Evolutionary Anachronisms in The Western Palearctic – Part I: Puzzling Pomes
This is a guest article contributed by Lars Hendrik Großgott, a bachelors student of Biological Diversity and Ecology at Göttingen University, who can be found as his profile on X, formerly Twitter, @GottGro.
Introduction
There seems to be a deep misunderstanding concerning the Quaternary extinctions: the notion of fathoming extinct animals in general as something “other”. For example, there is a long-standing tradition in palaeontology of giving extinct creatures archaic names. Think of Mammuthus primigenius, the “first mammoth”, Coelodonta antiquitatis, the “hollow-tooth of the good old days” or Palaeoloxodon antiquus, the “ancient paleo-elephant”. To me, this exemplifies how we tend to think of them as prehistoric peculiarities, and not as what we have begun to understand these animals for; functional and integral elements of ecosystems essentially comprised of the same elements as today. We now know that the time since the extinction of the straight tusked elephant (Palaeoloxodon antiquus) is significantly longer than what our forebearers thought of as the age of Earth itself. Yet, we also know that even this long period, roughly 28.000 years, (1) was not enough to erase the scars the pachyderm’s passing left behind. We can link the straight-tusked elephant’s extinction to a whole wave of large mammal extinctions, and we can trace the crippling effects that this demise had and continues to have. But in order to truly fathom what the extinction of the European megafauna has meant for the ecosystems they no longer traverse, it may be helpful to turn to something more tangible. Fruit dispersal is one of the most obvious ecosystem services performed by large mammals. Therefore, let us shed a light on some of the plants that, deprived of dispersal partners, have lingered in the Western Palearctic’s vast realms. Slowly fading into obscurity, some have been retrieved by the now unassailably dominant megafaunal animal of the region - us. These are the evolutionary anachronisms of western Eurasia and North Africa, with their puzzling pomes, proboscidean pulses and seemingly inexplicable ecological requirements.
(For an overview of extinctions in Europe, see The Extinction’s three-part series here, here and here.)
First, an important question needs to be addressed: what exactly is an evolutionary anachronism? In their 1982 paper Neotropical Anachronisms: The Fruits the Gomphotheres Ate, Daniel H. Janzen and Paul S. Martin defined it as a trait of a plant that is inexplicable unless seen in the backdrop of its evolutionary past. For instance, Gymnocladus dioicus, a tree in the legume plant family Fabaceae that is native to the eastern United States, produces seed pods that are poisonous to mammals, unbreakable to rodents and impervious to water, yet depend on all of these for dispersal since the seeds, the largest in the continental United States, are too heavy to be carried by wind anywhere far. As a result, each year the parent tree will produce pods that fall to the ground, where they slowly decompose over the years, even in seemingly natural habitat. This is odd because the fruit of any plant is always intended as a diaspore. Plants have outbid each other over millions of years in attempts to produce the most sophisticated designs that will allow their unborn offspring to travel and germinate a long distance away from the parent. So, if the Kentucky coffeetree, as it is also called, fails so miserably at dispersing seeds away from the parent tree, despite an elaborate diaspore, one is compelled to ask, why?
As it turns out, the most likely answer for Gymnocladus’ failure at distributing offspring is a lack of dispersal partners, which are now extinct (2). Species are products of the past. Or as Richard Dawkins put it in 1998 “They are beautifully built for survival in their ancestors’ environments”(3). Because any species is like a bucket of characteristics, many of which may have fallen out of selective favour a long time ago, traits that are observable today may be the result of selective pressures long-obsolete. Once a trait has fallen out of favour, it may still cling to an organism because it is not much of a hindrance either, or because a species did not have the time or opportunity to “catch on”.
Fig 1. Kentucky coffeetree pods, human hand for scale
Terms of use: This image is licensed under a Attribution 2.0 Generic. It is attributed to Вікторія Тротнер(Приймачук), and the original can be found here. The image is unedited
In their original paper, Janzen and Martin limited their scope, and thus the application of the term, to fruits from the subtropics of the American continent. This traditional approach I will transfer to the Western Palearctic, but otherwise adopt. However, since then the term has acquired a broader meaning that covers other characteristics as well. For example, the speed of the pronghorn (Antilocapra americana) (up to 88.5 km/h (55 mph)) has been associated with the extinct American cheetah (Miracinonyx spec.). In the absence of such a pacy predator, the pronghorn’s speed seems oddly exaggerated for the modern ecosystems of the American Midwest (4). Similarly, the regeneration requirements of light-demanding tree species in Europe (5) and the low abundance of forbs in contemporary grasslands, as compared to their prehistoric counterparts (6) have been explained by a lack of herbivory in modern ecosystems. In fact, by some measures, temperate woody plants at large can be considered anachronistic, since most of them originated and diversified in savannah ecosystems. Most temperate woody taxa are characterised by vigorous resprouting (coppice, pollards), rapid height growth (i.e. trees) or abundant basal shoots (i.e. shrubs), which probably evolved as responses to browsing and fire (7). Oaks (Quercus), for example, increased dramatically in diversity in the Late Miocene, concurrent with grasses (Poaceae), ruminants and the rise of savannah ecosystems (8).
The biogeographic region of the Western Palearctic is, for all intents and purposes, usually defined as encompassing Europe, the North African subtropical coastline and the Atlas Mountains, the Caucasus, Transcaucasia, Anatolia, Greater Syria and Mesopotamia. In addition, and in deviation from the usual definition I will include Hyrcania, a stretch of land lying between the Caspian Sea in the north and the Alborz Mountains to the south. This is because in many aspects, the Hyrcanian species assemblage is very similar to the European one. Many European taxa either have their easternmost occurrence in this region or are replaced there by closely related endemic equivalents. Therefore, I have opted to include this region, not least because it is home to some of the most fascinating anachronisms in question.
Fig 2. A map of the Western Palearctic. For the purpose of this article, a slightly different definition is adopted, which does not protrude so far into the Sahara and instead includes Iranian Azerbaijan and adjacent Hyrcania in the east (indicated by the black oval).
Terms of use: This image is licensed under a Attribution 2.0 Generic. It is attributed to Jimfbleak, and the original can be found here. The image is edited.
