The recovery programme for the critically endangered Spix's Macaw highlights the many challenges involved in wresting parrot species from the brink of extinction. Although the effort to save the Spix's Macaw has generated considerable achievements, there have also been serious disappointments. Now extinct in the wild, the Spix's Macaw faces an uncertain future. The experiences of the recovery programme to date can and must be used to map a route for the ultimate recovery of the species in the future. The interpretations and conclusions presented in this paper are the personal views of the author.
The first specimen of Spix's Macaw (Cyanopsitta spixii) was collected by Dr Johan Baptist Ritter von Spix on the banks of the Rio São Francisco in 1819¹. The region in which the specimen was taken is characterised by dry thorn and cactus scrub known as caatinga. At the specific locality where the bird was found, gallery woodland adjacent to seasonally inundated creeks was the dominant vegetation type. Despite his evident first hand familiarity with other blue macaws, Spix did not realise that he had taken a species unknown to science.
In 1832, Spix's assistant, Johann Wagler, a Professor of Zoology at the University of Munich, noticed that the bird found by Spix, and which he had named Arara hyacinthinus², was in fact a new species. In honour of its finder, he renamed the parrot Sittace Spix. Wagler realised that the bird collected by Spix was smaller than the previously-known Hyacinth Macaw and was a different colour too. It had a greyish head, black bare skin on its face instead of the yellow patches seen in the Hyacinth, and it had a smaller and more delicate bill than the bigger Hyacinth Macaw and its relatives. Later on, in the 1850s, the naturalist Prince Charles Bonaparte proposed that the species be included in a new monotypic genus -Cyanopsitta. Although the species was named, described and classified, there were very few further details available about the species' habits, distribution or status. Indeed, the next scientists to report the species in the field did so only more than 80 years after Spix's discovery. In June of 1903, Othmar Reiser saw Spix's Macaws during an expedition of the Austrian Academy of Sciences to north-eastern Brazil.
He wrote, "As I knew that Spix had discovered this rare and beautiful parrot in the area of the Francisco River near Joazeiro, I made sure to keep an eye out for it in the area described. Unfortunately without success. Any enquiries made to the local people were also negative"³. Finally, at the lake at Parnaguá in the state of Piauí more than more than 400 kilometres to the west of Juazeiro, Reiser and his companions were rewarded with two sightings: one of three birds and another of a pair.
The next report of the species in Brazil came from Kaempfer in 1927. Following a search for Spix's Macaws on the north shore of the Rio São Francisco near to Juazeiro, Kaempfer wrote that "All questions about Cyanopsitta Spixii that Spix discovered here a hundred years ago were fruitless, nobody knew anything about such a parrot". But he did track one down. A captive bird that he saw at Juazeiro railway station. Spix's Macaws were also supplied to collectors in Brazil itself and at least one was successful in breeding them. During the 1950s a parrot collector called Alvaro Carvalhaes built up a breeding stock of four pairs that between them produced 20 hatchlings (note 5). One of these later went to the Naples Zoo.
Although there was evidently a continuing flow of wild caught Spix's Macaws during the 1970s to meet international demand in bird collecting circles, the openness with which collectors declared the ownership of such rare creatures sharply declined in the late 1960s. Brazil banned the export of its native wildlife in 1967 (note 6) and the Spix's Macaw was prohibited in international trade under the Convention on the International Trade in Endangered Species (CITES) in I975. With effect from July lst that year, all international commercial trade in Spix's Macaws between countries that had ratified the Convention became illegal.
A critically endangered species
In the early 1970s, it was clear that the Spix's Macaw was very rare and possibly endangered. But there were few firm details concerning its true status (note 7).
In 1974, Helmut Sick, one of Brazil's leading ornithologists, and his field assistant, Dante Teixera, fleetingly sighted Spix's Macaws near to Formosa do Rio Prêto; a little town more than 500 kilometres west of Juazeiro on the River Prêto, a tributary of the São Francisco, in the northwest of the state of Bahia. Sick recorded two parties of birds there, one of three macaws, the other of four, flying over buriti palms (Mauritia flexuosa) (Note8). Sick and some other ornithologists believed that Spix's Macaws, like other kinds of blue macaws, fed on palm nuts, and because of where they had been seen preferred the fruits of buriti palms.
