Local tales of a song'aomby (Malagasy for 'cow that is not a cow'), or pygmy hippopotamus, led French naturalist Alfred Grandidier to follow a village headman to a marsh in southwestern Madagascar, a site called Ambolisatra, which became the first known subfossil site in Madagascar. In 1868, Grandidier uncovered the first subfossil remains of lemurs—a humerus from Palaeopropithecus and a tibia of a sifaka. The Palaeopropithecus remains were not described for several decades, and it took decades more for the remains to be correctly paired with other sloth lemur remains. It was not until 1893 that giant lemur species were formally described, when Charles Immanuel Forsyth Major discovered and described a long, narrow skull of Megaladapis madagascariensis in a marsh. His discoveries in various marshes of central and southwestern Madagascar sparked paleontological interest, resulting in an overabundance of taxonomic names and confused assemblages of bones from numerous species, including non-primates. Specimens were distributed between European museums and Madagascar, often resulting in the loss of field data that went with the specimens, if the data had been recorded at all.

In 1905, Alfred Grandidier's son, Guillaume Grandidier, reviewed subfossil lemur taxonomy and determined that too many names had been created. His review established most of the presently known family and genera names for the extinct lemurs. Despite the taxonomic clarification, subfossil postcrania from different genera, particularly Megaladapis, Palaeopropithecus and Hadropithecus, continued to be incorrectly paired and sometimes assigned to non-primates. Since subfossil remains were often dredged from marshes one by one, pairing skulls with other bones was often guesswork based on size-matching, and was not very accurate as a consequence. Even as late as the 1950s, bones of non-primates were attributed to subfossil lemurs. One reconstruction of the confounded subfossil remains by paleontologist Herbert F. Standing depicted Palaeopropithecus as an aquatic animal that swam near the surface, keeping its eyes, ears, and nostrils slightly above water. Postcranial remains of Palaeopropithecus had previously been paired with Megaladapis by Guillaume Grandidier, who viewed it as a giant tree sloth, which he named Bradytherium. Standing's aquatic theory was supported by Italian paleontologist Giuseppe Sera, who reconstructed Palaeopropithecus as an "arboreal-aquatic acrobat" that not only swam in water but climbed trees and dove from there into the water. Sera took the aquatic theory further in 1938 by including other extinct lemurs, including Megaladapis, which he viewed as a thin ray-like swimmer that fed on mollusks and crustaceans while concealed underwater. It was primarily the paleontologist Charles Lamberton who correctly paired many of the confused subfossils, although others had also helped address problems of association and taxonomic synonyms. Lamberton also refuted Guillaume Grandidier's sloth theory for Megaladapis, as well as the aquatic lemur theory of Standing and Sera.

Excavations during the early 20th century by researchers like Lamberton failed to unearth any new extinct lemur genera. Fourteen of the approximately seventeen known species had previously been identified from field work in southern, western, and central Madagascar. When paleontological field work resumed in the early 1980s, new finds provided associated skeletal remains, including rare bones such as carpal bones (wrist bones), phalanges (finger and toe bones), and bacula (penile bone). In some cases, nearly complete hands and feet were found. Enough remains have been found for some groups to demonstrate the physical development of juveniles. Standard long-bone indices have been calculated in order to determine the intermembral index (a ratio that compares limb proportions), and body mass estimates have been made based on long-bone circumference measurements. Even preserved fecal pellets from Archaeolemur have been found, allowing researchers to learn about its diet. More recently, electron microscopy has allowed researchers to study behavioral patterns, and DNA amplification has helped with genetic tests that determine the phylogenetic relationships between the extinct and living lemurs.

A new genus of sloth lemur, Babakotia, was discovered in 1986 by a team led by Elwyn L. Simons of Duke University in karst caves on the Ankarana Massif in northern Madagascar. Along with Babakotia, a new species of Mesopropithecus, M. dolichobrachion, was also discovered, but not formally described until 1995. The same team has also helped promote new ideas about sloth lemur adaptations and the relationships among the four genera. They have also provided evidence that living species, such as the indri and the greater bamboo lemur, have lost much of their original range. In 2009, a new species of large sloth lemur, called Palaeopropithecus kelyus, was described from northwestern Madagascar by a Franco-Madagascan team. The new species was found to be smaller than the two previously known species from the genus, and its diet reportedly consisted of more hard-textured food. The resurgence in subfossil lemur work has also sparked new interest in Madagascar's small mammals, which have also been found at the subfossil sites. This has led to new ideas about the origins, diversity, and distribution of these animals.

