The Origins of Flowers

good evening ladies and gentlemen can i welcome you to the royal society it's my great pleasure this evening to introduce our distinguished speaker professor sir peter crane who's going to be talking about the origins of flowers before i give a very brief introduction to peter can i just remind people that this talk is going to be webcast live and not only must we have our mobile phones off but we must have that we must have them completely off not just on silent otherwise it does terrible things to the web broadcast so there are three brief things i want to tell you about peter before he starts his talk first peter is one of our most distinguished botanical scientists darwin famously said and you'll hear more about this from peter that the origin of flowers is an abominable mystery and although that mystery isn't completely solved we are a long way down compared with darwin's day and this is through a combination of molecular studies morphological studies and paleo paleo i can't even say it fossil studies and peter has been one of the leaders especially in the fossils and the morphological side he's a member of the fellow the royal society foreign member of the national academy and many other academies around the world second thing i'd like to tell you about peter is that he is a hugely distinguished and excellent science leader peter first led a major organization in the states the famous field museum in chicago and from there he moved to kew gardens where he was director for seven years stepping down two years ago and queue has been enormously fortunate to have a series direct directors ian france then peter and steve hopper now who really make you one a world-class biodiversity institut institution peter returned to chicago as i said two years ago but has recently announced that he's moving again and is going to go to yale to become dean of the school of forestry and environmental studies the last thing i'd like to tell you about peter is a more personal thing in that peter is a handyman to have in the jungle i was hugely fortunate to spend a week with peter in the forest of malaysia and borneo about two years ago in fact looking at the royal society program out there and peter as well as being a fabulous scientist of the wonderful naturalist i i learned a huge amount especially about some of the less showy flowers and the ferns and the mosses and things and peter is also immensely resourceful we'd plan to go to this wonderful lost world platter the mario basin and it looked like we weren't going to be able to get there because meetings kept multiplying and things and we were about to abandon it when sir peter crane the director cue managed to rustle up a helicopter and so we rushed out of the meeting peeling off suits and getting to our field gear and got out there and also managing to avoid a 2000 feet climb through hot sticky weather so i've always thought peter that if you get bored of science that there's a career ahead of you it's i'm a botanical celebrity get me into there so without more do can i ask you to come and give the talk on the origins of flowers charles thanks very much indeed it's a very great pleasure to be here and um uh this evening i want to talk about um the origins of flowers but i want to preface my remarks first with some acknowledgements and my two long-term collaborators uh elsa marie fries and kai pederson are here in the audience and much of the work that you'll see this evening uh was work that we've done together over the last 25 or 30 years even so they deserve a huge amount of the credit for what progress we've made in some of the areas that i'll talk about this question of uh really where flowers come from uh in an evolutionary sense is obviously an important and fascinating one and it's tied up with the origin of the group that bears these wonderful flowers this is the american lotus photographed on the mississippi last summer but it was a this was a topic that uh darwin was very interested in and as uh charles has already mentioned this famous uh quote the rapid development as far as we can judge of all the higher plants within recent geological times is an abominable mystery it's a mystery to me how this quote has really sort of taken off uh in the literature because it was just a quote in a letter written to one of my predecessors at q uh joseph hooker but it was i think really popularized by the paleobotanist steward who with darwin's son published several of his letters and he drew attention to this quote very early on and i think it's worth considering what this quote means because it's been used in many ways by many different people for many different aspects of this general problem and i think the essential elements of this is is rapid development higher plants by which he means flowering plants within recent geological times these i think are the two key things so he's he's worried about the the speed and he's worried about the fact that they show up um rather late and it's also worth of course uh mentioning darwin's relationship uh with hooker and indeed his general interest in plants several of the books he wrote dealt with specific botanical topics and at least three of them dealt with uh particular issues related to the evolution of flowers and the relationship between darwin and hooker was a very close one uh uh there's a note in the queue archives uh that came to hooker from one of darwin's children after his death that says you were his closest friend so hooker and darwin were really very close and they shared an interest in botanical topics this is a nice north american prairie and the scene that i show you is really just to just to remind you that the world's vegetation is dominated by flowering plants and if you look at the species numbers and i should say there's only one number on here that we know for sure and that's that ginkgo only has one species but the point is really to simply to show you that flowering plants with three 350 400 000 living species far and away dominate the diversity of plants on land right now so they show up late in the fossil record all of these groups have a fossil history that goes back 200 million years or more whereas angiosperms have a fossil record that goes back about 100 million years and we'll look at that in more detail in a moment so that's the the recent origin piece of the darwin uh abominable mystery and the other piece is that um where this