One of the defining characteristics of Japan is that it consists of several smaller archipelagos within a larger archipelago. The 1,200-km-long chain of islands lying between Kyūshū and Taiwan is one of those lesser archipelagos, known as the Nansei Shotō and sometimes called the Ryūkyū Archipelago, comprising nearly 200 subtropical islands.
Japan’s straggling southern archipelago is located where the Philippine Sea Plate descends beneath the Eurasian Plate. The islands were created 15 million years or so ago, as a consequence of the formation of the Okinawa Trough. Over the last two million years, changing sea levels as a consequence of climatic variation, and the deposition of Ryūkyū limestone, have built up the island chain, along with the formation of the islands’ immense coral reefs. Repeated separation from and reattachment to the Eurasian continent, and to one another, have resulted in two very different types of island: ‘high islands’ composed of pre-Tertiary rocks with hills rising several hundred metres above sea level, such as Mt Yuwan (694 m) on Amami Ōshima, and rivers; and ‘low islands’, formed of Quaternary Era1 exposed coral-reef limestone, with plentiful groundwater.
Some 15 Mya, the island chain formed the eastern rim of the Eurasian continent, as a consequence of which the islands hold a similar composition of terrestrial species. As the islands became isolated from the continent by oceanic channels, the biota of those islands also became isolated and, over time, diverged from their source populations as a result of differential selection pressures and random genetic drift. While many of the species surviving today in the Nansei Shotō are relicts of those ancient times of connection to the Eurasian continent, the long isolation resulting from the barrier of the Kuroshio Current, separating the island chain from the continent and from other islands, has allowed a unique biota to evolve independently on these islands. As a consequence, Japan’s southern archipelago is rich in flora and fauna, and offers a clear example of the effects of geographical isolation resulting in genetic change. The islands are often described as a living museum, displaying the same natural phenomena as shown in the Galápagos.
Periodic connection and isolation have set the stage here that has resulted in the evolution of endemic species of plant, insect, bird and even mammal, and so created a biodiversity hotspot in these islands. By way of illustration, the warm-temperate forests of southwestern Japan may support 50 to 70 species of tree per hectare, whereas a typical subtropical forest in the Nansei Shotō supports about 100 species. Although these subtropical forests may at first seem superficially similar to the evergreen broadleaf forests of the warm temperate zone, their tree-species diversity is actually much higher and their dominant canopy species are evergreen broadleaf species, such as Itaji Chinkapin, a member of the beech family, and Needlewood, a member of the tea family.
Over geological times these islands have been linked to one another, and at other times they have been disconnected from each other and from the main island of Kyūshū to the north and Taiwan to the south and have evolved their own unique, endemic species.
The Nansei Shotō archipelago is a string of biological ‘pearls’ so rich in species and so fascinating in their biogeographical differences that they warrant being far better known. Here, in long isolation, live a number of species endemic to these small islands. These coral-fringed islands support subtropical forests, with tree ferns and cycads, where one can search for the extraordinary and highly localized, such as Iriomote Cat2.
In this archipelago, two regions in particular stand out as biodiversity hotspots. The first, the northern Nansei Shotō, which encompasses the islands of Tanega-shima, Yaku-shima, Amami Ōshima and Tokuno-shima and their associated islands, is the subject of this chapter. The second region, the southern Nansei Shotō, including the islands of Okinawa, Miyako, Ishigaki, and Iriomote and associated islands, is dealt with in the next chapter. The islands of the northern Nansei Shotō combine elements of Japan’s main islands, including unique races of Japanese Macaque and Japanese Deer for example on Yaku-shima, with unique single-island endemics such as Owston’s Woodpecker and Lidth’s Jay on Amami Ōshima.
The forests of the Nansei Shotō south of Yaku-shima are subtropical in character. They are dominated by Japanese Chinquapin, Okinawan Oak and Japanese Bay Tree.
The fauna and flora of the whole region are distinctive, differing dramatically from the main Japanese archipelago to the north, and characterized by high levels of endemism as a consequence of prolonged insularization. Some species occur throughout the island chain and no farther north, although they are not endemic. For example, the charismatic dragonfly known as Variegated Flutterer ranges from South Asia (e.g. Sri Lanka) to the Nansei Shotō, while Orange-tailed Sprite ranges from Southeast Asia to Kyūshū. Other species occur throughout the island chain, sharing their range with the more widespread flutterer and sprite, but are endemic to the Nansei Shotō. Examples include some that are restricted to islands of the northern Nansei Shotō or to the islands of the southern Nansei Shotō, such as Amami Damselfly which is endemic just to the Amami Islands; Japanese Matrona and Ryukyu Damselfly, both of which are endemic to the Amami Islands and Okinawa Islands; and Yaeyama Gossamer-wing Damselfly and the Yaeyama Clubtail, both of which are endemic only to the Yaeyama Islands.