The number of Malagasy subfossil sites containing subfossil lemurs has increased significantly since the mid-20th century. At that time, subfossil lemurs had only been found in the center, south, and southwest of the island. Since then, only the eastern rainforests have not been represented, and paleodistributions are now known for both extinct and living species around most of the island. Large quantities of subfossil lemur remains have been found in caves, marshes, and streambank sites in drier regions. The subfossil sites are clustered together geographically and are recent in age, mostly dating between 2,500 and 1,000 years old, with a few spanning back into the last glaciation, which ended 10,000 years ago.

At least 17 species of giant subfossil lemur vanished during the Holocene, with all or most extinctions happening after the colonization of Madagascar by humans around 2,000 years ago. Madagascar's megafauna included not only giant lemurs, but also elephant birds, giant tortoises, several species of Malagasy hippopotamuses, Cryptoprocta spelea (a "giant fossa"), large crocodiles (Voay robustus), and Plesiorycteropus, a unique digging mammal, all of which died out during the same period. Madagascar's megafaunal extinctions were among the most severe for any continent or large island, with all endemic wildlife over 10 kg (22 lb) disappearing, totaling approximately 25 species. The most severely impacted lemurs were generally large and diurnal, particularly the clade containing the living indriids and extinct sloth lemurs. Although only the indriids are alive today and represent only a small percentage of the living lemur species, this clade collectively contained the majority of the extinct giant lemur species.

By region, the Central Highlands lost the greatest number of lemur species. It has lost nearly all of its woodland habitat, but some lemur species still survive in isolated forest patches. Lemur diversity is tightly linked with plant diversity, which in turn decreases with increased forest fragmentation. In extreme cases, treeless sites such as the town of Ampasambazimba from the central region no longer support any of the lemur species represented in their subfossil record. Other locations no longer have giant subfossil lemurs, yet they still maintain forested habitat that could support them. Even though the giant lemurs have disappeared from these locations, while the smaller species survive in the forest patches that remain, the subfossil remains indicate that the living species used to be more widespread and coexisted with the extinct species. The Central Highlands saw the greatest species loss, but was not the only region or habitat type to witness extinctions. The least-understood region is the eastern rainforests, which have not yielded subfossil lemur remains. Consequently, it is impossible to know what percentage of lemur taxa were recently lost there; studies of Malagasy customs (ethnohistory) along with archaeological evidence suggests the eastern rainforests were more ecologically disturbed in the past than they are today. Hunting and trapping by humans may have severely impacted large lemurs in this region as well.

Comparisons of species counts from subfossil deposits and remnant populations in neighboring Special Reserves has further demonstrated decreased diversity in lemur communities and contracted geographic ranges. At Ampasambazimba in central Madagascar, 20 species of subfossil lemur have been found. At nearby Ambohitantely Reserve, only 20% of those species still survive. Only six of 13 species found at Ankilitelo and Ankomaka Caves in the southwest still survive at Beza Mahafaly Reserve. In the extreme north, the caves of Ankarana have yielded 19 species, yet only nine remain in the surrounding forests. In the northwest, 10 or 11 subfossil species have been found at Anjohibe, whereas only six species remain at nearby Ankarafantsika National Park.

As with the extinctions that occurred on other land masses during the late Pleistocene and Holocene (known as the Quaternary extinction event), the disappearance of Madagascar's megafauna is tightly linked with the arrival of humans, with nearly all extinctions dating to around the same time of the earliest evidence of human activity on the island or significantly later. The exact date of human arrival is unknown; a radius (arm bone) of a Palaeopropithecus ingens with distinct cut marks from the removal of flesh with sharp objects dates to 2325 ± 43 BP (2366–2315 cal yr BP). Based on this evidence from Taolambiby in the southwest interior, as well as other dates for human-modified dwarf hippo bones and introduced plant pollen from other parts of the island, the arrival of humans is conservatively estimated at 350 BCE. Measurements of stratigraphic charcoal and the appearance of exotic plant pollen dated from Holocene core samples confirm these approximated dates for human arrival in the southwestern corner of the island and further suggest that the central and northern parts of the island did not experience significant human impact until 700 to 1,500 years later. The humid forests of the lower interior of the island were the last to be settled (as shown by the presence of charcoal particles), possibly due to the prevalence of human diseases, such as plague, malaria, and dysentery. The entire island was not fully colonized by humans until the beginning of the second millennium CE.