Before embarking on our journey through the flora of western Eurasia and North Africa, however, it is important to note that anachronistic traits are no binary affair. In one way or another, every living thing can be considered anachronistic, to varying degrees (9). In the Northern Hemisphere at least, examples as extreme as the Kentucky coffeetree are rare. More commonly, one will find examples where a fruit is distributed to some extent by modern dispersal partners, but their service is insufficient and the fruit seems overbuilt for it. In other cases yet, a species’ legitimate disperser is no more available due to local eradication, making the species a local anachronism. In parts of Africa and Asia, for example, there are many anachronisms in the making as the world’s megafauna continues to dwindle in range and population (9). As we shall see, examples of substantial anachronisms, local anachronisms and of plants that are “merely” inefficiently dispersed can at present be found in the Western Palearctic as well. In this region, however, we find the peculiar situation that many of the anachronisms in question have been domesticated and utilised by humans for such a long time that they are not usually understood as natural species. Their existence has become synonymous with cultivation. This makes identifying them as anachronisms more difficult than it would be in other regions, as the natural origins of many of these species are usually overlooked and obscure.
Finally, I will use teleological language in this essay, implying that certain fruit shapes and characteristics are intended for certain animals to consume. This is not to say, however, that I believe that any of the plants discussed hereafter knows what kinds of animals it is trying to lure. Rather, the linguistic implication of intent is simply a convenient way of describing how natural selection favouring certain phenotypes will result in similar outcomes for distinct lineages. As a result, fruits that were originally dispersed by the same or similar dispersal partners may share certain similarities, even if the species in question are not closely related. This is why it is possible to identify dispersal syndromes such as anemochory (wind dispersal), vertebrate dispersal syndrome, avian dispersal syndrome and mammalian dispersal syndrome. Mammalian dispersal syndrome in turn comes in two varieties: epi and endozoochory. Epizoochory is what plants like the burdocks (Arctium) do, the seeds stick to fur (or clothing) and are carried away. Endozoochory, on the other hand, is internal dispersal, the seed being enclosed in a fruit. It is this last type of dispersal that shall henceforth be our concern. Now, without further ado, let us embark: who are the elusive anachronisms?
The Sorb Case
Let our first example be Cormus domestica, the service or sorb tree. Superficially similar and closely related to the rowans (Sorbus s.str.), it differs from them in that it has pursued an evolutionary pathway away from avian dispersal towards mammalian dispersal. In the process it has become large, possibly the largest of the Rosaceae (the rose family), rivalled in Europe only by the chequer tree (Torminalis glaberrima). Indeed, in relation to Sorbus species, the sorb tree is massive in virtually every regard. In dense forest, it can grow to a height of 35 m, and a crown height and width of 20 m in a free-standing position. What’s more, it can attain an age of 500-600 years. Rowans pale in comparison, often being shrubs or multi-stemmed small trees that rarely reach a height of more than 10 m and mostly live for only a few decades, rarely 100 years. Fruit size differs widely, from 1,5 to 5 cm in diameter, but while even the smallest are larger by magnitudes compared to Sorbus species, the largest overlap with wild pears (Pyrus pyraster) and crab apples (Malus sylvestris) in size (10), (See figure 6).
(Note: here and throughout this essay the updated, narrow definition of the genus Sorbus is adopted, which elevates several of the taxa formerly lumped together under Sorbus s.sl. into the status of genera in their own right (11). Nonetheless, the traditional definition is still widely used, with the service tree listed as Cormus domestica and the chequer tree as Sorbus torminalis.)
Moreover, again deviating from the Sorbus-norm, the fruit of the service tree is clearly not meant for birds, but for mammals. First of all, the fruit is large, too large for most of the notorious avian dispersal guild (thrushes and other passerines) to swallow whole. Then, where its Sorbus kin retain them on the branches through the winter, for birds to snatch, the sorb tree drops its fruit to the ground once they are ripe, ready for rummaging mammals to gobble up. Third, and most importantly, the fruit looks dull, but smells intensely. Avian-adapted fruits are conspicuously red, blue or black but virtually odourless, as epitomised by wild cherries (Prunus subg. Cerasus.), rowan berries (Sorbus s.str), blackberries (Rubus fruticosus agg.) or blueberries (Vaccinium spp.), as birds rely heavily on vision. Mammalian-adapted fruits, on the other hand, are often inconspicuous in colour but emit a strong odour, since the olfactory sense tends to dominate in mammals (12). This distinction is not complete, of course; brown bears (Ursus arctos) are diligent consumers of blueberries, and elephants have been observed to carefully pluck blackberries, given the chance (9). However, in general, fruit can be assorted into one category or the other, even though the picture is more complicated in regions where monkeys, apes and fruit bats play prominent roles in dispersal. In fact, the longstanding presence of Primates (barbary macaques (Macaca sylvanus), humans (Homo spp), and many others since the Miocene) in the Western Palearctic means that their contribution cannot be ruled out, and was probably substantial in some cases (13). To go back to the sorb tree, mammal dispersal is also consistent with another observation: for long, it was unclear how the sorb tree could be propagated artificially, to the point where it was deemed impossible to grow in cultivation. The answer, then, came in the form of very specific requirements the seeds demand to be fulfilled if they are to break dormancy. These requirements include a prolonged exposure to cold and moist conditions known as cold stratification. The seeds must be stratified for at least 8 weeks before they will germinate. The best results, however, have been observed when the seeds are first washed, then placed in warm water, then stratified and finally transferred into a substrate of peat (14). Rumour also has it that placing seeds in your mouth before treatment in warm water aids in germination. This sounds precisely like the kind of germination requirements a tree that depends on dispersal by large herbivores in the temperate zone would develop. First, the herbivore would consume the fruit, thereby separating seeds from flesh. Then the seed would pass through the mouth, oesophagus, stomach and intestines, where it would be exposed to acids and enzymes. Finally, it would be deposited in a pile of faeces, a cool and damp environment, to await the arrival of spring.