great transition occurs which is roughly at about 100 million years you move from fossil floras which are dominated by ferns and cycads and conifers and other seed plants into floras which are dominated by angiosperms and you can see this in the field if you're in late cretaceous deposits it's all angiosperms and if you're in early cretaceous deposits there's not an angious burn to be seen so darwin was was interested in the rapidity of this radiation and his interest was peaked even more by the fact that the fossils that were being described in the latter part of the 19th century from places like greenland and north america and southern europe were to the paleobotanist of the day very similar to living flowering plants so it seemed that you went from nothing to a kind of fully formed flora very very uh quickly and this was this was clearly at the heart of darwin's worries so what i'm going to do today is really deal with these uh five questions and uh uh perhaps the most interesting one of these is number five and that's the one that you'll hear least about i'll spend my time uh on the other uh four so let's first consider this question of uh the abominable mystery and the extent to which it's solved or not the abominable mystery i would argue at least in the sense that darwin was using that term in terms of the recency uh and in terms of the rapidity is to a large extent uh solved and that's because this uh apparently sudden change during the mid cretaceous looked at now with better stratigraphy looked at now with better techniques appears much more gradual and that work uh was really um brought to uh a kind of set of conclusions in the mid 1970s in two books one by norman hughes paleobiology of angiosperm origins and a volume edited by charlie beck the origin and early evolution of angiosperms they both came out in 1976. and i just show you one diagram from the doyle and hickey paper and this is across that early late cretaceous boundary and what it really shows is that down here in the aptian in the early cretaceous we have a relatively sparse uh diversity of angiosperm leaves a few angiosperm pollen grains but by the time we move up a couple of tens of millions of years we've got a much greater variety of leaves and a much greater variety of pollen grains and interestingly one of the places where you see this gradation the best at least based on the pollen grains not based on really anything uh else is in a variety of floras from uh southern and southern england and this diagram shows the phases record recognized by norman hughes through this interval in his book that came out a little later in the 1990s and then the incoming of different kinds of angiosperm pollen grains through this interval they come in a nice sequential appearance so when you look in more detail they don't come in quite as suddenly as darwin expected they come in more gradually and i think that dilutes part of the abominable mystery what's interesting though is that these very earliest angiosperm pollen grains are all based on a very similar plan and here are two pollen grains viewed with a scanning electron microscope of that age and they each have a single germinal aperture we call these things monocopa pollen grains a little bit later in the fossil record come other grains with three apertures and the symmetry of pollen grains which is quite different from those early forms and we now know based on the wonderful advances that have been made in understanding the phylogenetic relationships among this great group of flowering plants much of it done by mark chase who's here in the audience and his colleagues we now know that the groups of angiosperms fit together like this and those pollen grains with those three apertures are characteristic of this group are the eu dicots the eu dicots account for around 75 percent of living angiosperm species and pollen grains of that kind come in relatively late compared to pollen grains of the other kind so what we have then is exactly what we would have expected in the fossil record from the systematics there's complete concordance of these two lines of evidence with respect to this one particular feature of the pollen grains which is a very important one so you dicots with those pollen grains with the three apertures about three quarters of living angiosperm species and i won't dwell on how they get those three apertures but i'll simply go back to norman hughes's diagram these are different sections uh in some in the north of england most of them in the south of england his phase is coming through here and while we have lots of uh mono-aperturate pollen grains angiosperm pollen grains at this level we first get these triaterate pollen grains characteristic of eu dicots one grain in these samples that he looked at one grain in these samples and in other parts of the world those pollen grains appeared about that same level this is one of the pollen grains that norman hughes figured from his book and these are very important because they give us a one of our most robust tie points for the origin of a major clade of flowering plants so just to sort of roughly summarize what we see through the early cretaceous by the time we get down to the jurassic cretaceous boundary at about 144 million years we have no evidence of angiosperms at about 135 million years we see the first incoming of these monocoque pollen grains with a particular kind of pollen wall diagnostic of flowering plants and then a little later we see the incoming of this triaterate pollen of presumed eu dicots so to some extent part of that abominable mystery the part that deals with the rapidity is solved we can see a stepwise incoming of flowering plants into the fossil record and the order of appearance corresponds well to the order of appearance that would be predicted from our studies of living plants alone let's move on to the second issue of of what is a flower and i'll start with a wonderful illustration from the time of darwin this is victoria amazonica this is fitch strawing for joseph hooker's father actually when this plant flowered in the uk for the first time in cultivation it's a water lily a giant water lily we have the carpools with the ovules inside that develop into the seeds uh we have the stamens in this tyres in this flower around the edge of a cup and then we have a variety of petal-like perianth members around the outside and this is the common pattern that we see in