Three years after this sighting, Sick went back into the field again in search of the elusive blue birds, this time with American ornithologist Robert Ridgely. The two scientists gathered details that extended the possible range of the Spix's Macaw to embrace the north-eastern part of the state of Goiás and the southern part of Maranhão. When these details were placed alongside claimed sightings from other reliable ornithologists in the states of Piauí, the possible range of the Spix's Macaw covered a vast area of the dry interior.
But despite the huge area where the birds seemed to live, the few firm sightings that existed led Ridgely to conclude that the bird was rare. He wrote that it was a "poorly known species whose recent population may always have been small; there is no evidence of a recent reduction in numbers or contraction in range, but a thorough study of the current situation would be worthwhile". He added, significantly as it turned out, that the species was "vulnerable to the activities of bird trappers" 9.
Paul Roth, a Swiss ornithologist working in the biology department of the University of Maranhão at São Luis, Brazil, decided that it was time to undertake such a thorough study. He obtained modest financial backing from the International Council for Bird Preservation (ICBP), a network of bird conservation groups from around the world headquartered at Cambridge in England (note 10).
In June 1985, Roth set off for the first of a series of five expeditions - the last of which was to conclude in 1988 - to assess the status and distribution of the Spix's macaw. Roth searched in all of these areas, visiting some of them several times, but he only found conclusive proof for the existence of Spix's Macaws in one place. While searching in Maranhão he was directed by a bird trapper to Curaçá, a little town about 80 kilometres to the north-east of Juazeiro; near to the place where Spix collected his specimen. The trapper said that there were five Spix's Macaws there.
In April 1986, Roth set off to verify, or otherwise, the information from the bird trapper. He confirmed that there were indeed Spix's Macaws there, but only three, a pair and a single bird; the other two had already been trapped (note 11). The macaws were located in a patch of gallery woodland bordering one of the many seasonal watercourses that drained the otherwise dry country during the wet season.
Roth gleaned important information about the birds' habits. His most striking finding was how they seemed especially attached to the tall trees that grew along the creeks. In the particular little area of woodland where they lived, in a place called the Melância Creek, the three parrots were found to prefer some of the bigger caraiaba trees for nesting and as look out posts (note 12).
Roth found that the birds were creatures of habit, using time and again particular holes for nesting and specific places for sleeping. He saw how the macaws found a lot of their food on the trees that only grew by the creeks where conditions were moist compared to the dry thorny cactus scrub in the surrounding countryside. He also learnt from the local people that the African bees that had been introduced into Brazil were a problem for the Spix's Macaws. This very aggressive insect had recently colonised north-east Brazil and was said to have killed Spix's Macaws in their nest holes.
It also emerged that the Spix's Macaw, like any other bird or animal, were seen as fair game for the pot and had in the past been shot for food. But bird trappers posed the real danger. So as to help protect the few Spix's Macaws he had tracked down, Roth arranged a few of the local people to act as guards then set off again in search of others.
Roth made several more extensive searches across the north-east of Brazil investigating localities suggested by bird trappers, local farmers and ornithologists. But he found no trace of any more Spix's Macaws.
Roth arrived back at the Melância Creek in May 1987 to pay the guards he had hired to keep an eye on the few birds he had found there. There were now only two left. The month before he returned another one had been caught. This meant that the last ones he knew for definite to be left in the wild was just a single pair.
Later that year, on Christmas Eve 1987, Roth was told that a trapper had returned and taken one more, leaving only a single bird from the last three. In January 1988, Roth received news that even the last one had gone too, taken by the trappers.
The loss of this last tiny population was of grave concern in conservation circles. However, the species was not yet proven to be extinct in the wild. Indeed, there were still reports of other wild birds filtering out of Brazil. One came to ICBP from the Brazilian wildlife photographer Luiz Claudio Marigo. He provided information suggesting that the species remained in the Gerais region in a range of hills called the Chapada das Mangabeiras. ICBP decided to support an expedition to search in this region in June and July of 1990.
A long search took place but no birds were located in this region: neither was any evidence for their recent occurrence in this region. A bird dealer, however, suggested that the ICBP team should search far to the east - in the region of Curaçá.
Further leads gathered locally took the search team to the Melância Creek - the place where Roth had located the trio of birds some years earlier. A single bird was located there - probably the last one that was reported to Roth as having been trapped, but which evidently had not been.