Like the pedunculate oak Quercus robur, the service tree is relatively slow-growing and appreciates warmth, develops a deep root system and a fissuring bark when maturing and is generally impervious to biotic and abiotic stress, although light-demanding (10). All of these are characteristics that make it well-suited to a life in open, grazed environments. The distribution of the sorb tree is relatively extensive, with core regions in the Balkans, the Italian peninsula, France and the eastern Iberian Peninsula, and occurrences further afield in Central Europe, Anatolia and the Atlas Mountains (10). These days, however, the sorb tree is in trouble. It is rare and threatened throughout its range, suffering from insufficient progeny (15). In this regard it is interesting to note that even though the geographical centre of the tree’s distribution lies in Mediterranean Europe, the largest, oldest and most magnificent individuals can be found in temperate Europe. It is here where the species seems to find optimal conditions, resulting in significantly longer life spans (10). This suggests two things: First, that the sorb tree, even though it is able to cope with Mediterranean conditions, is primarily adapted to temperate climates and second, that its modern distribution pattern is likely to be the result of insufficient long-distance dispersal since the Last Glacial. This means that its modern distribution centres are likely to reflect former glacial refuges. When the climate warmed at the dawn of the Holocene, the temperate zone shifted northwards but the sorb tree, bereft of efficient dispersal partners, could not follow, having to now make do with a Mediterranean climate that it is imperfectly adapted to. Likewise, oral tradition has it that the service tree was first deliberately introduced into Central Europe during Roman times (10). There is much to be said for a limited filling of the sorb tree’s potential range. Limited range-filling is prevalent in many European tree species (16), and the sorb tree is fruiting seamlessly as far north as Denmark, well outside its “natural” range (10). Populations of service trees have been identified on the rugged cliffs of southern Wales and Cornwall, small and stunted but apparently healthy (10). This demonstrates a considerable adaptability of the species and indicates the fundamental potential for much wider distribution.
Nowadays service trees, if not in orchards, grow mostly on sites extreme enough to eliminate more vigorous competition, particularly dry, south-facing slopes on shallow soils (17). As with other competition-sensitive species, it is generally understood that this represents a natural realised niche, as opposed to the wider potential niche. (Broadly speaking, a species’ potential niche is the hypothetical ecological niche that a species would occupy in the absence of competition. It is contrasted with the realized niche, which takes competition from other species, in this case most notably European beech (Fagus sylvatica), into account). However, if the sorb tree were indeed adapted to these extreme environs, it would be expected to fare best under these conditions. True xerophytes (dry-specialists), such as cacti, are so well adapted to dry environments that they simply cannot tolerate moisture. Yet, the sorb tree can and readily does grow on rich, limy soils in mesic (balanced supply of water) warm-temperate conditions, prime habitat for virtually all Central European tree species. These are the conditions in which it reaches its full physiological potential, resulting in an old age, large proportions and massive fruit crops. Well, if it can, could it be that it would have done so naturally in the past? Could it be that the sorb tree’s modern realised niche is nothing more than the last thread to which it clings onto, for lack of alternatives?
There is another characteristic of the tree that seems off kilter; while most species in the genus Sorbus readily engage in apomixis (the production of seeds without fertilisation), hybridisation and self-pollination, the sorb tree does not. It will not hybridise with any other species (18). In fact, this makes it something of an outlier within the Rosaceae at large, where hybridisation is often commonplace (19). Also, the sorb tree is extremely reluctant to self-pollinate. It is not dioicous (i.e. male and female flowers on separate individuals), so isolated individuals will inevitably produce autogamic (self-fertilized) seeds. These, however, exhibit dramatically reduced fitness (10). In principale, this can be explained by the sorb tree’s long potential lifespan. A main purpose of self-pollination is reproductive assurance. For this reason, annual plants will often commit to autogamy. However, with a longer lifespan, reproductive assurance becomes less of a problem (successful reproduction will happen, given time), making autogamy and its associated risks like inbreeding depression less desirable. Still, beggars can’t be choosers, and this applies to plants as well. Since the sorb tree is also pollinated by generalist insects, which can only cover so much distance, its strong preference for outcrossing (i.e. avoiding autogamy) makes it unlikely that the plant was originally rare (2). Unfortunately, pollen of insect-pollinated plants rarely ends up in places that aid preservation (such as peat bogs). This means that wind-pollinated plants are commonly over-represented in pollen diagrams, whereas insect-pollinated plants are either under-represented or missing entirely. Accordingly, direct extrapolations of former abundance on the basis of deposited pollen, a common means of inferring former plant occurrences, are not possible for the sorb tree. However, the species boasts a surprisingly high molecular diversity, (20) suggesting that it may have been more common in the past, and there is indeed indication that this may have been the case. Leaf imprints of the species are known from two sites in Italy and southern France dating to the Early Pleistocene, and at both sites, it constitutes a principal component of the woodland community (52)(53).
For all the above, it seems probable that the sorb tree once found optimal conditions in the wide wood-pastures, grasslands and open woodlands of Europe, criss-crossed by vast herds of herbivores that not only created the open conditions the tree requires for its growth, but also happily consumed its fruit, carrying the seeds far and wide on their seasonal migrations. After the herds had disappeared, displaced by humans and replaced by their livestock, the sorb tree nonetheless persisted. Humans favoured it for its fruit, which make for good jam, help preserve and lend a special flavour to cider and wine and were used to treat diarrhoea. The sorb tree also yields one of the heaviest types of wood in Europe, valuable for woodworking (10). Then, however, the industrial revolution swept across Europe. The fruit’s contents were no longer valued, coppice was largely abandoned in favour of modern forestry, resulting in denser, taller forests, and the livestock disappeared from pastures to pens. For the sorb tree, the combination of these processes led to catastrophic losses. Today, it is ecologically practically extinct.
Fig. 3 A service tree in autumn near Kronberg, Germany
Terms of use: This image is licensed under a Attribution 2.0 Generic. It is attributed to Heinz-Vale, and the original can be found here. The image is unedited.
Anachronistic Apples
To give more substance to the identification of a common pattern, let us now turn to another example; the European wild apple (Malus sylvestris), also known as the crab apple. A small tree of up to 10 m in height, it often stays smaller, developing into a multi-stemmed bush. Although it is not the primary ancestor of the cultivated apple (Malus domestica), it is nonetheless close enough to produce fertile hybrids – one of the many threats this species faces (21)(22). A 2013 survey by the German Ministry of Agriculture and Forestry recorded a mere 5500 individuals in the country, by and large in a detrimental state of preservation (23). Apart from its susceptibility to hybridisation with domesticated stocks, other problems include the demise of mature individuals in closing forests and insufficient progeny (22). Like the sorb tree, the crab apple is light demanding, but unlike it, the crab apple is also simply too small to persist in a competitive forest environment. The symptoms are all too familiar. What about the causes?