the waterlily family but we also see great variation in this uh waterlily family and the reason that i'm focusing on this family is that it's one of the very earliest lineages to branch from the main line of flowering plant evolution so looking at what that lineage looks like today is a very important framework for interpreting some of the early cretaceous fossils so this is another genus and this has much smaller more simple flowers and indeed this one is wind pollinated rather than insect pollinated and then just recently incredibly we've had the discovery of this remarkable plant much of the original work done by paula radol from queue who's here in the audience the discovery of this thing which for a long time we thought might be related to grasses we now find that it's related uh to this water lily uh family and over the last couple of days we've had a a wonderful symposium here on floral biology and floral evolution and one of the things that paula pointed out in her paper is how difficult it is actually to define the flower in these complex reproductive units here so it's not always straightforward uh to define a flower and and um we see this in in other groups that i'll allude to a little later this is the molecular phylogenetics um showing where this curious plant tritheria fits in it fits in as the sister group to all others in the waterlily clade and the only branch that appears below it down here at the base of the angiosperms is a thing called amborella a single species that lives in new caledonia more diagnostic or more straightforward than the flower itself are the components that go to build up uh an angiosperm flower and there i really want to highlight these these three uh components because these are the things that we should be trying to explain the origin of uh in the fossil record if we're interested in the origin of flowers and the origin of flowering plants the pollen producing organs of flowering plants the stamens are very characteristic structures i think for a long time we overlooked just how distinctive they were but they're quite unlike the pollen producing structures of any other seed plant that we know they have two pairs of pollen sacs and then the feature that we've always focused on is the origin of the carpal itself this is a structure that encloses the ovule and ultimately matures into the fruit and then there's the ovule itself because this ovule has two surrounding coverings two integuments so as we'll see in a moment the female parts of a flower a typical angiosperm uh flower is like a sort of matrushka doll with one thing inside the other we've got um the megasporangium uh at the center surrounded by a new cellus surrounded by one layer surrounded by another layer both integuments and then surrounded by uh the carpal so if we go to another very simple flower at the base of flowering plants you can see this uh curious angiosperm stamen you're only seeing one side of it here but this is one of the thicky in which there are two uh pollen sacs there would be two more on the other on the other side here there are no uh petal-like period parts just a little bract here's the carpel and inside would be the ovule this beautiful picture is by peter andres who's in the audience here it's a genus called sarcandra in the chloranthy so just to remind you again a cross section through that stamen then would look something like this with these two pairs of pollen sacks and the ovule inside the carpal would have these two layers around the uh around the new what we call the new sellers so what were the earliest flowers like well this is a paleo botanical question and here i think we've made uh quite a bit of progress over the last uh 20 years or so flowers are not the kind of thing that one would ordinarily expect to find in the fossil record it's been known for a long time that flowers can be beautifully preserved in amber and there are several important localities the baltic amber for example from about 40 million years ago and the dominican amber from somewhat a younger time but in the 1970s david dilcher and his colleagues and i was a postdoc in tilter's lab back in the early 80s started to work on fossil flowers from eocene uh beds down in kentucky and tennessee and this is a specimen that i collected years ago on a field trip and this is actually a a mimisoid legume that was described by dilcher and his colleagues and these little things here are the act are the anthers of the stamens with these long filaments and it's clearly related to modern legumes so the approach to kind of get more fossil flowers was initially to collect them in the same way that you would collect other fossils to split rocks basically like a convict and open up the slabs and hope that you could find a a fossil flower it's tough going you have to split a lot of rocks but david deutsche was very successful in finding some material from the mid cretaceous and this is some material now that's not 50 million years old as that legume was that i just showed you this is about 95 million years old from the dakota formation of kansas and i won't go into the detail but simply there's an axis here it's got these uh little pods on and each of these pods has a has seeds inside here a couple of the seeds macerated out and i say i won't go into the detail of it but all i can say is that i think it's very possible to re reconstruct very convincingly uh from this flower this little flowering shoe we know the leaves we have the stipules we have various kinds of teapores we have the scars of the stamens and we have the carpels as they develop into fruits and this is clearly related to the modern family magnoliaceae and you might get a as you know your woody plants might get a hint of tulip tree leaves uh from these uh remarkable lyrium philharmonies and that approach was quite successful there are a few rather important fossils that were described in this way this is just one of them that we called archaeanthus but the real breakthrough in studying these early cretaceous flowers and pushing the fossil record back still further came at around about the same time uh in the early 1980s uh through a new approach and that was basically taking the approaches that several of us had used and had been used for many years in studying cenozoic fossil floras which was to break down the sediment and sieve out the plant fossils and else marie applied this technique to sediments that look a bit