The ICBP team determined that the species preferred a unique gallery woodland habitat found only on dry plain fringing the middle reaches of the São Francisco River. Earlier suggestions that it might be a specialist of buriti palm groves seemed erroneous. Another nearby area of similar gallery woodland, fringing another creek - the Vargen Creek - was searched and there were strong suggestions that Spix's Macaws had lived there too, but all had been trapped and removed.
Given that there were only these two areas of apparently suitable habitat still in existence, it was deemed likely that the single bird found in the Melãncia Creek was probably the last one of its kind left in the wild. The researchers who located the last bird suggested that reports of the species over the wider interior of north-eastern Brazil were mistaken - as was its assumed dependence on buriti palms. It was, it seemed, a specialist inhabitant of a very restricted habitat type (note 13). The São Francisco River was one of the main routes for colonial exploitation of the interior of Brazil. In its middle reaches around Juazeiro it became important for ranching. Towns developed and large estates emerged. The land gradually changed as people began to settle there during the 17th century. The larger trees were felled to provide timber for buildings and boats, while cultivation was practised where some of the better soils lay around Curaçá, for example. The main focus, however, was on cattle and by the start of the 18th century there was already extensive ranching. To open land for grazing, the colonists used fire.
More than a hundred years of intensive burning, logging and grazing had already preceded the early 19th century passage by Spix and Martius. What state the special creeks were in by then we can only guess at, but given the rare moisture they held, the presence of big timber trees and the potential for cultivation on their fine soils, it must be that their degradation was already well advanced by the early 19th century when the first Spix's Macaw was collected by a scientist. Roth believed that the wooded creeks once extended for 50 kilometres into the caatinga either side of the São Francisco River and occurred along a significant stretch of its middle reaches. By the time the last wild Spix was found in 1990, only two little patches of wooded creek remained, the rest was gone.
During the 20th century, further pressures on the local ecosystems emerged. In the 1950s, the World Bank funded extensive water management projects. By then the valley of the São Francisco had land-use patterns typical of much of Brazil: large estates and plantations owned by landlords with the majority of the population living in extreme poverty as tenants, labourers and squatters. The arrival of irrigation heralded the large-scale cultivation of sugar cane, soya and maize.
In 1974 the World Bank provided finance for the Sobradinho dam. The reservoir created above the barrage some 50 kilometres upstream from Juazeiro and Petrolina flooded a large area of the São Francisco valley, including any potential habitat of Spix's Macaw that might have been there. Later on, the pressures arising from oil and gas prospecting and mining would join the impact of agriculture and hydropower (note 15).
Thus the Spix's Macaw appears to be the victim of long term habitat degradation as well as more recent direct unsustainable exploitation to meet the world demand for rare captive birds. Functionally extinct in the wild, it was confirmed in 1990 that the only chance to save the species rested with a tiny captive population.
The official recovery programme
Before the final confirmation of the true status of the Spix's Macaw in the wild, efforts were already underway to establish a recovery programme. The first serious attempt to get a conservation effort agreed took place in Loro Parque in Tenerife in August 1987. At that time, there were a total of 12 birds confirmed in various locations in captivity outside Brazil and a further 12 in Brazil - 24 in all. Rumours circulated about as many as 19 others. But there was no confirmation for the existence of any of these.Although the meeting was a helpful focal point in drawing attention to the desperate situation apparently faced by the Spix's Macaw, it was not able to deliver an agreement sufficient to save the species. This was mainly because there were too few of the owners of the captive birds involved (Wolfgang Kiessling, the proprietor of Loro Parque, was the only owner in attendance) and there were no representatives from the Brazilian government in there either.
A series of further meetings took place in the months ahead. These involved various combinations of interested parties including the Brazilian Government, scientific experts, international agencies such as CITES and TRAFFIC, as well as the owners of the captive birds. These discussions led to the establishment of a Working Group for the Management of the Spix's Macaw. The Working Group in turn proposed the establishment of a Permanent Committee for the Recovery of the Spix's Macaw (notes 16 and 17). It met for the first time in July 1990 - coincidentally the very week that the ICBP team had established that the single bird in the Melância Creek was the last of its kind remaining in nature.
The objectives of the Committee were to consolidate information about the Spix's Macaw in the wild, to identify the species' historical area of occurrence, to determine the sex of the captive birds and compile a studbook and to elaborate a management plan to ensure the species' survival. This last aim was to include the determination of basic husbandry, veterinary and reproductive data so as to increase the productivity of Spix's Macaw in captivity. The first meeting also identified those centres with sufficiently well equipped facilities to breed the birds and recommended a couple of transfers of birds between owners.