The situation is complicated by the fact that small-fruited cultivated apples, “feral” cultivates, hybrids between the cultivated and the wild, and even entirely different Malus species are commonly misidentified, labelled and sold as European wild apples (55). A factor that surely contributes is that knowledge of the pure wild apple and of how to identify it is broadly lacking. But how can it actually be distinguished? The bad news is that a reliable distinction between wild and hybrid individuals in particular is not always possible without genetic testing. That being said, there is a number of features that are strongly associated with the pure wild tree. When approaching it, the mature European wild apple has a distinctly dense crown structure and an overall dark appearance, which is an effect of both the crown structure and the dark bark. Indeed one may be forgiven to mistake it for a large common buckthorn (Rhamnus cathartica) or a hawthorn (Crataegus monogyna) even. While cultivated apples look like small trees with open, see-through crowns, wild apple trees appear like large shrubs, with dense, thorny crowns. The overall similarity with buckthorn in particular is upheld by the leaves, which in the wild apple are smallish and leathery, and mostly hairless on the downside. Overall, they do not have the “fluffy” feel of apple leaves. The aforementioned bark is darker and less scaly than in the cultivated apple, and cracks like the bark of a pear. The apples, finally, are small (<35 mm in diameter) and of a green or lightly yellow colour; Red fruit is strongly indicative of hybrids (56).
The crab apple bears small, mostly green apples which it produces in abundance in fall. From personal experience I can attest that these apples often litter the ground around the tree in winter, and eventually just rot. For the tree, this is surely undesired. Who, then, has been distributing seeds up to the present day? Contrary to the claims made in various publications, the crab apple’s primary intended disperser surely is not to be found in the bird guild. The fruit is too big, too dull and, again, drops to the ground once it ripens. If not birds, what about small mammals? Foxes (Vulpes vulpes), badgers (Meles meles) and beech martens (Martes martes) indeed do eat crab apples and disperse seeds. They are probably the primary reason for the limited dispersal sorb tree, wild pear and crab apple still enjoy (24). However, all of these mesopredators are territorial, making them inapt for long-distance dispersal. In addition, they are obviously overburdened with the task of eating fruit in bulk every autumn. Finally, even when dispersing seeds there is another task that badgers simply cannot shoulder - the creation of suitable habitat. The best bed of turd is of no worth to the seed when it is dropped under a closed canopy.
Once again, the most likely answer to the riddle is mammals, preferably large ones. Since the obliteration of elephants, rhinos and other megafauna from Europe, this leaves us with horses (Equus spp.), bovids (Bos taurus, Bos bonasus, Bubalus bubalis), cervids (Capreolus capreolus, Cervus elaphus, Dama dama), suids (Sus scroffa) and ursids (Ursus arctos). In this regard, it comes in very handy that the crab apple, uniquely, has been the subject of in-depth research on fruit consumption and dispersal not only by wild animals, but also domesticated livestock. A particularly thorough examination is provided by Buttenschøn & Buttenschøn (1998). Their findings are significant in two respects: firstly, they determined that in their study area cattle and horses accounted for 98% of seed dispersal. This was despite the presence of other potential dispersers (sheep (Ovis aries), roe deer, red deer, hare (Lepus europaeus)), which brings us to the second aspect of their study that is significant for our purpose; apart from cattle and horses, ungulates were poor at dispersal, but still superior to other animals, apart from the aforementioned mesopredators. The reason for this may be that deer and sheep process their food more thoroughly than cattle or horses do, as evidenced by their fine-grained turd. To the seeds, this is presumably detrimental. What this means for the crab apple in practice, however, is daunting; since cattle and horses are no longer ecologically present in modern Europe (they are either penned or fenced in, with few exceptions), the species is deprived of by far its two most eminent dispersers (25).
I have myself conducted trials that seem to assert this assumption. When I gave a local herd of zebu cattle, grazing in the name of landscape protection, a bag full of locally sourced crab apples, they hastily gobbled them up in no time. Although no proof for an interaction restored by me, I see this as a strong indication that crab apples are indeed intended to lure cattle or, more broadly, bovines. Crab apples are incredibly astringent for the human palate, but not to ungulates. Forefront fermenters in particular, (bovids, deer) need to be wary of acidosis, a condition that results from them eating foods rich in sugar. I suspect this to also be the reason why, in another feeding trial, the zebus would not eat the wild plums (Prunus cerasifera) I had offered them. At first this puzzled me, as I had not expected the indifference. However, against the backdrop of acidosis, the cattle probably acted wisely. What to my gusto were delicious plums was probably rather repelling to them. This clearly demonstrates that cattle are conscious of what they eat, which, in turn, allows a fruit tree to adapt to their dispersal. Since cattle seem so fond of crab apples, the species is most probably an anachronism that misses herds of large herbivores in the landscape, for the provision of both suitable half-open habitat and dispersal services.
Figure 4 A multi-stemmed crab apple tree. Note: pictured is a morphologically wild specimen. Due to the frequent hybridisation and the imperfect reliability of diagnostic features, it is not possible to reliably distinguish apples of wild and hybrid parentage without genetic testing
Terms of Use: The photograph was taken by the author
Although hard to attest from the sparse literature, the crab apple’s Western Palearctic kin are also probably suffering from mechanisms similar to those that make the European crab apple an exceedingly rare sight in modern Europe. These other species are Malus orientalis, Malus crescimannoi, Malus trilobata and Malus florentina. Malus orientalis, the Caucasian crab apple, has been genetically shown to have contributed to the domestic apple (Malus domestica), although less than the European crab apple or the Central Asian Malus sieversii (26). It is also one of the large-fruited Malus species, and generally similar to its European relative. Indeed, the apple genus contains both species with large fruit which drop, and with small, retaining fruit, intended for birds.
Malus crescimannoi is a relatively recently described species endemic to the mountains of northern Sicily. It too is doubtless close to the European wild apple, but is currently recognised as a distinct species. Although this status may change in the future, it is clearly distinct from other populations and is therefore evidence of the genetic diversity of the genus, which still exists in Europe but is in imminent danger of disappearing without suitable dispersal partners.