The situation of these islands, straddling the faunal boundaries between the Palaearctic and the Oriental Regions, could hardly be more provocative. Put an entomologist, a botanist, a herpetologist and an ornithologist together in these islands and they will hotly debate where the regional boundary should be drawn. Biogeographical maps of this region should be annotated with ‘here lies confusion’. That such debate takes place is witness to this region’s extraordinary natural history.
Whereas the Galápagos and Hawaiian Islands comprise collections of volcanoes which emerged from the ocean floor and have been isolated since birth, the Ryūkyū Islands are entirely different. All living things surviving on the former, namely Galápagos and Hawaii, isolated volcanic islands are descended from life forms that flew, rafted or were blown there. Additional factors, however, come into play in the Ryūkyū Islands, which did not arise uninhabited from the ocean floor. Until about 1.7 Mya, the Ryūkyū chain was part of an already inhabited great peninsula extending eastwards from the Asian continent.
This island arc has three unequal parts (northern, central and southern). As sea levels have fallen and risen, the three segments have sometimes been joined as one unit, sometimes isolated as three. At times of lowest sea levels, they have been partly reconnected to the north, to mainland Japan, and at times to Taiwan to the south, allowing the spread of plants and animals from those areas. At times of higher sea levels, what were once hills on larger islands became many separate islands. Confusing? Yes, but just imagine what that confused geological history means in terms of biogeography.
Living on these few small islands today are wide-ranging Southeast Asian species that just reach their northern limits here, such as Purple Heron and Cinnamon Bittern. Other species reach their southernmost limit here, such as Japanese Macaques on the island of Yaku-shima. Of particular interest are the endemic species. Some species are endemic to the whole island chain, such as Ryukyu Green Pigeon, while others are endemic to just one major sector of the island arc and occur nowhere else. Finally, some species are restricted to a single island, such as Pryer’s Woodpecker and Okinawa Robin. That amounts not just to an enormous biodiversity in a small area, but to some fascinating foci of isolation and evolution. Isolation on an island from other populations of the same species over a long period of time results in genetic change, so that the isolated population becomes distinctive from the source population. The isolated population may be regarded as a race of the original species, a new subspecies, or even a new full species. Such new species on these islands are by definition endemic: they occur nowhere else.
Birds, bats and flying insects colonize islands rather easily; their flight enables them to disperse to the most isolated of islands. Reptiles may take longer to spread to new areas but, because of their ability to go for long periods without food or water, they can survive to reach even remote islands, such as the Galápagos. Amphibians on the other hand, having permeable skin, have little hope of surviving a sea passage. Large mammals, too, display a poor track record of long-distance dispersal across seawater. The ancient connections between the Ryūkyū Islands and continental East Asia, however, have enabled certain amphibians and mammals, such as the ancestors of our Amami Rabbit, to disperse through the region. During their subsequent long isolation, they have evolved into the distinct forms which we see today.
Because of its 20th-century history, Okinawa is the best-known of the Nansei Shotō. Far less well-known are Yaku-shima and Amami Ōshima. The latter is a gem of an island that lies to the north of Okinawa and which, like Okinawa, is a major centre of biological endemism. Here, in lush subtropical forests dotted with tree ferns and sago palms, occur many of the Ryūkyū regional endemics such as Sword-tailed Newt, Anderson’s Crocodile Newt [VU], Narrow-mouthed Toad, Ryukyu Kajika Frog and Hallowell’s Tree Frog, as well as species more closely restricted to the Amami group of islands. For example, these islands are home to several ancient endemic amphibia, such as Amami Green Tree Frog, Amami Brown Frog, Amami Tip-nosed Frog [VU], Amami Ishikawa’s Frog [EN], and the enormous Otton Frog, the last reaching 14 cm in length. Birds found nowhere else in the world except on this small island group include the extremely attractive Lidth’s Jay, a relative of the crows, resplendent in dazzling cobalt and chestnut plumage, and the elusive, more sombre-clad Amami Thrush. Other near-endemic species, i.e. those confined to Amami Ōshima, Okinawa and the islands and islets in between, include Amami Woodcock [EN], Pryer’s Keelback Snake (sometimes called the Ryukyu Water Snake), and Ryukyu Brown Frog.
Geology, geography and mammalogy move hand in hand here in the Ryūkyū Islands, with clear demarcations subdividing the archipelago into northern, central and southern parts. The mammalian fauna of the northern sector shares affinities with mainland Japan, while that of the southern sector shares affinities with Taiwan and East Asia. In contrast, the mammals of the central islands are highly distinct. Birdwise, we find the same pattern, with no (or very few) locally endemic species in the northern or southernmost island groups, and just the regional endemics occurring throughout; whereas in the central islands there are a number of endemic species, such as Okinawa Rail, Lidth’s Jay, Pryer’s Woodpecker and Owston’s Woodpecker, and Amami Thrush. So, here on subtropical Amami Ōshima, surrounded by coral reefs and situated in the Ryūkyū Islands 400 km southwest of Kyūshū, lives an extraordinary collection of endemic species, the largest of which is a unique rabbit.