At this time, there were some 15 birds known in captivity, some of the others believed to have existed some years previously had not materialised, others had died. Of the 15 known in mid 1990, there were six inside Brazil: one was in São Paulo with the private breeder Nelson Kawall, another in Recife in the north-east at the aviaries of Mauricio dos Santos and four more were in São Paulo Zoo. The rest were scattered around outside; two were in Tenerife at Loro Parque, six were in the Philippines at Antonio de Dios' Birds International facilities and one more was at Vogelpark Walsrode in Germany (note 18).
In October 1990, the Brazilian Government stated that if any other owners of Spix's Macaws were to come forward and to cooperate with the Committee, their birds would not be confiscated. The amnesty coincided with further birds emerging in Switzerland. Three birds were involved, the possession of a private Swiss aviculturist: Joseph Hämmerli.
During 1990, and following several failed attempts to do it, most of the captive birds that remained of unknown sex were examined to determine whether they were male or female. The absence until this point of even this basic information underlined how very far the embryonic efforts to save the species were from being effective.
Conservation Activities - field and aviary
In the field:Until 1990 it had been assumed that the Spix's Macaw had a vast natural range in the interior of Brazil embracing several different habitat types including buriti palm swamps, cerrado and dry caatinga. Following the ICBP expedition, it was now believed to be a specialist inhabitant of the dying gallery woodlands. Any serious reintroduction effort would therefore necessitate actions to save its habitat as well. This meant that there was every reason to carry out conservation activities in the gallery forests of the caatinga as well as through the various bird collections.
Thus, soon after the last bird was located, a field scientist was appointed to work in the local area promoting the cause of the Spix's Macaw and to gather information about the single wild parrot. He was later to be joined by others who would carry out vital and pioneering work in the native woodlands of the last wild bird.
When the last bird was located in the Melância Creek, there was a clear choice facing the Recovery Committee - to release a partner to join it, or not? On the one hand, the bird might be predated or trapped and the species placed a stage closer to final extinction; not only through becoming lost completely in the wild, but through a further and potentially fatal contraction in numbers. On the other, if the wild bird was removed, then what would be the argument for saving the last remnants of woodland? Moreover, what would be the effect on local support for the recovery plan? Also, what would be the prospects for novice captive bred birds let out into an alien world where there would be none of their kind to learn from?
The last known wild bird knew where to find water in the dry season, he knew which seeds and fruits were good to eat and he knew where to get them. He knew where the predators hid and from where they might attack. He was a vital cultural lifeline, a link between the captive birds and realising the prospect that one day there might be a successful release of macaws to repopulate regenerated and protected gallery woodlands.
During a technical meeting in late 1992, more than two years after the discovery of the wild bird, the Recovery Committee decided that it would be best to leave the last macaw out there. In addition it was agreed to follow the ICBP proposal and select a bird from the tiny captive population to be released to be his partner. If a wild pair successfully bred, then releasing captive reared macaws to team up with a small wild flock would have an even better chance of leading to the successful recovery of the Spix's Macaw in its native woodlands. But before one of the irreplaceable captive birds was released, it was absolutely vital to determine for certain the sex of the wild one.
It was believed that the last Spix's Macaw was male. But in order to check this for certain, moulted feathers were collected and sent to Oxford University for DNA analysis. The verdict took some months to reach as a completely new test had to be devised. The wild bird was indeed male.
Following the decision to leave the last wild bird at liberty and to release a partner to join him , facilities were built in the Melância Creek to assist with the reintroduction effort. A female bird was selected from the captive population for release. This parrot was captured i in 1987 and was most likely the wild bird's former partner. She was shipped to the holding facilities and prepared for release. In March 1995, she was set free.
There was a period of tension following the release as it was by no means certain that the wild bird and the newly released individual would form a pair. The uncertainty was heightened by the fact that the wild macaw had formed a cross-species pair with a Blue-winged Macaw (Propyrrhura macarana), a smaller species of macaw that also lived in - but was not confined to - the local gallery woodlands where the last Spix's Macaw spent most of its time. This unlikely pair had formed a strong bond and was often seen together. Rather than trap or kill the Blue-winged Macaw, however, it was felt safer to leave it where it was lest its removal further stress the last wild Spix.