like this from southern sweden of about 80 million years before present and to her and everyone else's astonishment and delight out came literally hundreds or even thousands of beautiful fossil flowers these fossil flowers are preserved in various ways but a very common way is that they're preserved as charcoal and charcoal is a an extremely uh gives you extremely fine uh preservation some shrinkage but uh combustion in the absence of oxygen retains the structures beautifully they're very fragile but they're beautifully uh preserved so this material this kind of material has been become the kind of focus of the botanical much paleo botanical work uh in this area and this is another slide from else marian this is actually very early early cretaceous material and it shows that that often the material is pretty tiny and a general feature of these early flowers is that they're really small not as quite as small as this but on the order of a millimeter or two so they're very small and since then those kinds of fossil occurrences have been found in several localities uh throughout the world not only in southern sweden but also in kazakhstan and also in japan and then several new localities in eastern north america and this is just a selection of some of these uh flowers from uh these floras and what i want you to notice is it's just the size these are millimeter scale bars so these are tiny tiny little flowers and these why are these flowers important they're important for three main reasons one reason is that because much of the systematics of living flowering plants is based on flowers if you've got the flowers you can do more in terms of understanding how those fossils might be related to living plants another important reason is that if you've got the flower you might be able to say a little bit more about the pollination biology than trying to do that just from the pollen grains alone and then a third and very important reason that these flowers are important is that you can get the pollen grains out of the anthers and you can therefore start to pin down what the parent plants of some of these dispersed pollen grains are this is a particularly characteristic kind of pollen grain it's part of a large group that we call the norma polys that are common in the late cretaceous of eastern north america and western europe and one of the early breakthroughs from the study of these flowers again by elsa marie friese was the discovery of these norma polly's pollen grains in now a variety of different flowers which we can clearly relate to the modern group of flowering plants that includes the walnuts the hazel's the birches and so on and these are clearly the the precursors of these of these very common uh temperate trees uh that we know today so that's really about the lower the upper cretaceous and the kinds of material that that we can find and i think another very important point that i want to make and this is just for a flora that we studied from georgia around about 80 million years before present we have thousands of specimens but as is often the case a relatively small number of taxa account for most of the material and then you go out on a very very long curve and you've often only got one specimen of a particular kind and that raises some some issues when you've got lots of specimens you can tear them apart and section them and so on and indeed some of these specimens were sectioned in a conventional way using thin sectioning embedding and thin sectioning approaches but if you've only got one specimen or you've only got a few specimens um you'd really rather not destroy it and more recently elsa marie and her colleagues have pioneered the application of x-ray synchrotron microtomography the swiss light source just outside of zurich and this now allows the here's a scanning electron micrograph of a single flower and this now allows digital cross sections of the same specimen longitudinal sections or transverse sections with my colleagues pat herrandine and masamichi takahashi we've been using the same technique essentially although not with quite as good resolution as at the swiss light source this is at the spring 8 facility in just outside of osaka in in japan i'll just show you quickly a couple of preliminary results so here is a here's a lovely flower bud from a locality in georgia here are the the sepals and these are the twisted petals and here is a section through the um the flower down towards the base for the section through the carpals and then the stamens and then the petals and the remains of the the sepals around the outside i'll just show you here's a section near the top through the petals itself a little lower down and a little lower down still and here's a longitudinal section of the same specimen so we have the opportunity now to kind of capture digitally in three dimensions with pretty good resolution the structure of even rare specimens that we don't particularly want to dissect or otherwise destroy so to move on rather quickly with this array of approaches and with the principle established that we could now get fossil flowers from late cretaceous sediments we turned our attention to these early cretaceous sediments and we had two seasons of field work one in the potomac group going back to this these classic localities of doyle and hickey to look for fossil flowers that might have been producing some of these pollen grains or that might have been attached to some of these leaves and we also spent a summer field season in portugal looking at sediments of around the same age so this is what the kinds of localities look like on inspiring sand pits and when you look carefully occasionally you find some muddy layers in amongst the sand you can see a little bit of the organic material through here and then when you sieve that down and break it down in water you can float out the floral parts this is a dish full of stamens and these are the pollen sacks with the pollen still inside them and you get beautiful details i won't go in but here's a little flower the stamens are the same flower the pollen grains of the stained flower the carpels that belong to this flower the stigmatic surface of the of those carpals and then the appropriate pollen grains nestling among the stigmatic papillae on those uh carpals and then a reconstruction of the floral structure and in this case it's uh a particular kind of little inflorescence with a terminal female flower