The extinction of Madagascar's megafauna, including the giant lemurs, was one of the most recent in history, with large lemur species like Palaeopropithecus ingens surviving until approximately 500 years ago and one bone of the extinct Hippopotamus laloumena radiocarbon dated to about 100 years BP. An even wider extinction window for the subfossil lemurs, ranging up until the 20th century, may be possible if reports of unidentified animals are true. As recently as the early 17th century, dwindling populations of subfossil lemurs may have persisted in coastal regions where tree-cutting and uncontrolled fires had less of an impact. By that date, the Central Highlands' forests were mostly gone, with the exception of scattered forest fragments and strips. Along the northwest coast, forms such as Archaeolemur may have survived for more than a millennium after the arrival of humans. This is supported by radiocarbon dates for Archaeolemur from the Ankarana Massif dating to 975 ± 50 CE as well as archaeological data that show there was little human activity in the area until a few centuries ago, with low human population density along the northwest coast until nearly 1500 CE.

In the 20th century, six hypotheses for explaining the extinction of the giant subfossil lemurs have been proposed and tested. They are known as the "Great Fire", "Great Drought", "Blitzkrieg", "Biological Invasion", "Hypervirulent Disease", and "Synergy" hypotheses. The first was proposed in 1927 when Henri Humbert and other botanists working in Madagascar suspected that human-introduced fire and uncontrolled burning intended to create pasture and fields for crops transformed the habitats quickly across the island. In 1972, Mahé and Sourdat proposed that the arid south had become progressively drier, slowly killing off lemur fauna as the climate changed. Paul S. Martin applied his overkill hypothesis or "blitzkrieg" model to explain the loss of the Malagasy megafauna in 1984, predicting a rapid die-off as humans spread in a wave across the island, hunting the large species to extinction. That same year, Robert Dewar speculated that introduced livestock outcompeted the endemic wildlife in a moderately fast series of multiple waves across the island. In 1997, MacPhee and Marx speculated that a rapid spread of hypervirulent disease might explain the die-offs that occurred after the appearance of humans worldwide, including Madagascar. Finally, in 1999, David Burney proposed that the complete set of human impacts worked together, in some cases along with natural climate change, and very slowly (i.e., on a time scale of centuries to millennia) brought about the demise of the giant subfossil lemurs and other recently extinct endemic wildlife.

Since all extinct lemurs were larger than the ones that currently survive, and the remaining large forests still support large populations of smaller lemurs, large size appears to have conveyed some distinct disadvantages. Large-bodied animals require larger habitats in order to maintain viable populations, and are most strongly impacted by habitat loss and fragmentation. Large folivores typically have slower reproductive rates, live in smaller groups, and have low dispersal rates (vagility), making them especially vulnerable to habitat loss, hunting pressure, and possibly disease. Large, slow-moving animals are often easier to hunt and provide a larger amount of food than smaller prey. Leaf-eating, large-bodied slow climbers, and semiterrestrial seed predators and omnivores disappeared completely, suggesting an extinction pattern based on habitat use.

Since the subfossil bones of extinct lemurs have been found alongside the remains of highly arboreal living lemur species, we know that much of Madagascar had been covered in forest prior to the arrival of humans; the forest coverage of the high plateau region has been debated. Humbert and other botanists suggested that the central plateau had once been blanketed in forest, later to be destroyed by fire for use by humans. Recent paleoenvironmental studies by Burney have shown that the grasslands of that region have fluctuated over the course of millennia and were not entirely created by humans. Similarly, the role humans played in the aridification of the south and southwest has been questioned, since natural drying of the climate started before human arrival. The marshes of the region (in which subfossil remains have been found) have dried up, subfossil sites have yielded a host of arboreal lemurs, and site names, such as Ankilitelo ('place of three kily or tamarind trees') suggest a recent wetter past. Pollen studies have shown that the aridification process began nearly 3,000 years ago, and peaked 1,000 years prior to the time of the extinctions. No extinctions occurred prior to the arrival of humans, and the recent climatic changes have not been as severe as those prior to human arrival, suggesting that humans and their effect on the vegetation did play a role in the extinctions. The central plateau lost more species than the dry south and southwest, suggesting that degraded habitats were more affected than arid habitats.

Over-hunting by humans has been one of the most widely accepted hypotheses for the ultimate demise of the subfossil lemurs. The extinctions and human hunting pressure are associated due to the synchronicity of human arrival and species decline, as well as the suspected naïveté of the Malagasy wildlife during the early encounters with human hunters. Despite the assumptions, evidence of butchery has been minimal until recently, although folk memories of rituals associated with the killing of megafauna have been reported. Archeological evidence for butchery of giant subfossil lemurs, including Palaeopropithecus ingens and Pachylemur insignis, was found on specimens from two sites in southwestern Madagascar, Taolambiby and Tsirave. The bones had been collected in the early 20th century and lacked stratigraphic records; one of the bones with tool marks had been dated to the time of the first arrival of humans. Tool-induced bone alterations, in the form of cuts and chop marks near joints and other characteristic cuts and fractures, indicated the early human settlers skinned, disarticulated, and filleted giant lemurs. Prior to these finds, only modified bones of dwarf hippos and elephant birds, as well as giant aye-aye teeth, had been found.

