracing ancestral traits is not always that easy. Just imagine the flower colors of the seven species had this distribution on the phylogeny. Now we can tell whether the ancestor was red or yellow. So using this approach how much can we learn about the early flowering plants from tracing morphological traits on a phylogeny? Well, first of all, we obviously need to have the phylogenetic tree and despite that also within the flowering plants, there are some uncertaincy, we do have a fairly good idea about the relationships of the early branches in the phylogeny. The sister group to all of the other flowering plants is a single species, Amborella trichopoda. A small shrub or tree found on the island of New Caledonia. It has small unisexual flowers that is, they are either male or female. In the photo you see the male flowers. The next branch in the tree is the Nymphaeales, the waterlily order. A group composed entirely of water plants, many with much larger and more showy flowers, which are all bisexual. In the photo of the waterlily flower, you can see both numerous stamens and the top of the ovary in the center. The next branch in the tree is the Austrobaileyales, a small group entirely composed of woody plants. Most well known is the star anise, as you can see here, which produces the fruits and seeds that we use as a spice. But these flowers from the Kadsura and Schisandra also belong to the group. You can see that the flowers are unisexual. Both of these are female flowers with many small carpels. The next branch in the tree is a group composed of the Magnoliids and the Chloranthales. The Chloranthales is a small group of shrubs or herbs with rather small, inconspicuous flowers, which may be bisexual or unisexual. The Magnoliids are a much larger group and more diverse, the trees, shrubs, and herbs. Some have larger showy flowers, like these magnolias. Inside the flower, you can see the short red stamens and the green carpels. So the flower is bisexual. But other Magnoliids look completely different. This is a Drymus with slightly smaller flowers, and others have even smaller and inconspicuous flowers like this Pepperomia. This is the whole inflorescence and the individual flowers are barely visible, but whether they are large or small, all Magnoliid flowers are bisexual. For now we will need the rest of the phylogeny, and I will try to sum up what the combined evidence from fossils and from phylogenetic character tracing can tell us about the first flowering plants. Most likely they were either herbaceous or small shrubs. Flowers were likely small, that is less than a couple of centimeters, perhaps like the Kadsura. Then there was no differentiation yet between petals and sepals. That is, all the flower leaves looked the same as you can see it in the flowers here. All the flower leaves are more-or-less petal-like. The flowers were also regular, or had radial symmetry. That is, any two halves of the flower are mirror images, as in the flowers here. Another ancestral trait shared by all the flowers is the position of the ovary above the petals. As you will see later, many flowers have changed that position so the petals and sepals will be on top of the ovary instead. The flowers that we see here having the ancestral trait are all hypogynous. The ovary or the carpels themselves also have an ancestral trait, the stigmatic tissue, where the pollen will be attached, is located just on top of each of the carpels. You can see it in the magnolia as a pale tip of the carpels. Later you will see how other flowers develop a long style to raise the stigma above the ovary. Finally, the stamens of early flowers were slightly laminar. That is, they were somewhat broad and leaf-like. And in the picture you can see it in the Amborella. So these things we are all pretty certain about. Other things we still don't know. We don't know whether the first flowers were unisexual like this or if they were bisexual. We don't either know how many floral leaves they had. How many stamens, or how many carpels they had. And we don't know what kind of fruit they had. Was it a fleshy berry or a dry capsule? Hopefully, new evidence will answer these questions to give us a complete picture about the very first flowering plants. [MUSIC]