and two little lateral male flowers that are very similar to the flowers of modern boxes the bucks ac so this kind of material we now have from the early cretaceous this particular example is from the potomac group lots of material from the early cretaceous of eastern north america and from portugal too and there are a few other localities that are very important and these famous beds in china that have been producing the feathered dinosaurs and early birds have also produced uh some interesting plant material this is uh architructus described by david dilcher in science several years ago and here's a reconstruction of architructus probably a water plant but it adds to the picture that we get from the meso fossil occurrences and there's another rather important set of localities from brazil of albion age a little younger than the material from china but the material from china also kind of rounds out our picture of what some of these early flowers were like so back to our little stratigraphic diagram we now have flowers from around about 120 million years and maybe even perhaps a little bit older in the oldest uh localities right around when the first triaterate pollen comes in so far we don't have uh flowers that go back into this very earliest phase but many of these dispersed monocolpate pollen grains we do have in reproductive structures from light slightly younger rocks so um what kinds of things are there so we start we now have just to show it in a different way we now have floral material the very earliest mesofloras from portugal of around this age they could contain some interesting things contains a quite a lot of angiosperms of various kinds um but among them are many that we so far can't assign to living groups for one reason or another often we understand them pretty well but it's not clear to which living group they should be related but in a few cases we can assign them to living groups and this one is particularly interesting this is a well-known kind two well-known kinds of dispersed pollen grains and the pollen grains occur inside angiosperm stamens which are born in these little inflorescences and the inflorescences um if that's indeed what they they are have no associated so far anyway no associated uh bracks or or petals or periamp parts associated with them and they're very very similar although a lot lot smaller than to the um uh pollen producing organs of the modern family chloramthaceae and particularly the genus hedy osmond and we also have hedy osman like fruits so in these the point is that in these very earliest floras there are a few things that we can relate to modern groups of flowering plants and as i said at the beginning the results are rather consistent with the molecular evidence on the phylogenetics chloranthaceae is indeed one of these lineages that comes off very early from the main line of angiosperm evolution other material down there is a little uh water lily flower with uh carpools in the center stamens around the outside and and t pours and i won't go into the details but we can relate this i think now very securely to modern water lilies based on various details of the flower including now with the new uh tomography techniques the ability to look inside that single specimen and see the seeds sitting inside the gynesia and this is the same specimen in longitudinal view and you can again see the seeds in position and this helps secure its relationship to the nymphaeales so what do we have in the let's say by around 95 million years what groups can we recognize that we can recognize in in in living floras well this is the sort of base of the angiosperm tree we have good representation of of uh this nymphy alien clade we heard that we have um in this meeting we reported the presence of monocots good reliable monocots from these very early floras for the first time and those monocots this is the mon a breakout of the monocot phylogeny belong to the araci which falls in this very early clade of the monarchs we have good evidence of chloranthy as i've just mentioned perhaps piperellis and and good strong evidence of both of these groups laurely's and magnolias by about 95 million years and the initial diversification of eu dicots this is a breakout of the eu dicots and we have material that's clearly referral to this group and also to this group so in general what we see is very consistent as i said with the uh molecular phylogenetics so for the last 10 minutes i'd like to deal with this fourth and then very briefly with the fifth question where did flowers come from and if this is what we mean when we talk about the abominable mystery it's still abominable we haven't made much progress at all on this one for a moment about 15 years ago it seemed like we had it cracked and then it all came apart again and it remains to be seen how it will get put back together one of the issues here is that we have had increasingly the application of molecular data not just to relationships within flowering plants but how flowering plants relate to other groups of seed plants unfortunately there are only four other groups of living seed plants cycads ginkgo conifers and a peculiar group called the nitaily is a very important group that we'll talk about in a moment and to cut a long story short the morphological analyses that have been done point towards a close relationship of angiosperms and knee tailees and the molecular work that's been done including by many people in this room points to anything but a close relationship uh with between angiosperms and knee tails and this is a particularly tricky uh problem because we've got a sampling problem whichever way you look at it when we deal with living plants we can get an enormous amount of information the whole genome if we want but what's for a potentially rather poor sample of all the seed plants that have ever lived if we want to include the fossils we can get a better sample but then we're restricted to using morphological data and either way we're going to be stuck with this missing data and at the moment uh what we've got is two different scenarios emerging a molecular set of results that point in one direction and a morphological set of results to point in another uh direction and for the last few minutes i just want to tell you a little bit about uh some of the morphological uh work here is a uh a fairly recent uh tree based on morphological data uh that evaluates not only the relationship