Connection and isolation have played an important role in the Nansei Shotō, such that the Kerama and Tokara Straits effectively divide the archipelago into three parts. In times past, the northern section retained a connection with Kyūshū, whereas the southern section retained a connection via Taiwan to the continent. At other times, all islands have been connected, and at still other times of high sea levels all have been isolated.
Not all of the islands share the same species, because each has a unique history. Few of Japan’s familiar mammals occur here. There are no squirrels, no foxes, and no Tanuki. Instead, the largest mammals are rather small subspecies of Japanese Deer and Wild Boar. There are, however, other mammals on the islands, and among them are several endemic rodents, such as Amami Spiny Rat, Okinawa Spiny Rat and Ryukyu Tree Rat or Ryukyu Long-haired Rat, and the extraordinary forest-dwelling Amami Rabbit. The largest predator of the region, Iriomote Cat, no larger than a Domestic Cat, was discovered only as recently as 1965 and lives on just one small island, Iriomote-jima.
Some animals of the northern and southern islands of the Nansei Shotō remain genetically similar to their source populations on the main islands of Japan or of southern Asia. Some others living on the islands of Okinawa and Amami Ōshima tell a different story. Genetic analysis has shown that Large Japanese Field Mouse belongs to a group of species that is widespread in the temperate zone. It occurs on all four of Japan’s main islands and as far south as the northern Nansei Shotō, but it is entirely absent from the central and southern islands of that chain. Although those Japanese field mice on the various islands of the northern Nansei Shotō have been isolated for some time from those in Kyūshū, they still share many similarities with each other, and are not very distinct from those of other areas. At the other end of the Nansei Shotō, in the extreme south, lives Iriomote Cat. Although distinctive, it is closely related to the widespread Leopard Cat of Asia, and research has shown that there has been a relatively recent exchange of genetic material between the continental and island populations. Although unique, Japan’s Iriomote Cat is, therefore, a relatively recently derived species.
So, if the animals of the northern and southern islands of the Nansei Shotō remain closely related to those of the main islands of Japan or of southern Asia, from which they have only recently diverged, what of the animals living on the central islands of Okinawa and Amami Ōshima? Here, the situation is very different indeed. Amami Spiny Rat, Okinawa Spiny Rat, Ryukyu Long-haired Rat and Amami Rabbit are four very distinctive mammals of these islands. Research has shown their levels of endemism to be so high that they are considered to belong not only to distinct species but also to distinct genera (the next highest category into which species are grouped), and bear witness to the ancient isolation of these fascinating islands.
Ryukyu Spiny Rat of Amami Ōshima, Tokuno-shima and Okinawa is particularly special, as its lineage is as distinct as that of the various groups of mice, rats, voles and field mice. Ryukyu Long-haired Rat, however, has been found to be much more closely related to the better-known rat species that belong to the familiar genus Rattus. Despite occurring on the widely separated islands of Amami Ōshima and Okinawa, Ryukyu Long-haired Rat populations differ only slightly. This is a consequence of relatively recent gene flow between them, even as recently as during the latter half of the Pleistocene, and perhaps as recently as during the last glacial maxima 15,000–24,000 years ago. Conversely, the degree of divergence between the populations of Ryukyu Spiny Rat on the much more closely situated islands of Tokuno-shima and Amami Ōshima is very high.
It seems that, although populations of Ryukyu Long-haired Rats mixed across the land bridges formed between Okinawa, Tokuno-shima and Amami Ōshima during periods of low sea levels, spiny rat populations did not experience the same gene flow. So, what was it that kept these spiny rat populations separate, despite the islands on which they lived having been joined together? It seems that a major divergence between the spiny rat populations of Tokuno-shima and Amami Ōshima had already taken place, such that even when land bridges reconnected their ranges they were unable to breed together. Although to the untrained eye spiny rats from the Tokuno-shima and Amami Ōshima populations look identical, an examination of their chromosomes reveals an enormous difference. Whereas those from Tokuno-shima have just 25 chromosomes, those from Amami Ōshima have 45. Such a massive difference would have made it impossible for individuals from the two populations to breed together, and so they would have maintained their differences, even while the long-haired rats were blending theirs.
As genetic research has already revealed such a major difference between the geographically close populations of spiny rats on Amami Ōshima and Tokuno-shima, it is more than likely that future research will prove the population on Okinawa (which have 44 chromosomes) to be equally distinct. In fact, all three populations have been elevated to separate species on the basis of their genetic differences.
In addition to those two rats, the other highly distinctive mammal of the region is Amami Rabbit. This forest creature, although immediately recognizable as a rabbit of sorts, has, on closer inspection, so many distinctive characteristics that it is obviously not closely related to other rabbits and hares, either of Eurasia or of North America.