Malus florentina, the Florentine crab apple, and Malus trilobata, the Lebanese wild apple, are two enigmatic species with fragmented distributions. Malus florentina occurs mainly in Italy and the Balkans, especially in North Macedonia and Kosovo, (27). M. trilobata has only been identified in a few widely disjunct populations in the Levant, Anatolia and southeastern Thrace (28). To account for their uniqueness, both are placed in sections of their own as distantly related sister groups within the genus Malus, or perhaps even as genera in their own right (29). Although they are receiving growing attention as of late - M. trilobata having become an appreciated ornamental tree – knowledge of both species’ ecological characteristics is still lacking. The fruits of M. trilobata are reported to reach sizes of up to 2.5 cm and weights of more than 9g, (30) which would render it well on the upper limit of what native frugivorous birds can swallow. Fittingly, the fruit of both species apparently gets dropped before ripening, (31)(32) which is indicative of mammal dispersal. For M. trilobata, also known as the deer apple, it is reported that fruit fall is in October and November, and that the harvest remains on the ground for months, but is eaten by herbivores such as deer and horses (28). Both species’ fruit is reportedly good in taste, and is used to some extent in rural cuisine (31). This fits with the fact that both species bear fruit that is inconspicuous in colour and undergoes a process known as bletting, where it changes texture, becoming mushy and aromatic. This is a feature it shares with sorb apples (Cormus domestica) and other mammalian-adapted fruits, as we shall see. This could mean that bletting is indicative of mammal dispersal, too, turning the fruit from repellent, (protection against seed predators) to palatable to the intended disperser. Mammalian dispersal, or lack of it, may also explain why both species occur over relatively extensive geographic areas, but consistently in small, isolated and widely disjunct populations. Needless to say, both species are in serious peril, suffering, apart from dispersal troubles, from reproductive isolation (33) and forest canopy closure (34).
Figure 5 A Merck’s Rhinoceros (Stephanorhinus kirchbergensis) forages underneath an old crab apple, alongside a roe deer (Capreolus capreolus), a fieldfare (Turdus pilaris) and two redwings (Turdus iliacus, next to the tree’s trunk). Later, the rhino will excrete the seeds, securing the crab apple’s posterity. Similar interactions would have probably been a regular sight in Europe’s not too distant past.
Terms of use: Artwork by Hjalte Kjaerby and used with the kind permission of the artist
Pears in Peril
In many respects pears (genus Pyrus) are very similar to apples. However, while the genus Malus is found throughout the temperate zones of the Northern Hemisphere, Pyrus is (originally) unique to the Old World. The genus has two centres of diversity: East Asia and the Western Palearctic. Central Europe may for the most part host only one species, Pyrus pyraster, but the genus as a whole is very species-rich in a region stretching from Iberia in the west to Iran in the east (35). Like the genus Malus, Pyrus includes both large-fruited species (such as P. pyraster and P. spinosa) and small-fruited species. Again, as with genus Malus, the small-fruited species are bird-dispersed, as evidenced by the rapid invasion of Pyrus calleryana into North America, often with the assistance of the equally novel European starling (Sturnus vulgaris) (36). At the same time, many pear species suffer from insufficient dispersal in their native habitat. As with the crab apple, the population of European wild pear (Pyrus pyraster) in Germany has been the subject of meticulous investigation by the German Ministry of Agriculture and Forestry. Although not quite as desolate as the crab apple, the count still identified a total of just 14.000 individuals. While largely vital, two thirds of stands were lacking in regeneration (23). Pyrus bourgeana, a species from Iberia, occurs at low densities and has a fragmented distribution. Where it does occur, stands are strikingly aggregated, the most likely reason for this being limitations in dispersal (37). In Doñana National Park, the species is mostly dispersed by wild boars, red foxes and European badgers (38). For this reason, Fedriani et al. consider badgers, foxes and wild boar to be legitimate dispersers of the tree, despite the fact that their insufficient roaming distance seems to be at the heart of the pear’s awkward stand structure (37). While short-distance dispersal can help a population survive mid-term, it will, in the long run, bring about the accumulation of inimical effects. As long as long-distance dispersal occurs every now and then, a population may be able to offset the effects of inbreeding. When it is lacking, however, the deleterious effects of inbreeding in fragmented aggregations of closely related individuals can in the long run cause population breakdown through inbreeding depression.
While dispersal and population structure in other species of pear is less well documented, extrapolation is probably appropriate where the fruit is aiming for mammalian dispersal. Given that the Western Palearctic constitutes a major centre of global diversity for the genus, the region also holds a substantial conservation responsibility. It is home to a number of highly endemic species such as P. anatolica, P. hakkiarica and P. oxyprion, all from Anatolia alone. Another species, Pyrus gergerana (a large-fruited species), is restricted to the immediate environs of the namesake village Herher, Armenia. Despite close scrutiny, the species’ total population amounts to no more than 49 individuals on record (39). It is one of several geographically restricted species in Armenia, all facing uncertain future prospects (40). At the same time, agricultural globalisation tends to promote the penning of livestock away from ancestral grazing grounds, often resulting in the virtual disappearance of livestock from the landscape. This process is currently making itself felt in many regions of the world, the country of Romania among them. Here, it often leads to the abandonment of ancient common pasture grounds (41). Assuming that the findings Buttenschøn & Buttenschøn produced for the European crab apple also hold true for mammal-dispersed wild pears, the outlook for the unique and diverse assortment of pears of the Western Palearctic is bleak.
Figure 6 From left to right: 3 Sorb apples (C. domestica), 2 wild pears (P. pyraster), 2 crab apples (M. sylvestris), some of them already bletting. Despite no closer relationship other than being Maleae (pomes), the fruits show striking overlap in shape, dimensions, texture and smell. More strikingly, being situated on disparate branches of the Maleae tribe, European crab apples and sorb apples are more similar to each other than they are to some of their close, bird-dispersed kin. This suggests convergent evolution, driven by similar or the same dispersal partners. Note also the considerable variation in size in sorb apples.
Terms of Use: The photograph was taken by the author
A Haw Peculiar
Like the sorb tree, the medlar (Mespilus germanica) represents a unique evolutionary lineage that departed from an otherwise mostly bird-dispersed group, in this case the hawthorns (genus Crataegus), to try its luck with mammals as dispersers instead (42). (Crataegus mexicana, the tejocote, is another of the few mammal-dispersed hawthorns). Evolution then did its work, and the result is strikingly alike, at least in function. Like the sorb tree, the medlar bears fruit much larger than those of its closest relatives. Moreover, these fruits adopt an inconspicuous brown colour as they ripen, blet, smell intensely and get dropped to the ground. By now it should be clear where this is going. The medlar’s native (historical) range is obscure, and usually identified with western Asia and the southern Balkans (43). However, the medlar has been in cultivation for such a long time that the origins of the species are unclear. This is problematic because it means that finding information about the wild stock is seriously impeded. Like many of the species discussed so far, the medlar exhibits a wide ecological amplitude and is tolerant of a variety of soils and climates. Unlike the cultivated forms, the wild medlar is reported to be thorny, however, it is also reported to be rare (43). As a shrub or small tree uncompetitive, light-demanding and relying on large herbivores for dispersal, modern herd-deprived Europe and western Asia make for a tough place to live in. At the same time, interglacial leaf imprints have been reported from Burgtonna, Thuringia, (44) which may tell of a time when the medlar proliferated throughout Europe, benefitting from the omnipresence of large herds of herbivores.
