, , ,

10346553834_ee9bad1e12_hFlowering plants, including the kinds we associate with edible fruit and berries, only diversified near the end of the time of dinosaurs. For most of the Mesozoic, anything like avocadoes and grapes, blackberries or raspberries simply did not exist. Accordingly, paleoartists don’t depict these in their Mesozoic scenes.

But this does not mean Mesozoic animals didn’t eat sweet fruit-like plant parts, as flowering plants are not the only ones that produce them (photo: Juniperus communis, Sarah Gregg/Flickr).

Fruits and berries are fleshy, energy-rich (usually sugary, sometimes fatty) plant tissues that the plant uses to bribe animals into distributing their seeds. In an optimal situation, the animal will swallow the tasty morsel, seeds and all, and later drop the intact seeds into a new habitat, along with a convenient serving of manure.

The tactic is so useful it has been independently aquired all over the family tree of plants. And not just flowering plants. Only the reproductive structures of flowering plants are properly called fruit, however. The technical term for non-angiosperm ones is false-fruit or diaspore.

Ginkgoes are perhaps the oldest group of plants to produce edible false-fruit. They have been around since the Permian, though there is plenty of variation in the reproductive structures of prehistoric ginkgoes.


Seed coats of the modern maidenhair tree, Ginkgo biloba, are yellowish and said to smell (and probably taste) like carrion or vomit. The toxic seed is contained within a soft, edible if smelly husk (photo by Fritz Geller-Grimm/Wikipedia, from a park in Germany).

As the tree is almost extinct in the wild, it’s hard to say what kind of animals are supposed to eat the fruit. According to locals, the fruit of the last (possibly) wild trees in Tian Mu Shan Reserve, Eastern China, are occasionally eaten by small carnivores. These include palm civets, raccoon dogs and even leopard cats. The seeds survive the journey through a carnivore’s digestive system, though actual new seedlings are extremely rare. Sadly, little actual research has been done on ginkgo seed dispersal.

Squirrels are disgusted by the smelly flesh, but often extract the seeds and hoard them, apparently indifferent to the toxins. Many of these seeds will get eaten, but some are forgotten in underground stashes and germinate.

If the fruit of Mesozoic ginkgoes were similarly disgusting, they could have been eaten by a wide variety of carrion-loving animals from dinosaurs to pterosaurs. It has also been suggested they were hoarded by multituberculates, superficially rodent-like Mesozoic mammals only distantly related to anything living today.

717px-Ginkgo_apodesThe fruit of modern Ginkgo are larger than those of many Mesozoic forms, though there was plenty of variation. Ginkgo adiantoides known from the latest Cretaceous is apparently morphologically identical with modern trees. Ginkgo apodes, from the Early Cretaceous Yixian Formation, has tiny fruit less than a centimeter across (photo by B. M. Begovic/Wikimedia Commons). Tiny fruit weren’t necessarily much of a problem even for moderately large seed dispersers, since the fruit have a habit of falling down and rotting in great heaps.

Taxus baccata, Yew

Another lineage of plants fond of edible parts are conifers. While the hard, scaly cones of pines and spruces may be more familiar, many conifers grow fleshy female cones that are eaten by animals such as birds, squirrels and sometimes humans. The fleshy layer is actually a highly modified cone scale (photo: European yew Taxus baccata, Jonathan Tyler/Flickr).

While botanically different, these female cones resemble berries in all practical purposes. They are present in a number of most ancient living lineages of conifers: Taxacaea and Cupressaceae, fossils dating all the way from the Early Jurassic, and Podocarpaceae, which is even older.


Japanese plum yew (Cephalotaxis harringtonii) false-fruit are about 2,5 cm long. Photo by François Guibert/Flickr.

The taste and colour of fleshy cones varies, as do the animals they attract, but they generally prefer dinosaurs. The pungent, blue ‘berries’ of the common juniper (Juniperus communis) are mostly dispersed by thrushes (Turdus spp.), at least in the Mediterranean. The sweet, bright red cones of yew (Taxus) are similarly favoured by trushes and waxwings (Bombycilla). The kakapo (Strigops habroptila) of New Zealand will only breed when the orange-red cones of the podocarp known as rimu tree (Dacrydium cupressinum) are plentiful.

Conifers are generally conservative, and haven’t changed all that much since the Mesozoic. Some Mesozoic conifers are actually placed into the same genera as modern fruit-bearing ones. For example, the yews Taxus and Torreya are known from the Early Cretaceous Yixian Formation. It is likely at least some of them also produced similar cones.

While there were no trushes and waxwings in the Jurassic and Cretaceous, a variety of small herbivores and omnivores would probably have been willing seed dispersers: small theropods and ornitischians, early birds, forest-dwelling pterosaurs, lizards and early mammals. As there were no true fruit trees to compete, perhaps conifers or other gymnosperms also, at some point, explored the niche of large fruit dispersed by herbivorous megafauna: a niche today filled with such plants as mangoes, avocadoes, plums and durians. This is pure speculation, however: I’m not aware of fossil finds suggesting big conifer false-fruit existed.