Amami Rabbit’s lineage is an ancient one. The divergence of the genus Pentalagus, of which this rabbit is the only living species, occurred some 10 or 20 Mya, making it a truly ancient relict species. Its continued survival, along with both species of Ryukyu spiny rat and Ryukyu Long-haired Rat, in such a small cluster of islands is remarkable, and it forms a vital part of Japan’s biodiversity legacy. Unfortunately, rapid forest destruction and the introduction of alien predators to its home islands may prove its undoing.
Japan is extraordinary for its size in that each region has endemic mammalian species, but whereas Hokkaidō, at the northern extremity of the Japanese archipelago, despite having the richest mammal fauna of any part of Japan, has only one small endemic rodent, Hokkaidō Red-backed Vole, Central Japan has many more endemics. So, what makes the Nansei Shotō exceptional? Although other regions of Japan have endemic mammals, in the Nansei Shotō a very high degree of endemism has survived for millions of years despite the very small area of the islands.
What remains unclear about the evolutionary history of the animals of the Nansei Shotō is whether their ancestors arrived in the Nansei Shotō during ancient times, at least 10–20 Mya, and diverged after they had arrived, or whether their lineages diverged first, somewhere outside the region, and they then reached what are now the Nansei Shotō Islands. Whether they arrived and then diverged, or diverged and then arrived, the animals of the Nansei Shotō are, nevertheless, very special indeed.
The mammals of the Nansei Shotō may be inconspicuous and rather difficult to observe, but their distributions, and the relationships between the different species and populations on these islands, provide a fascinating insight into the past history of the archipelago, with its past connections to the north into Honshū, to the south into Taiwan and the Asian continent.
It takes no stretch of the imagination to picture how rising sea levels flooded lowlands, separating larger islands into smaller ones, and marooning creatures on the separate parts of what was once a larger homeland. Similarly, it takes little mental effort to envisage small mammals, rodents, rabbits and the like, being unable to swim the deeper channels that were formed in this way between the new islands. In such isolation they might easily evolve slowly in time in tune with local conditions, and so end up distinctly different from their earlier close relatives. We might expect, therefore, that the closer islands are together the more likely they are to share the same fauna and flora and, conversely, the farther they are apart the more likely it would be that their fauna and flora would become more and more distinctive over time. These very patterns are observable among the mammals of the Nansei Shotō.
Amazingly, other groups of animals show the same pattern. Amazingly because, while it is easy to imagine terrestrial animals being divided by water channels, it is much harder to think of birds being affected by the same ‘barriers’, yet that is exactly what we find. Given that typical birds can fly, why then would they be so affected by the distribution of islands and the distance between them? Of course, by no means all birds are affected. Many species are migratory, and many migratory birds fly from one island to the next seemingly with relatively little effort. Many species of small bird pass through on migration, and they are quite obviously able to cross water, so how is it that some are unable?
Even within the same species, some populations migrate longer distances than others, while some populations may not migrate at all. Migratory avian populations and migratory species have longer wings than non-migratory ones have. Although non-migrants have wings for flight, they do not have wings suitable for migration. In particular, those birds living in a rich forest environment may have only small territories and so they may need to fly only short distances. In general, fewer predators live on islands, especially small islands, so the birds living there have less reason to fly; they no longer have predators to flee from, so their unnecessary ability to fly long distances may decline over time. In fact, among the birds that live on islands quite a high proportion are entirely flightless, having given up the ability. Typical among these are the rails. These sturdy, omnivorous, ground-dwelling birds do well on islands, or at least they did until humans introduced rats, cats, pigs and the like. Flightless island rails are especially vulnerable to alien predators. The endemic rail of Guam was completely wiped out in the wild by 1987, as a consequence of Brown Tree Snake being introduced to its home island. The snakes, which were alien to the island, were intended as controllers of rodents but, finding the native birds easy to catch, they consumed them quickly. Thankfully, captive breeding has saved the Guam Rail and made it possible to introduce these birds to a nearby snake-free island. Many other flightless island species, however, have disappeared before such conservation efforts could be put into action.
The flightless Okinawa Rail is confined to the forests of the northern part of Okinawa, which it shares with the equally restricted, and even rarer, Pryer’s Woodpecker. And on Amami Ōshima is the very attractive forest-dwelling Lidth’s Jay, the secretive Amami Thrush and Amami Woodcock. The last-mentioned is seemingly more reluctant to fly than its widespread relative Eurasian Woodcock, but it can fly and it is found also on Tokuno-shima and in northern Okinawa, in addition to Amami Ōshima. But did Amami Woodcock evolve on one of these islands, and manage somehow to cross later to the others, or did it previously have a wider range on what was once a larger island that now has been divided into several islands by changes in sea level? Either way, here in these central Nansei Shotō islands there are at least five distinct bird species that occur nowhere else in the world, not even on the northern or southern islands of the same archipelago. That is an extraordinary coincidence, for they live in the very same central region, within the Nansei Shotō, where the most unique mammals are also found.