of flowering plants to living seed plants but also their relationship to fossil seed plants and angiosperms start around here and their closest relatives are an extinct group called the beneditales this strange group the knee tailees and then on out and this is the morphological hypothesis that's held sway now for nearly 30 years what do these plants look like well very interestingly and back to our main topic uh the benetti taylors have flower-like reproductive structures which is i think pretty fascinating and how they exactly work and what they're exactly like is is um something that we're that we're still uh working on but this is a beautiful classic specimen from the stockholm museum collections uh showing these gorgeous uh sort of periamph like structures but inside are some pollen producing organs and it's not entirely clear what they are from a morphological standpoint and some of these uh bene titales here's another one um from the middle jurassic of yorkshire um appear to have been bisexual producing seeds in the center and then pollen grains around the outside and then some kind of perianth around the outside of that and then in needles there's a hint of flower-like reproductive structures in the genus well witcher with an aborted ovule in the center surrounded by um pollen producing organs surrounded instead by some brax the problem though is that we really don't know how to compare these things we don't really know what these different structures represent how these pollen producing structures should be compared to these pollen producing structures how these ovules should be compared to these ovules and how they should be compared to the structure seen in other seed plants and in flowering plants it's worth going back to our thinking about the components of the angiosperm flower and i'll just say a little bit about this issue the ovule with two surrounding coverings and i have to do it very very quickly then i'll maybe touch on those okay if we take a if we take a wheat grain uh for for example i mentioned we've got one structure inside uh another we start with a modified megaspore which is inside the nucellus surrounded by an inner integument an outer integument enclosed within a carpal so we've got a number of layers and figuring out how these layers should be compared to the layers that we see in other seed plants is really the crux of one of these problems accounting for these layers is really the story of the origin of seed plants which took place 350 million years ago accounting for these layers is really the story of the origin of flowering plants which seems to have taken place around 135 million years ago let's say a little bit about the new tales this is baines one of the first collectors of well witcher in the namibian desert it's a self uh portrait of him with this remarkable plant well with you it's a little easier to get to these days although my son seemed a little less impressed than i was with these rather strange plants growing in the namibian desert at queue we have uh some of the original specimens and this is the original specimen brought back by baines and we also have some of his field sketches that were turned into remarkable illustrations again by that great artist fitch and well witcher is related to uh these two genera ephedra and neetum neetum a tropical climber ephedra a plant with a kind of switch habit of dry places around the world now this group the nitailis do have very distinctive uh pollen grains with rugby ball shaped pollen grains with these striations and these kinds of pollen grains do occur in the fossil record and this is back from a locality in virginia at around about um 120 or a little younger million years and back in 1987 gary upchurch and i described a little plant which we assigned to the knee tailees you can see it's tiny this is a millimeter scale we had lots of specimens and from that same locality but not from the same bed elsa marie and kai and i have recently and katarina ridden have recently been able to describe seeds that clearly belong to the knee tailees too and they are these seeds have two layers around the outside here's the outer layer with the inner layer and the seed itself gone and they have a characteristic ephedroid pollen grains in the micropile and these seeds occur not only in eastern north america but as you can see from this name they also occur commonly in early cretaceous floras from portugal they're very comparable to the seeds of modern ephedra and this is the hard outer layer out of which pokes the micropile the entrance to the seed it's itself and we've got lots of other records of this group now appearing back in 1987 there was almost nothing known of the fossil record of new tales and i don't have time to go into the details but now we have a lot and i particularly like this specimen from the albion of brazil so these are these deposits that are of similar age to some of the ones that we're looking at in portugal east and north america very similar to the cones of modern female cones of modern world witching so i kid my colleagues who write uh paleo botanical textbooks that in 1983 the nitaily is one of five living groups of seed plants got a quarter of a page in a book of 396 pages in the next edition in 1993 it got a page and a half i hope when they do it next time that they'll get a chapter we've got enough uh we now know enough about the fossil history of this group to deal with them in a bit more detail but we have um a number of seeds in these early cretaceous uh deposits that we're now able to study using this new uh x-ray microtomography uh technique and they reveal their structure very nicely and the central point is that they have these two layers here's the micropile itself and then here's what we call the outer envelope and this is a section through and you can see the micropile protruding out and then the outer envelope this structure is uh extremely consistent through a huge range of these seeds that we now have from the early cretaceous i'll just show you some of them here's another one that shows very beautifully the extension of the integument through up through this region forming the micropile up to the top with this thick outer envelope and it seems that the micropile itself becomes blocked by the growth of the cells in this area but you can see very clearly the two layers at the very top there and you can see that the inner layer gets a little