Intriguingly, there is another haw from the Mediterranean that may well be on its way to becoming an obligate mammalian disperser; the azarole (Crataegus azarolus). Its English synonym name, the Mediterranean medlar, refers to the fruit, which are usually around 2 cm in diameter (45), but can take on larger proportions in cultivation. Revealing an underlying potential that breaks through if selected for, this could hypothetically mean that the species used to be more reliant on mammals in the past, but reverted to avian dispersal in the face of a paucity of mammalian candidates. Either way, the fact that the fruit has a place in Mediterranean cuisine is a good pointer for at least partial mammalian dispersal, for which there is also evidence in the scientific literature (46).
Figure 7. Bletted (centre) and ripe but still unpalatable medlars (left and right)
Terms of use: This image is licensed under a Attribution 2.0 Generic. It is attributed to Istvan Takacs, and the original can be found here. The image is unedited.
A Hyrcanian Enigma
Like the medlar, the quince (Cydonia oblonga) is an enigmatic, small fruit tree whose native range is obscure, probably located somewhere in western Asia. The perhaps most probable candidate is Hyrcania, the stretch of land in modern Iran north of the Alburz Mountains. Hyrcania is a special region, with a unique flora as its main claim to fame. It is here where Parrotia persica, the Persian ironwood, has found its last refuge and where the tallest known oak tree in the world, a Quercus castaneifolia at a towering height of 60.4 m, has been found (47). In many ways, the quince is very similar to the pomes discussed above; it is a small tree or multi-stemmed bush that grows up to 8 m tall, light-demanding and producing fruit too large to be consumed effectively (meaning swallowed whole) by birds. However, quinces are exceptionally large. Even though the wild quince bears smaller fruit than its many cultivars, (48) the fact that quince cultivars have been bred to produce fruits of exceptional size in a relatively small time frame may indicate that the potential was already there, lying dormant since the extinction of animals large enough to eat fruit of this size. Additionally, the quince’s restricted native range could be indicative of a lack of dispersal. Incidentally, the quince occurs as an unusual invasive on Baozhong Mountain, China. Here, it is probably dispersed by macaques (Macaca spp.) and wild boar (49).
Restoring Ties
To date, I have also carried out similar feeding trials as for the crab apple for wild pears and sorb apples. In each of these, the fruit was received with fondness by the cattle (mainly zebu, Galloway and Highland cattle), and consumed in large quantities in a matter of minutes. Striking in any case was the eagerness the cattle expressed. When I was feeding crab apples for the first time, the cattle came running to eat them. Particularly powerful, however, was a trial I conducted another time, with a different herd of zebu cattle, this time with sorb apples as bait. When I arrived, the herd was a short distance away from me, and was being fed by someone else. I do not know what it was that the other person was feeding. All I know is that as soon as my bag of sorb apples was open, the other person was immediately ignored. Without me calling the cattle, making a gesture or throwing a fruit, the cattle came running towards me, seemingly in eager anticipation. I can only assume that they must have smelled the ripe sorb apples from afar. Each of the fruit I threw was diligently sought in the grass or eaten right straight out of my hand. When the bag was empty, the cattle became indifferent to my presence and started to walk back to where they had come from earlier. I had not visited this herd before in this year, but I cannot rule out the possibility that they remembered me from feedings in previous years. In any case, even if the cattle had associated me with feeding, their fondness of sorb apples seemed sincere. To me, this serves as a strong indication that sorb apples are designed to attract bovids, and most likely not just them.
Figure 8. Video of feeding of zebu cattle with Sorb apples
Conclusion
Among the many quirks the presented fruits sport, there are some common to all or most of them. Some of these we have discussed already; the fruits are rather large, furnish distinct smells, drop to the ground, tend to be dull or brown in colour, blet, are astringent when unripe and not too sweet when ripe. But there is one more characteristic that we have not paid due attention to so far: late fruiting. Sorb apples, crab apples, wild pears, medlars and quinces all shed their fruit in late fall. Typically, fruiting will start in September and end in November for these species. This is by no means representative, neither for pomes nor Rosaceae at large. Rowan berries, hawthorn haws and serviceberries (Amelanchier), for example, all ripen in summer, although they may remain on the branches for much longer. So why the late fruiting? As we have also established already, the mammal-dispersed pomes of the Western Palearctic do not cluster together within the Maleae. Instead, most of them count bird-dispersed species with widely different properties among their closest allies. Among them such species that fruit in summer. Fruit size alone does not suffice as an explanation either, although it seems cogent at first glance. Large fruit will naturally take longer to ripen, but plums (Prunus subg. Prunus) demonstrate that getting large fruit ready in summer is no principal contradiction. All it takes is a somewhat earlier flowering season. Since neither taxonomy nor fruit biology seem to hold clues for this ecological conundrum, perhaps evolutionary ecology will. Suppose that dispersal by one particular type of dispersal partner selects for certain characteristics, then species sharing this dispersal partner will, inevitably, turn out to be quite similar with respect to the selected feature. Additionally, if the intended dispersal partners are large mammals of the temperate hemisphere, these mammals will have to cope with what follows fall; winter. The long weeks of biting frost and scarce resources put a heavy strain on herbivores in particular. It is precisely during the weeks of preparation for this time of privation, that a nutrient surplus in the form of fruit will be most eagerly received. This alone would constitute a compelling argument for any endozoochorous mammal-dispersed species of northern latitudes to adopt a late fruiting season. But there may be another argument more inciting even. In regions with pronounced seasonal variation in climate, large herbivores commonly respond with mass migration. Some populations of blue wildebeest (Connochaetes taurinus) in east Africa constantly move in wide arcs on the heels of rain clouds. Barren-ground caribou (Rangifer arcticus arcticus) (50) in particular undertake seasonal migrations of epic proportions to evade adverse conditions. In a similar vein, it has been suggested by others that transhumance practices – seasonal livestock migration – may be the effigy of age-old natural migrations turned cultural practice (51). If true, this would suggest that wild herbivore herds in the Western Palearctic also used to conduct seasonal migrations, from summer grounds to winter quarters and back. Bletting – the process that makes medlars, sorbs and wild pears palatable – is either initiated or aided by frost. A year’s first cold spell may have, in turn, been the signal that triggered herbivore migrations. For a fruit tree, timing fruit fall to coincide with the onset of migration would be the ultimate prize, ensuring that its offspring are transported en masse to germinate in far-flung places recently cleared by the herds. What else is there to wish for?