The northernmost and southernmost islands of the Nansei Shotō have few endemic mammals, and that pattern is repeated among the birds. Although the ranges of a number of birds, such as Watercock and White-breasted Waterhen, creep north just far enough to include the southernmost Nansei Shotō islands and nowhere else in Japan, none of the species there is locally endemic. Similarly, some species more typical of Kyūshū and farther north extend just down into the northern islands and no farther south, but again no birds are locally endemic to those northern islands. It seems that the basic pattern of distribution of endemic species on these islands is very much the same for both mammals and birds, even though the latter can fly.
Among the mammals, there are also those that occur on a number of islands in the Nansei Shotō archipelago but not beyond it. In other words, they are endemic to the archipelago as a whole, but they are not so restricted as some of the other species that occur on only one or two of the islands of the Nansei Shotō. Ryukyu Fruit Bat is one such example among the mammals. Once again, there are amazing parallels among the birds, species such as Ryukyu Robin and Ryukyu Green Pigeon having very similar, archipelago-wide ranges without being restricted to any particular islands. Further investigation may show that it is species which have more general, broader habitat requirements that have the wider ranges within the islands, while the more restricted species may be confined because of their narrower habitat requirements.
There is still so much to learn about these fascinating islands, and every piece of the puzzle that is found and fitted in to place makes them seem even more like ‘the Galápagos of Asia’.
Now let us look at each island group in turn.
The highest point between the peaks of the Japanese Alps of Honshū that exceed 3,000 m and the even higher peaks of Taiwan is a distinct, lonely massif: the mountainous island of Yaku-shima. Rising from the sea, approximately 70 km from Cape Sata, in south Kyūshū, and isolated since its formation 15 Mya, Yaku-shima has not merely a distinctive character, and a particular local climate, but also a unique blend of natural history. For this reason this small piece of coral-fringed paradise was made a national park and was further honoured in 1993 as a World Heritage Site, Japan’s first World Cultural and Natural Heritage site, listed by UNESCO.
Formed by an upthrusting of subterranean granitic magma, it is now roughly circular, and conical. Yaku-shima is only 505 km2 in area, with a 132-km circumference, yet it is pinnacled and turreted with 40 granitic peaks rising above 1,000 m. The highest mountain, Mt Miyanoura, reaches a challenging 1,936 m, although Kurio (1,860 m), Okina (1,850 m) and Kuromi (1,836 m) are not far behind, making Yaku-shima a miniature oceanic and forest-clad Alps, and a wonderful wandering ground for the avid hiker.
These ‘mountains of the sea’ generate their own climate – a very wet one. Warm Yaku-shima ‘enjoys’ an incredible rainfall and is renowned (jokingly) for having ‘35 wet days a month’, creating superb waterfalls and rivers amid these misty, moss-covered peaks. Yet within a very short distance there are staggering climatic changes. Whereas the coastal climate is subtropical, mild, frost-free and relatively dry, with a mere 400 cm of rain each year and temperatures reaching 35°C, the subalpine mountain peaks, with their granitic boulder landscapes, experience the southernmost regular snows in Japan, and yearly precipitation amounting to a staggering 1,000 cm.
The diverse topography and climate of Yaku-shima leads in turn to a diverse and exceptional plantlife. Uniquely in Japan, plants at home in the subtropics and others equally at home in subarctic climates occur within a few kilometres of each other along a very short vertical distribution. Tropical plants, including hibiscus, bougainvillea, bananas, passion fruit, papayas and Japan’s northernmost groves of banyan trees, grow near sea level around the coastline. Not far above grow typical temperate mixed forests, home to conifers and broadleaf trees, rich with blossom in spring and coloured foliage in autumn, but additionally carpeted with mosses. Higher still there are species at home in the cool-temperate zone. Above 700 m, the forests consist largely of the island’s world-famous cedars and cypresses. These towering, lofty and long-lived trees create a calm cathedral-like atmosphere.
Of the 1,900 species and subspecies of flora recorded from the island 94 are endemic, while Yaku-shima’s most memorable botanical feature is its ancient cedars. Even some of the ‘younger’ ones are more than 1,000 years old, while standing among (and before) them is what is reputedly the world’s oldest tree, the Jōmon Sugi. This venerable tree, measuring 25·3 m high, 16·4 m around the trunk and an astonishing 43 m in circumference at the base, is reputed to be as much as 7,200 years old. It even outranks California’s oldest Giant Sequoia, known as the General Sherman Tree, which is estimated to be 3,000–4,000 years old; it outspans the 4,000-year-old Patagonian Cypress, and easily outstrips the ‘ancient’ 4,800- to 5,100-year-old Bristlecone Pine of the high mountains of California, Nevada and Utah.
While at mid-elevations Japanese Cedar, Jolcham Oak, Itaji Chinkapin and Isu Tree are typical of the laurel forests, at lower elevations and beside the coast there are Curtain Fig and Sea Fig and forests of Oval-leaf Mangrove.