thicker by the time you get down to this level and as you go down still further it becomes harder to separate those two layers well i won't go through them in detail but we have many many seeds of this kind so on the one hand these seeds compare very nicely with the fossil seeds of ephedra but they also compare very nicely with these seeds of another group that have different kinds of pollen grains associated with them pollen grains of the eucomia dieties type enigmatic gymnosperm pollen grain that occurs in the micropiles of these seeds so we now have this group of fossil seeds that are simmer on the one hand to the seeds of ephedra similar on the other hand to the seeds of these eukobia dietitians producing plants and they have various kinds of pollen grains in eucomia dietitis or even simple monocolpate pollen grains and this little beautiful photograph of elsa marie shows the some of the diversity of these seeds what's interesting however is how these seeds can be compared to the seeds of some of the other seed plants that are around and in particular in these lovely dinosaur reconstructions you often see these plants that look a bit like christmas puddings down on the the floor of the forest here and with flowers on the surface and these cycad-like leaves coming up and these are a fossil plant called cycadioidea and they're some of the best-known fossil plants of the uh beneditales so we went back to look at one of the very best known of these benetitalian flowers cycadioidea moriari which is from uh from france actually and um it's really quite well preserved and here's a a single seed and it can be studied with that same tomographic technique and to cut a long story short it shows the same structure the tubular long tubular micropile and then this thick outer layer even with the same kinds of cells that we see in some of these dispersed seeds so what's the significance of all of this i just slipped through a couple of slides and then i'll come to the conclusion the significance is that that these new fossil seeds that we have on the one hand are similar to seeds of living ephedra on the other hand are similar to seeds of bene titales and also compare to seeds of eukomia dieties type plants and it's a very unusual seed structure it has this inner layer with a hard outer layer so the question really is does this seeds structure which um under the analysis the morphological analyses that have been uh put forward here seems to unite this group does this seed structure really extend to angiosperms and is the second layer that we're picking up here in benedi's knee tailees and this group of fossil seeds is that the same second layer that we see in the seeds of angiosperms and that i think is the real question so where do we go from from here well first of all we need to reconcile this interpretation with the seeds of some other extinct plants this is pentoxilon from the early cretaceous of india and uh i'll wager that it will end up being interpreted in the same way we also have some other interesting and intriguing fossil plants this is caytonia how does this double layer that we see from the jurassic of yorkshire how does that compare with what we see in caytonia and other seed firms so and then of course if this hypothesis was true and if this second layer is is a feature that unites these groups then we ought to be able to reconcile how their pollen producing structures fit together and we know a little bit about that but it's not something that we can readily solve at the moment but it's something that can potentially be solved by looking at more of this fossil material so finally then i just say a few brief words about about insects um obviously the diversification of flowering plants has been linked in very large part to the diversification of insects about a third of the approximately 1.5 million species of insects known feed at some point in their lives on plants and most of the plants today are angiosperms and so there's clearly a strong ecological association not just on the pollination side but also in terms of herbivory there are various ways to look at it that i won't go into but i just want to make uh one important point i think some of these groups of pollinating insects have a fossil record that goes back way before angiosperm beetles and flies in particular we have evidence from living groups such as ephedra and cycads of insect-mediated pollen transfer in those groups and as i've said we have evidence of flower-like reproductive structures among some of the seed plants that predate anxious berms in the fossil record living flowering plants of course show a wide range of showy adaptations that we've spoken about over the last few days and wonderful work being done uh on aquilegia reported uh at this meeting that definitively show very close relationships between the evolution of these flowers and the evolution of insects so back to uh the lotus and i just finished this is from darwin's home this was a platter that he was given by the wedgwood family it shows the lotus there it shows the water lilies uh that we've been talking about um i wonder if he contemplated this and as he was thinking about his uh abominable uh mystery i don't think we've solved the aspect of where flowers come from but i think we've solved some aspects of uh the abominable mystery in terms of uh timing those of us in chicago have great hope that we will be able to solve this problem in the fullness of time and that the fossil record will ultimately reveal the very gradual change that we can believe in thank you very much so that's fabulous thanks very much indeed peter we have time for a few questions and if people would like to ask questions if they could stand up so the camera can spot you and if you could give your name so that we know who you are so can i call for any questions at the back there rupert sheldrake thank you for this wonderful lecture very illuminating can i ask what may be an impossible question have you any idea whether the first flowering plants were herbaceous trees or shrubs is there any indication at all about what kinds of plants they were we could say a little bit about that and we can uh we can do we can make some guesses about it from two different directions one way is to look at the modern relatives uh of the plants that we know uh were present in the early cretaceous and by and large they're herbaceous to shrubby things they're not large trees