False-fruit of cultivated Gnetum gnemon in Malaysia. Photo by Ahmad Fuad Morad/Flickr.

A third, separate group of fruit-bearing gymnosperms are the Gnetales. These comparatively small plants are members of an ancient gymnosperm lineage. Today they contain three genera: the vine-like Gnetum, shrubby Ephedra and desert-dwelling Welwitschia. They diverged from the lineage leading to other gymnosperm somewhere around the Triassic. While their early fossil record is fragmentary, fairly modern representatives are known from Early Cretaceous. For example, plants classified as Ephedra are known from multiple locations from the Early Cretaceous Europe, Asia and North and South America.

10134787224_41adcdb49f_kOf gnetophytes, both Ephedra and Gnetum produce fruit-like cones. The fifty or so living species of Ephedra can be classified into three dispersal syndromes, and different lineages hop between them apparently easily. There are species with dry, winged seeds that are dispersed by wind. Those with thicker, but still dry husk are cached by rodents. The third group includes species with fleshy, brightly coloured cones that are dispersed by frugivorous birds, such as pictured E. distachya from Ukraine (photo by: Marek Michalski/Flickr). Interestingly, a fossil of a fleshy Ephedra cone was described from the Yixian Formation in 2010. Ephedra ‘berries’ thus date from the Early Cretaceous at least.


Finally, there are the cycads. Paleoartists draw cycads all the time, being as they are among the most famous of ancient plant groups and a major part of many fossil floras. They rarely feature what the mature, seed-bearing cones of female cycads look like. They come in an amazing variety of yellows, reds, oranges and purples. And they are huge: they often weigh tens of kilograms (photo by: Steve Crane/Flickr. Probably Encephalartos villosus).

Why exactly are they colourful? Because the seeds have a thin, but tasty fleshy husk – they are false-fruit too. Today, cycads are not particularly good dispersers. According to a study conducted in Northern Australia, fleshy parts are eaten by smallish animals such as opossums, but the seeds are most often left uneaten around the parent plant, where they won’t survive. Few seeds will make it ten meters away or so, where the seedlings have a chance, and form cycad groves.

While the groves are long-lived, isolated seeds that somehow manage to find themselves in new habitats are less lucky. Cycads are either male or female, and they need others of their kind nearby to reproduce. Basically, many living cycads are confined to small habitat patches and rarely if ever are able to colonise new areas. They are stuck, in a way: in the absence of truly big animals, evolving slightly smaller seeds would probably only increase the amount of isolated seeds ending here and there.

Enter large animals. Now-extinct megafauna – were it giant marsupials or sauropods – were able to eat dozens of seeds at a time and move them much further away than a small mammal ever could, planting a whole new grove at once. Perhaps the best modern dinosaur analogues are African elephants, which are known to eat whole cones of the cycad Encephalartos poggei, dropping a huge pile of viable seeds a day or so later. Possibly thanks to the elephants, E. poggei is one of the most widely distributed and common cycads in existence.


Mammals, aside from the vervet monkey above (Pim Stouten/Flickr), don’t have much use for bright colours. The colours might actually be a relict from the time big animals had tetrachromatic vision – from the Mesozoic, that is. Even if they weren’t, there’s no reason to depict animal-dispersed plant parts any less colourful than they are today. If anything, they should have been more so.

While there were no flowers before the evolution of angiosperms, the Mesozoic world was not all greens and browns. At a time when dominant terrestrial animals of all groups had full colour vision (as opposed to the bichromatic vision of most mammals), colouration was probably an important means of communication between plants and animals.

This article again wrote itself far longer than I intended it to be, but hopefully readers have managed to reach the end of it and continue their days with some new ideas about paleoart. Next: Mesozoic plants that look like grass.

First part:  Not All Ferns Look Like Ferns.

References and further reading:

The Gymnosperm Database: Ginkgo biloba.

Connie Barlow (2000) The Ghosts of Evolution. Basic Books. Features a chapter on ginkgo seed dispersal.

Willis & McElwain (2002) The Evolution of Plants. Oxford University Press.

García 2001: Effects of seed dispersal on Juniperus communis recruitment on a Mediterranean mountain. Journal of Vegetation Science.

Hollander et al. 2010: Evolution of seed dispersal in North American Ephedra. Evolutionary Ecology.

Yang & Wang 2013: The Earliest Fleshy Cone of Ephedra from the Early Cretaceous Yixian Formation of Northeast China. PLOS One.

Hall & Walter 2013: Seed dispersal of the Australian cycad Macrozamia miquelii (Zamiaceae): are cycads megafauna-dispersed “grove forming” plants? American Journal of Botany.

RedOrbit: Cycads Evolved To Grow In Groves With Seed Dispersal By Large Frugivores.

Northern Territory Government: Cycads in the Northern Territory.