In ancient times, the mountains of Yaku-shima were thought to be populated by gods, and the cryptomeria forests were considered sacred. By the late 16th century, however, their value as timber was already recognized, especially for the massive Yakusugi boles required for the building of temples in Kyōto, then Japan’s capital.
Today, the serpentine roots of the ancient cedars embrace a forest floor of rocks and fallen timbers that are thickly moss-covered and slippery. In the ever-present dampness, the rich, loamy smell of the forest pervades the air through which the gnarled cryptomeria trunks struggle slowly skywards.
At higher levels, where dwarf bamboo is abundant, secretive diminutive Japanese Deer browse in small herds. At lower levels, the woods echo to the sounds of Japan’s most vocal groups of Japanese Macaques. Here, in such dense forest, individuals call more often than in other habitats in order to maintain contact with other group members. Another unique feature of these macaques is their big cheek pouches. They nibble off large quantities of tree fruits, and these they store in their pouches for consumption later. Spitting seeds as they go, it is the monkeys that contribute to the spread of the next generation of trees. The southernmost population of Japanese Macaques on Yaku-shima share their range with the Japanese Deer. They live with them in a loosely mutualistic relationship, following each other, in a relationship reminiscent of that between Spotted Deer and Rhesus Macaque and Northern Plains Langurs of India. Both deer and monkeys live in social groups and quickly give alarm calls when disturbance or danger threatens; in the case of these species’ pairings, deer and monkeys each understand and respond to the other’s alarms. It is the deer that reap the further benefit, however, because monkeys are messy eaters. As they climb into trees they select only certain leaves or fruits to eat and send a steady rain of detritus, discarded leaves and the like, down on to the forest floor below. To the deer, the miraculous appearance of fresh food from above must seem like manna from heaven – delicious floral morsels falling from above. While the deer appreciate the rain of leaves, buds and flowers that the monkeys let fall, the monkeys appreciate advance warning of disturbance at ground level.
The heavy rainfall that regularly drenches Yaku-shima makes life miserable for the soaked macaques, which huddle together in family groups against the rain. The same dousing, however, is ideal for damp-loving plants such as mosses and ferns. There are 300 species of fern living on this island alone. Typical of Yaku-shima as a botanical stepping stone, 42 of them are at their northern limit here and 43 are at their southern limit. At the highest levels of all, the lush moss-draped conifer forests give way to Yaku-shima’s astonishing mountain-top alpine zone, where subarctic plants capable of withstanding buffeting winds and cold temperatures grow. When the rain clouds deign to disperse, magnificent views include all the peaks across the island, the forests below and, in turn, the turquoise ocean below them.
It was for this exceptional natural beauty, its outstanding natural phenomena, and the diversity of its flora and fauna that Yaku-shima was included in the World Heritage List. It is because of its height and geographical location that Yaku-shima has a unique transitional ecology, combining communities that reach their northernmost and southernmost limits. For example, a large number of tropical plants and butterflies, along with other tropical insects and animals, exist at their northernmost limits here, while certain temperate and alpine species occur here at their very southern limits. Neither Japanese Macaque nor Japanese Woodpecker, for example, occur farther south than Yaku-shima.
Loggerhead Turtle [VU] and Green Turtle [VU] visit to nest on the beaches, and the laying frequency of the former is so marked that the nesting beach was designated a Ramsar Site in 2005.
It is the coexistence, on one small island, of plants and animals from such widely different climatic zones that makes Yaku-shima so special. In profile, from the sea, it is Yaku-shima’s mountains that stand proud but, up close, it is the great and venerable cryptomeria trees and the richness of life at this crossroads in the ocean that make Yaku-shima so very special. For Japanese Macaques, Japanese Deer and the many other animals and plants that live here, however, Yaku-shima is not merely special, it is home.
Detailed observation and analysis of a small part of the Japanese islands can reveal a representation of the patterns that are written large across the archipelago as a whole. There is an apparent nested pattern of distribution, in which species with larger ranges overlap species with smaller ranges, which in turn overlap species with smaller ranges within them, reminding me of nested Russian dolls, known as Matryoshka. Hence I refer to this situation as ‘Matryoshka Distribution’.
When we look, for example, at Japan’s southwest islands, we find a particularly high level of endemism in the central Ryūkyū region spanning the Amami and Okinawa Islands, separated from other regions by the Tokara Strait to the north and the Kerama Gap to the south. These deeper channels reflect the longer isolation of this central group of islands and help to explain the high level of endemism locally.
An examination of the avifauna of the Amami Islands, in Japan’s northern Nansei Shotō, reveals what we might expect from an island with a much smaller area, namely that it supports fewer species in total than do Japan’s main islands, although a similar range of species – some resident, some migrant, some widespread, some local and some endemic – can be found.