Although there is evidence that habitat loss, hunting, and other factors played a role in the demise of the subfossil lemurs, prior to the synergy hypothesis, each had its own discrepancies. Humans may have hunted the giant lemurs for food, but no signs of game-dependent butchery have been found. Madagascar was colonized by Iron-age pastoralists, horticulturalists, and fishermen, not big-game hunters. The blitzkrieg hypothesis predicts extinction within 100 and 1,000 years as humans sweep across the island, yet humans lived alongside the giant lemurs for more than 1,500 years. Alternatively, habitat loss and deforestation have been argued against because many giant lemurs were thought to be terrestrial, they are missing from undisturbed forested habitats, and their environment was not fully forested prior to the arrival of humans. Anthropologist Laurie Godfrey defended the effects of habitat loss by pointing out that most of the extinct lemurs have been shown to have been at least partly arboreal and dependent upon leaves and seeds for food, and also that these large-bodied specialists would be most vulnerable to habitat disturbance and fragmentation due to their low reproductive resilience and their need for large, undisturbed habitats. Still, much of the island remained covered in forest, even into the 20th century.

Linking human colonization to a specific cause for extinction has been difficult since human activities have varied from region to region. No single human activity can account for the extinction of the giant subfossil lemurs, but humans are still regarded as being primarily responsible. Each of the contributing human-caused factors played a role (having a synergistic effect) in varying degrees. The most widespread and adaptable species, such as Archaeolemur, were able to survive despite hunting pressure and human-caused habitat change until human population growth and other factors reached a tipping point, cumulatively resulting in their extinction.

While it is generally agreed that both human and natural factors contributed to the subfossil lemur extinction, studies of sediment cores have helped to clarify the general timeline and initial sequence of events. Spores of the coprophilous fungus, Sporormiella, found in sediment cores experienced a dramatic decline shortly after the arrival of humans. Since this fungus cannot complete its life cycle without dung from large animals, its decline also indicates a sharp decline in giant subfossil lemur populations, as well as other large herbivores, starting around 230–410 cal yr CE. Following the decline of megafauna, the presence of charcoal particles increased significantly, starting in the southwest corner of the island, gradually spreading to the other coasts and the island's interior over the next 1,000 years. The first evidence for the introduction of cattle to the island dates to 1,000 years after the initial decline of coprophilous fungal spores.

The loss of grazers and browsers might have resulted in the accumulation of excessive plant material and litter, promoting more frequent and destructive wildfires, which would explain the rise in charcoal particles following the decline in coprophilous fungus spores. This in turn resulted in ecological restructuring through the elimination of the wooded savannas and preferred arboreal habitats on which the giant subfossil lemurs depended. This left their populations at unsustainably low levels, and factors such as their slow reproduction, continued habitat degradation, increased competition with introduced species, and continued hunting (at lower levels, depending on the region) prevented them from recovering and gradually resulted in their extinction.

Hunting is thought to have caused the initial rapid decline, referred to as the primary trigger, although other explanations may be plausible. In theory, habitat loss should affect frugivores more than folivores, since leaves are more widely available. Both large-bodied frugivores and large-bodied folivores disappeared simultaneously, while smaller species remained. Other large non-primate grazers also disappeared around the same time. Consequently, large body size has been shown to have the strongest link to the extinctions—more so than activity patterns or diet. Since large animals are more attractive as prey, fungal spores associated with their dung declined rapidly with the arrival of humans, and butchery marks have been found on giant subfossil lemur remains, hunting appears to be a plausible explanation for the initial decline of the megafauna.

By region, studies have revealed specific details that have helped outline the series of events that led to the extinction of the local megafauna. In the Central Highlands, dense forests existed until 1600 CE, with lingering patches persisting until the 19th and 20th centuries. Today, small fragments stand isolated among vast expanses of human-created savanna, despite an average annual rainfall that is sufficient to sustain the evergreen forests once found there. Deliberately set fires were the cause of the deforestation, and forest regrowth is restricted by soil erosion and the presence of fire-resistant, exotic grasses. In the southeast, an extended drought dating to 950 cal yr BP led to fires and transition of open grasslands. The drought may also have pushed humans populations to rely more heavily on bushmeat. Had humans not been present, the subfossil lemur populations might have adjusted to the new conditions and recovered. Had the drought not reduced the population of the subfossil lemurs, the pressure from the small number of people living in the region at the time might not have been enough to cause the extinctions. All of the factors that have played a role in past extinctions are still present and active today. As a result, the extinction event that claimed Madagascar's giant subfossil lemurs is still ongoing.