so at this meeting we reported aroid monocots obviously herbaceous chloranthy woody plants but not generally uh large plants water lilies herbaceous of course the other argument that we can make is um it's very interesting that there's not much angiosperm wood in the fossil record from the cretaceous whereas there's lots of angiosperm water in the fossil record from the cenozoic obviously angiosperm wood preserves extremely uh well that's not to say that there were no large angiosperm trees in the lake cretaceous but they don't seem to have been diverse and they don't seem to have been very common we do have wood fragments that we can relate to the modern bay laurel family loracy and we have other wood fragments that we can refer to the modern plane tree family that we see growing outside but in general there's very little evidence of large angiosperm trees uh in the late in the late cretaceous a question just there and there's a microphone just on its way thank you my name is david plumstead and i'm at the moment researching uh various aspects of the development of um of the honeybee and relay trying to relate it to its history and how it's developed and what might account for the current hiatus in the uh worldwide colonies uh of bees in trouble uh and i just wondered if uh if you could kindly aim me at the sort of literature that i could look at because obviously the bee preceded the plants that we've been talking about today and i just want to see how how it came in and how they got together thank you yeah i mean um on the if i've had more time to talk about the insect side i said that we had beetles and flies probably pollinating kinds before angiosperms lepidoptera with chewing mouthparts appeared around the jurassic cretaceous boundary and then a little later true uh lepidoptera with uh with the sucking uh mouth parts and as we at the regard to the bees the earliest i mean bees obviously don't have a great fossil record but they do have a fossil record in the cenozoic the earliest bee is from new jersey amber and it's thought to be around about uh it's thought to be in the top of the of the late cretaceous around about 70 million years it's clear that the bees and lepidoptera if you wanted to choose two groups that really probably have evolved in concert with flowering plants those would be two very obvious insect groups that that you would point to so we don't have uh any record of of bees prior to about 70 million years a question there um paul rymer imperial college um a couple of weeks ago i heard a talk by graham bell who um reviewed some of the work by darwin and one of the claims that he made was that darwin actually led us astray by saying that evolution was slow and required long amount of time and in fact his argument which he pushed forward very strongly was that we have strong selection operating and quite rapid evolution and he provided some case studies can you comment on that please yes i mean i don't think that i don't think the uh that what the fossil record will provide and this really last slide is a little bit of fun but i mean you don't see uh gradual evolution in the fossil record in general um what you see is things coming into the fossil record and staying relatively unchanged for long periods of time i mean i i would i would certainly subscribe to to the sort of punctuated equilibrium view as being close to what we see it's harder in flowering plants to judge this because obviously if you're dealing with a trilobite or something you've got a reasonably whole organism to look at but with flowering plants where you've only got a pollen grain and a leaf and so on you may miss changes that are occurring in the other organ systems but what i think we could expect to see in the fossil record and where my little bit of hope comes in because we see it in other areas where we've got a group of defining features like the two integuments like the carpal like the stamen in other parts of the fossil record we can see those features coming in sequentially in other words we find groups that have some but not all of the defining features of the modern group that's what one would hope to find here with angiosperms so we might expect to find a group that has two coverings around the ovule but that hasn't yet got a carpel and that hasn't yet got a statement that's the experience that we would bring from looking at other aspects of the plant uh fossil record so we won't see gimme sperms changing imperceptibly and gradually into flowering plants but i think what we could see is some of these intermediate transitional forms and i think we've probably already described them without actually recognizing what they are that would be my guess time for a final question if anyone has one just in the front in the aisle there i just wondered if you it's martin sands robert hannigan whether the microdermography had shown up any of the vasculature and if it has has it helped in any way to interpret the relationship between the floral parts we do see uh vasculature both in in the flowers and also uh in the seeds it doesn't really provide us with any additional evidence one way or another at this at this point but certainly the technique and particularly um uh the very high resolution approach from uh villigan from the swiss light source which has got better resolution than we than we've been able to obtain in japan you can certainly see the vascular truth the vasculatures there in these charcoal fossils you will definitely be able to see it thank you i'm afraid we have to bring it to a close there but let me just finally thank peter again for a wonderful talk i think one of the great things about the research that we've heard this evening is that it's research that is sort of based on techniques that go back 100 150 years that darwin would have been would have been aware of but then brought bang up to date with these wonderful new ways of visualizing micro fossils something peter hasn't talked about the complex statistical analysis that goes into building those phylogenetic trees and then the link with the new molecular techniques so it's a wonderful link between bang up today cutting edge biology but then going all the way back to things that darwin would would have recognized so peter thank you again for a really fascinating talk you

Channel: The Royal Society

Views: 33,227

Keywords:

Id: POv-TPX7REw

Length: 63min 25sec (3805 seconds)

Published: Tue Dec 24 2013