Analysis also reveals species with a wider distribution, for example species that are widespread throughout Japan from Hokkaidō to Okinawa, such as Japanese Pygmy Woodpecker and Brown-eared Bulbul. There are also species that range fully through Japan’s southern archipelago, the Nansei Shotō, including the island of Amami Ōshima, but are absent from Japan’s main islands, such as Ryukyu Scops Owl, Ryukyu Green Pigeon and Ryukyu Robin. Then, there are those species which occur only on Amami Ōshima and its very closely associated islands, and nowhere else in the world, such as Owston’s Woodpecker, Lidth’s Jay and Amami Thrush. In prolonged isolation these endemic species have evolved in situ.
The Amami Islands are now registered as the Amami Guntō National Park, which includes the island’s coastal coral reefs, its mangroves, coastal mudflats and rivers and its forest ecosystems. The island offers dramatic coastal scenery and sea cliffs of Ryukyuan Limestone. While some islands have limestone caves, others support some of Japan’s largest subtropical laurel forests. The park protects a host of unique taxa, including the endemic Amami Ishikawa’s Frog, the endemic Amami Tip-nosed Frog, the regional endemic Ryukyu Tube-nosed Bat and regional endemic Yanbaru Myotis (which, despite its name occurs also on the Amami Islands), and the endemic Amami Damselfly.
Recent studies of the evolutionary histories of some of the endemic species living in the Nansei Shotō have added considerably to our understanding of the evolution that has occurred in this little-known archipelago. Mammals from the northern part of the island chain closest to Kyūshū (for example, Japanese field mice) and mammals from the southernmost part of the island chain closest to Taiwan (for example, Iriomote Cat) exhibit relatively low levels of endemism, whereas those from the more distant, central islands show high levels of endemism. Three mammals, namely Ryukyu Long-haired Rat, Ryukyu Spiny Rat and Amami Rabbit, are so genetically distinct from other populations of similar species that each is assigned to its own unique genus, indicating that these species have been living here for a very long time. All three are regarded as endangered, and each is protected as a Japanese Natural Monument. The lineages of these species have been traced back to their origins in the Tertiary Period (from approximately 66 million to 2.6 Mya); their continued survival in such a small island cluster is nothing short of miraculous.
Recent research on the mammals of the Nansei Shotō has shown that Ryukyu Spiny Rat, once considered a single species ranging across Amami Ōshima, Tokuno-shima and Okinawa, should be reassigned as three separate species: Amami Spiny Rat, Tokunoshima Spiny Rat and Okinawa Spiny Rat. It seems that the spiny rat populations on each of the three islands (Amami Ōshima, Tokuno-shima and Okinawa) have diverged significantly from one another, the population on Okinawa even having twice as many chromosomes as that on Amami Ōshima. The Okinawa population also has very large sex chromosomes, whereas sex chromosomes are absent from both the Amami Ōshima and the Tokuno-shima populations. This absence makes these two populations particularly odd, because sex chromosomes are normally essential in mammals for creating distinct males and females.
Research on sequence variation of mitochondrial DNA of Amami Rabbit has shed light on its long-disputed phylogenetic position, revealing that its lineage has been as independent as that of any other genus in the Leporidae.
With tyres spinning on a muddy gravel track through the forest, I tightly grasped the steering wheel with one hand; I was searching for – rabbits. In the other hand I held a searchlight, playing the beam around curves into the blackness unreached by the headlights – not just any rabbits. A third hand, to prevent my binoculars from crunching into my ribs as I bounced up and down in the vehicle while I drove the rutted trail, would have been useful; and even a fourth would have helped with changing gears, and occasionally wiping away the annoying trickle of sweat which ran down from my forehead into my eyes – very special rabbits indeed. Dense vegetation swept the sides of my van, and trees overhung the track, so my scanning pattern with the searchlight had to be a swift sweep around each corner, along each side of the track, under the overhanging brush, along the exposed branches and trunks overhanging the route, and up and down the banks at each side of the track. No, I was not in a jungle-warfare drivers’ training camp; I was searching for the mysterious black Amami Rabbit of Amami Ōshima.
Unusual in many ways, the elusive rabbit on Amami Ōshima is, for starters, entirely nocturnal. Furthermore, it is a relaxed creature, alert but generally unhurried, ready to melt into the darkness at any moment. It is large, measuring up to 50 cm in length, and weighs up to 2 kg, appearing heavy and thickset. Just occasionally, these rabbits use forest tracks as routes, too, lolloping or hopping across them, lingering briefly under the dense overhanging vegetation at the edge, revealing red-reflecting eyes and rather short, pinkish rounded ears. Another feature always conspicuous for its absence is the tail. This rabbit has no flashing scut with which to signal, in fact nothing at all – it is tailless.
At first sight, it is the extremely dark blackish-brown fur that seems most striking, but coat colour in fact is extremely variable among mammals. More distinctive during a closer view is Amami Rabbit’s proportions. It lacks the rangy, long-limbed appearance of any of the hares, and it does not share the delicate proportions of the European Rabbit. Instead, it has a rather large head with small ears and eyes, a rather short body and very stout hind legs. Its short, rounded ears are more like those of a pika than those of any other rabbit species, and tipping its stout legs are surprisingly prominent claws. All these points make for a most distinctive creature, and one that is instantly recognizable.