Recent radiocarbon dates from accelerator mass spectrometry C dating, such as 630 ± 50 BP for Megaladapis remains and 510 ± 80 BP for Palaeopropithecus remains, indicate that the giant lemurs survived into modern times. It is likely that memories of these creatures persist in the oral traditions of some Malagasy cultural groups. Some recent stories from around Belo sur Mer in southwestern Madagascar might even suggest that some of the giant subfossil lemurs still survive in remote forests.

Flacourt's 1658 description of the tretretretre or tratratratra was the first mention of the now extinct giant lemurs in Western culture, but it is unclear if he saw it. The creature Flacourt described has traditionally been interpreted as a species of Megaladapis. The size may have been exaggerated, and the "round head and a human face" would not match Megaladapis, which had an enlarged snout and the least forward-facing eyes of all primates. The facial description, and the mention of a short tail, solitary habits, and other traits better match the most recent interpretation — Palaeopropithecus. Malagasy tales recorded by the 19th-century folklorist Gabriel Ferrand describing a large animal with a flat human-like face that was unable to negotiate smooth rock outcrops also best match Palaeopropithecus, which would also have had difficulty on flat smooth surfaces.

In 1995, a research team led by David Burney and Ramilisonina performed interviews in and around Belo sur Mer, including Ambararata and Antsira, to find subfossil megafaunal sites used early in the century by other paleontologists. During carefully controlled interviews, the team recorded stories of recent sightings of dwarf hippos (called kilopilopitsofy) and of a large lemur-like creature known as kidoky; a report of the interviews was published in 1998 with encouragement from primatologist Alison Jolly and anthropologist Laurie Godfrey. In one interview, an 85-year-old man named Jean Noelson Pascou recounted seeing the rare kidoky up close in 1952. Pascou said that the animal looks similar to a sifaka, but had a human-like face, and was "the size of a seven-year-old girl". It had dark fur and a discernible white spot both on the forehead and below the mouth. According to Pascou, it was a shy animal that fled on the ground instead of in the trees. Burney interpreted the old man as saying that it moved in "a series of leaps", but Godfrey later claimed that "a series of bounds" was a better translation — a description that would closely match the foot anatomy of monkey lemurs, such as Hadropithecus and Archaeolemur. Pascou could also imitate its call, a long single "whoop", and said that kidoky would come closer and continue calling if he imitated the call correctly. The call Pascou imitated was comparable to that of a short call for an indri, which lives on the other side of Madagascar. When shown a picture of an indri, Pascou said kidoky did not look like that, and that it had a rounder face, more similar to a sifaka. Pascou also speculated that kidoky could stand on two legs and that it was a solitary animal.

Another interviewee, François, a middle-aged woodcutter who spent time in the forests inland (east) from the main road between Morondava and Belo sur Mer, and five of his friends, reported seeing kidoky recently. Their description of the animal and François's imitation of its long call were virtually identical to Pascou's. One of the young men insisted that its fur had a lot of white in it, but the other men could not confirm that. François and his friends reported that it had never climbed a tree in their presence, and that it flees on the ground in short leaps or bounds. When Burney imitated the sideways leaping of a sifaka moving on the ground, one of the men corrected him, pointing out that he was imitating a sifaka. The man's imitation of the gallop kidoky used was very baboon-like. The men also reported that imitating its call can draw the animal closer and cause it to continue calling.

Burney and Ramilisonina admitted that the most parsimonious explanation for the sightings was that kidoky was a misidentified sifaka or other larger living lemur species. The authors did not feel comfortable with such a dismissal because of their careful quizzing and use of unlabeled color plates during the interviews and because of the competence demonstrated by the interviewees in regards to local wildlife and lemur habits. The possibility of a wild introduced baboon surviving in the forests could not be dismissed. The descriptions of kidoky, with its terrestrial baboon-like gait, make Hadropithecus and Archaeolemur the most plausible candidates among the giant subfossil lemurs. At the very least, the stories support a wider extinction window for the giant subfossil lemurs, suggesting that their extinction was recent enough for such vivid stories to survive in the oral traditions of the Malagasy people.

• Subfossil lemurs at Wikibooks
• Primates portal
• Madagascar portal

Books cited

• Media related to Category:Subfossil lemurs at Wikimedia Commons