In this first part, we have introduced the concept of evolutionary anachronisms, briefly touching on its history and famous examples, to then dive deeper into what the Western Palearctic has to offer. So far, we have only discussed pomes, the “apple fruits”. In the next chapter, we will broaden the taxonomic scope of this investigation to look at some other examples. Some of them are substantial, some more tentative anachronisms. But all are intriguing, in one way or another.
Sources
1. Carlos Neto de Carvalho, Silvério Figueiredo, Fernando Muniz, João Belo, Pedro P. Cunha, Andrea Baucon, Luis M. Cáceres & Joaquín Rodriguez-Vidal (2020) Tracking the last elephants in Europe during the Würm Pleniglacial: the importance of the Late Pleistocene aeolianite record in SW Iberia, Ichnos, 27:3, 352-360
2. Zaya, D.N., Howe, H.F. The anomalous Kentucky coffeetree: megafaunal fruit sinking to extinction?. Oecologia 161, 221–226 (2009).
3. Richard Dawkins, 1998
4. Byers, John A. (1997). American Pronghorn: Social Adaptations and the Ghosts of Predators Past. Chicago University Press. ISBN 978-0-226-08699-6.
5. BAKKER, E.S., OLFF, H., VANDENBERGHE, C., DE MAEYER, K., SMIT, R., GLEICHMAN, J.M. and VERA, F.W.M. (2004), Ecological anachronisms in the recruitment of temperate light-demanding tree species in wooded pastures. Journal of Applied Ecology, 41: 571-582.
6. Bråthen, K., Pugnaire, F. I., & Bardgett, R. (Accepted/In press). The paradox of forbs in grasslands and their legacy of the Mammoth steppe. Frontiers in Ecology and the Environment.
7. Adie, H. and Lawes, M.J. (2023), Solutions to fire and shade: resprouting, growing tall and the origin of Eurasian temperate broadleaved forest. Biol Rev, 98: 643-661.
8. Kremer, A. and Hipp, A.L. (2020), Oaks: an evolutionary success story. New Phytol, 226: 987-1011.
9. Barlow, Connie (2001). The Ghosts of Evolution: Nonsensical Fruit, Missing Partners, and Other Ecological Anachronisms. New York: Basic Books. ISBN 978-0-465-00551-2.
10. Hrdoušek V. (ed), Špiešek Z., Krška B., Bakay L. (2023): The Service Tree - The Tree for a New Europe, Petr Brázda – vydavatelsví
11. Jiahui Sun, Shuo Shi, Jinlu Li, Jing Yu, Ling Wang, Xueying Yang, Ling Guo, Shiliang Zhou, "Phylogeny of Maleae (Rosaceae) Based on Multiple Chloroplast Regions: Implications to Genera Circumscription", BioMed Research International, vol. 2018
12. Janson, Charles H. “Adaptation of Fruit Morphology to Dispersal Agents in a Neotropical Forest.” Science, vol. 219, no. 4581, 1983, pp. 187–89. JSTOR
13. Spengler, R.N., Kienast, F., Roberts, P. et al. Bearing Fruit: Miocene Apes and Rosaceous Fruit Evolution. Biol Theory 18, 134–151 (2023).
14. Wedig Kausch-Blecken von Schmeling: Der Speierling. 2. Auflage. Verlag Kausch, Bovenden 2000
15. Rotach P. 2003. EUFORGEN Technical Guidelines for genetic conservation and use for service tree (Sorbus domestica). International Plant Genetic Resources Institute, Rome, Italy.
16. Svenning, J.-C. and Skov, F. (2004), Limited filling of the potential range in European tree species. Ecology Letters, 7: 565-573.
17. Paganová V.: Ecology and distribution of service tree Sorbus domestica (L.) in Slovakia. Ekológia (Bratislava), Vol. 27, No. 2, p. 152–167, 2008.
18. E. Nelson-Jones, Derek Briggs and A. Smith (2002): The origin of intermediate species of the genus Sorbus, Theoretical and Applied Genetics, year 2002, volume 105, pages 953-963
19. Zhang SD, Jin JJ, Chen SY, Chase MW, Soltis DE, Li HT, Yang JB, Li DZ, Yi TS. Diversification of Rosaceae since the Late Cretaceous based on plastid phylogenomics. New Phytol. 2017 May;214(3):1355-1367. doi: 10.1111/nph.14461. Epub 2017 Feb 10. PMID: 28186635.
20. George, Jan-Peter, et al. “High Molecular Diversity in the True Service Tree (Sorbus Domestica) despite Rareness: Data from Europe with Special Reference to the Austrian Occurrence.” Annals of Botany, vol. 115, no. 7, 2015, pp. 1105–15. JSTOR
21. Schnitzler A, Arnold C, Cornille A, Bachmann O, Schnitzler C (2014) Wild European Apple (Malus sylvestris (L.) Mill.) Population Dynamics: Insight from Genetics and Ecology in the Rhine Valley. Priorities for a Future Conservation Programme. PLoS ONE 9(5): e96596.
22. Wagner, Iris, Maurer, W.D., Lemmen, P., Schmitt, H.P., Wagner, M., Binder, M. and Patzak, P.. "Hybridization and Genetic Diversity in Wild Apple (Malus sylvestris (L.) MILL.) from Various Regions in Germany and from Luxembourg" Silvae Genetica, vol.63, no.1-6, 2014, pp.81-93.
23. Erfassung und Dokumentation genetischer Ressourcen seltener und gefährdeter Baumarten in Deutschland“, Teillos 2: Wild-Apfel (Malus sylvestris) und Wild-Birne (Pyrus pyraster), Bundesanstalt für Landwirtschaft und Ernährung (BLE), 2013
24. Herrera, Carlos M. “Frugivory and Seed Dispersal by Carnivorous Mammals, and Associated Fruit Characteristics, in Undisturbed Mediterranean Habitats.” Oikos, vol. 55, no. 2, 1989, pp. 250–62. JSTOR
25. Buttenschøn, Rita Merete, and Jon Buttenschøn. “Population Dynamics of Malus Sylvestris Stands in Grazed and Ungrazed, Semi-Natural Grasslands and Fragmented Woodlands in Mols Bjerge, Denmark.” Annales Botanici Fennici, vol. 35, no. 4, 1998, pp. 233–46. JSTOR, http://www.jstor.org/stable/23726629.