Few species come stranger than this mysterious creature, the largest wild animal on the Amami Islands. It was discovered in 1896, and described to science in 1900, and it occurs only on Amami Ōshima and adjacent Tokuno-shima. In the century since being described, the secret life of Amami Rabbit has eluded most biologists. So different is its morphology from that of other members of the family Leporidae that the ancestors of Amami Rabbit are considered to have diverged as an early side branch from the main leporid branch some 10–20 Mya, or about half as long ago as the pikas and the ancestral rabbits separated.
This rabbit occurs in dense hill forest. There, it excavates simple dens rather than the burrow systems of other rabbit species. Its strong hind legs and conspicuously heavy claws (at 1–2 cm they are unusually long for any rabbit) are ideally adapted to this difficult terrain and digging habit.
During daytime, Amami Rabbits retreat to holes among rocks or under tree stumps, or to burrows that they have excavated in the heavy soil of the mountain slopes. Each evening at dusk, they leave their dens and forage all night on the forest floor, then return at dawn. They move slowly about the forest, following traditional trails, browsing on ground-level vegetation, including growing and fallen leaves, ferns and grasses. All the while they are constantly alert, regularly sniffing and looking around, seemingly very sensitive to both sound and smells.
Just how did this extraordinary creature come to be so different from other rabbits, and how did it come to be so isolated in the Ryūkyū Islands?
While other rabbits may rear large litters of young several times in a year, Amami Rabbit is a very slow-reproducing animal, breeding just twice, from October to December and again during April and May, and then rearing only one young each time.
Amami Rabbit burrows are simple, short, and typically occupied by just one animal. The female rabbit digs a special birthing and rearing den about 1·5 m long, and ending in a chamber about 30 cm wide and high and 60 cm long and lined with vegetation. Amazingly, she seals her newborn youngster into it. Once every two nights she approaches, sniffing cautiously around the sealed den to make sure that all is well, and repeating a whistled call. Then, using her mouth and forelegs, she digs away the soil that plugs the den, creating an opening. Once her youngster comes to the entrance, the two nuzzle each other, and then, eventually, the female rises up on her haunches facing her youngster, enabling it to suckle. The female signals the end of suckling by lowering her front legs, and the baby returns to the safety of its den. She usually grooms near the entrance before closing the den again. Sealing is a prolonged process, requiring 20–30 minutes for her to scoop up loose soil, fill in the hole and tamp it into a hard plug with her feet. She then wanders off into the forest.
Each spring and autumn, the secretive forest rabbits of Amami Ōshima breed, rearing just one youngster each breeding season. Rearing takes 30–50 days, and, with feeding as infrequent as just once every two nights, Amami Rabbit’s milk must be especially rich and creamy. As the youngster develops, it spends more time outside the den after suckling, eventually learning to dig its own way out. Finally, when the female decides that the youngster is large enough, she no longer seals the den when she heads off into the forest to forage. A couple of nights later, when the mother calls her youngster, it leaves the isolated birth den for the first and last time, and sets off with her. As mother and youngster wander together in the forest, they communicate by calling and by thumping the ground with their hind legs as a warning signal. If left well alone, even this species’ distinctively slow rate of reproduction is just one more way that it remains in tune with the ancient forest environment of its island home.
The unique mammals of the Ryūkyū Islands have co-evolved with venomous snakes, in particular various pit-vipers (Okinawa Habu, Princess Habu and Tokara Habu), but not with predatory mammals. Rats accompanied human settlers to the islands and, since agriculture has spread so, too, have the rats, providing increased food for the Habu, which apparently increased in response. In order to control the snakes, local people introduced Small Asian Mongoose to the islands. This failed attempt at biological control has threatened the integrity of the natural forest ecosystem, because the mongoose, instead of eating the snakes, eats various indigenous and endemic mammals, although perhaps even that threat was not so drastic as clear-felling the forests, which was undertaken at an astonishing speed over the same time frame. Today, however, the forests are better protected, and efforts are underway to eradicate or at least limit introduced alien predators. Thus, the future for Amami Rabbit is looking brighter.
The slow pattern of Amami Rabbit’s life ticks away with the seasons, marking the years of an ancient forest that has been isolated for millions of years, and which is home to a fascinating suite of creatures found nowhere else on earth. The ancient mammalian, avian, reptilian and amphibian lineages of the Nansei Shotō, and the natural forests that they inhabit, are not merely a unique Japanese legacy, they are symbolic of the significant biodiversity of the central Nansei Shotō, and of the need for effective conservation here; they are of global significance.
1 Beginning approximately 2 Mya and lasting until the present.
2 Although described originally as a unique species in a unique genus, the taxonomic position of Iriomote Cat (once Mayailurus iriomotensis) has caused much debate. Morphology suggests that it is unique, although more recent genetic research has led to its being considered a subspecies of the widespread Leopard Cat Prionailurus (or Felis) bengalensis.