26. Cornille A, Gladieux P, Smulders MJM, Roldán-Ruiz I, Laurens F, Le Cam B, et al. (2012) New Insight into the History of Domesticated Apple: Secondary Contribution of the European Wild Apple to the Genome of Cultivated Varieties. PLoS Genet 8(5): e1002703.
27. Gordana Tomovic, Marjan NiketiNiketic and Vladimir Stevanovi, 2003},Stevanovic, “Malosorbus florentina (Rosaceae-Maloideae) - Distribu- tionDistribution, Synecology and Threatened Status in Serbia}, , 2003,
28. Giorgos, Korakis., Kostas, Poirazidis., Nikolaos, Papamattheakis., Aristotelis, C., Papageorgiou. (2006). New localities of the vulnerable species Eriolobus trilobatus (Rosaceae) in northeastern Greece.
29. GUAN-ZE QIAN and others, Taxonomic study of Malus section Florentinae (Rosaceae), Botanical Journal of the Linnean Society, Volume 158, Issue 2, October 2008, Pages 223–227
30. Tashev, Alexander, Petkova, Krasimira, “Plant structure, function and development Fruit and seed morphological peculiarities of the critically threatened Eriolobus trilobatus (Rosaceae)” 2020/04/19
31. DEMİRCAN, H., SARIOĞLU, K., SAĞDIÇ, O., ÖZKAN, K., KAYACAN, S., US, A. A., & ORAL, R. A.. (2022). Deer apple (Malus trilobata) fruit grown in the Mediterranean region: identification of some components and pomological features. Food Science and Technology, 42, e116421.
32. Sutton, J. & Dunn, N. (2021), 'Malus florentina' from the website Trees and Shrubs Online (treesandshrubsonline.org/articles/malus/malus-florentina/).
33. Tashev, Alexander, Petkova, Krasimira, “Plant structure, function and development Fruit and seed morphological peculiarities of the critically threatened Eriolobus trilobatus (Rosaceae)” 2020/04/19
34.Maccaronia, Luca, “Study of the population of Malus florentina and of the main ecological drivers responsible for its structure within the natural reserve of the Abbadia di Fiastra (MC)”, 2018/19
35. Rubtsov, G. A. “Geographical Distribution of the Genus Pyrus and Trends and Factors in Its Evolution.” The American Naturalist, vol. 78, no. 777, 1944, pp. 358–66. JSTOR
36. Clark, Olivia, "A Starling in a Pear Tree: Assessing the Influence of Bird Dispersal on Callery Pear (Pyrus calleryana)" (2022). Honors Theses. 347.
37. Jose M. Fedriani & Miguel Delibes (2009) Seed dispersal in the Iberian pear, Pyrus bourgaeana: A role for infrequent mutualists, Écoscience, 16:3, 311-321,
38. Fedriani, Jose M., et al. “Spatial Pattern of Adult Trees and the Mammal-Generated Seed Rain in the Iberian Pear.” Ecography, vol. 33, no. 3, 2010, pp. 545–55. JSTOR
39. Fayvush, G. 2014. Pyrus gergerana. The IUCN Red List of Threatened Species 2014: e.T200400A2656627.
40. Asatryan, Anna (2019) Wild Pears of Armenia : Diversity, endemics and conservation. Acta Biologica Plantarum Agriensis. 7. pp. 19-31. ISSN 2061-6716 (Print), 2063-6725 (Online)
41. Plieninger, Tobias; Hartel, Tibor (eds.). European Wood-pastures in Transition. Routledge.
42. Phipps, James B. "Studies in Mespilus, Crataegus, and× Crataemespilus (Rosaceae), I. Differentiation of Mespilus and Crataegus, expansion of× Crataemespilus, with supplementary observations on differences between the Crataegus and Amelanchier clades." Phytotaxa 257.3 (2016): 201-229.
43. Baird, J.R., Thieret, J.W. The medlar (Mespilus germanica, Rosaceae) from antiquity to obscurity. Econ Bot 43, 328–372 (1989).
44. Vent, W. 1977. Neue Ergebnisse über die Flora des Travertins von Burgtonna. Gleditschia 5:53-57.
45. 'Crataegus azarolus' from the website Trees and Shrubs Online (treesandshrubsonline.org/articles/crataegus/crataegus-azarolus/).
46. Khiari, Salma, Mohamed Boussaid, and Chokri Messaoud. "Genetic diversity and population structure in natural populations of Tunisian Azarole (Crataegus azarolus L. var. aronia L.) assessed by microsatellite markers." Biochemical Systematics and Ecology 59 (2015): 264-270.
47. Pieter De Frenne on X, then Twitter https://twitter.com/PieterDeFrenne/status/1388213327856377856
48. Abdollahi, H. A review on history, domestication and germplasm collections of quince (Cydonia oblonga Mill.) in the world. Genet Resour Crop Evol 66, 1041–1058 (2019).
49. Xie, Y., Li, J., Zhao, L., Liu, W., Gong, Q., Deng, M., Zhao, M., & Huang, S. (2023). Naturalization of an alien ancient fruit tree at a fine scale: Community structure and population dynamics of Cydonia oblonga in China. Ecology and Evolution, 13, e9703.
50. Harding LE. Available names for Rangifer (Mammalia, Artiodactyla, Cervidae) species and subspecies. Zookeys. 2022 Aug 26;1119:117-151. doi: 10.3897/zookeys.1119.80233. Erratum in: Zookeys. 2022 Sep 28;1122:173-174. PMID: 36762356; PMCID: PMC9848878.
51. Transhumance: a link to natural migration? (May 16, 2012), Rewilding Europe https://rewildingeurope.com/blog/transhumance-a-link-to-natural-migration/
52. Adroit, B. et al. (2016). Bernasso, a paleoforest from the early Pleistocene: New input from plant–insect interactions (Hérault, France). Palaeogeography, Palaeoclimatology, Palaeoecology, 446, 78-84
53. Denk, T., Sami, M., Teodoridis, V., Martinetto, E. (2022): The late Early Pleistocene flora of Oriolo, Faenza (Italy): assembly of the modern forest biome. – Fossil Imprint, 78(1): 217–262, Praha.
54. Ruhsam, Markus; Renny, James; Worrel, Rick (2022): Verification of wild apple (Malus sylvestris) nursery stock sold in the United Kingdom. Plants, People, Planet, 5, 203-208
55. Worrel, Rick; Ruhsam, Markus; Renny, James; Jessop, Will; Findlay, Graeme (2018): "The Ecology and Genetics of Scotland's native wild apple: Malus sylvestris" (Report)