The mixed pair and the released female were soon seen flying together as a trio. Later on the female and male Spix's Macaw were observed spending more time together. It seemed that the daring experiment had worked and that a wild pair had been established. Then disaster struck - the female bird disappeared. She had been at liberty for less than three months.
Her fate was not immediately known. However, some years later a local cowboy came forward with the the claim that he had seen the bird's dead body beneath some overhead power cables with which she had apparently collided. The male was once more on his own and it was decided that no more of the precious captive females could be spared for another release attempt.
Despite this setback, the field team working in the Melância Creek continued with their monitoring of the last wild bird. Among other interesting observations, they noted in the breeding season of 1996-1997 that the hybrid pair of parrots were spending more time around a nest hole in one of the old caraiba trees. The researchers suspected eggs had been laid. They looked and found three. With the backing of the Recovery Committee, the field team took the decision to remove the clutch and to replace the eggs with those removed from a nearby Blue-winged Macaw's nest.
It was assumed that the eggs removed from the hybrid nest would be sterile. But a close examination by scientists in São Paulo revealed that one had indeed developed an embryo. The embryo was dead and was probably not viable but was certainly a hybrid - it contained DNA from a Spix's Macaw.
Even if this unique pair of birds could not produce viable or fertile offspring, or in the judgement of the scientists be risked to rear hybrid young, perhaps they could still act as foster parents for pure Spix's Macaw eggs taken from a captive pair. It might even be that they could rear captive-hatched babies placed in their nest. Since there was little else that could now be done to reintroduce more birds back into the creek side woods, the scientists believed it was worth trying.
The first job was to determine whether the two parrots could actually hatch fertile purebred eggs and then raise infant parrots. Spix's Macaws' eggs were considered too valuable to risk in the experiment. But if they could hatch Blue-winged Macaw eggs and rear the babies, it would indicate that they were capable of rearing young Spix's.
The hybrid pair accepted replacement Blue-winged Macaw eggs taken from a nearby wild nest and successfully hatched them out, but predators ate the young. The researchers had to wait another year before the hybrid pair would try to nest once more. A further cross-fostering attempt with pure Blue-winged Macaw eggs was tried in the 1997-1998 breeding season, but this also met with failure when the eggs were once more eaten, either by mammalian predators or by a snake.
In the next breeding season the hybrid couple made two nesting attempts. The first failed, but in December 1998 the pair laid another clutch and the researchers launched a fourth effort to get the birds to rear young.
This time, instead of replacing the eggs as before with those from a pure Blue-winged Macaw nest, they were exchanged for artificial wooden ones. In January 1999, after a 23- day 'incubation' period, the researchers removed the wooden eggs and replaced them with two three-day-old hatched chicks. In March, the foster parents flew in the creek with the two youngsters. Following this success, a further experiment was now considered possible - replacing the hybrid pair's eggs with pure Spix's Macaw eggs from captivity.
In the meantime, however, another daring experiment was tried. This involved the release of captive bred Blue-winged Macaws. This was conceived as a dry run for the future release of captive-reared Spix's Macaws. The assumption was that the challenges would be quite similar in the release of both species but that this bold move should be tried with the less endangered species first.
Birds were donated by the Loro Parque Foundation to take part in the experiment. Some were hand-reared. These were deliberately selected for release because the results of how they fared would be especially useful in planning the coming phases of the recovery of the Spix's Macaw, which would inevitably involve hand-reared birds.
In November 1997, 11 Blue-winged Macaws were set up in the adaptation aviary near the Melância Creek originally built for the female Spix's Macaw that was released in 1995. In order to maximise the information that could be gathered from the birds to be set free, they were fitted with radio-collars so that their daily movements could be followed and their behaviour more closely studied.
A little over a year later, following various difficulties in selecting the correct form of radio tracking equipment, the birds were finally released. The idea was to release only one partner from each of the firm pairs that had formed, so that the birds already at liberty would stay in the local area. Food, water and nest boxes had been placed outside to ease the hardships of a wild existence on the newly released birds. The macaws were closely monitored and, as hoped, they stayed around the aviary. The other half of the pairs were released in January 1999.
Despite some early deaths, seven lived and made a successful transition to a wild existence. So successful was the release that immediately after being set free some of the Blue-winged Macaws made a breeding attempt. In Loro Parque the birds would breed from February or March until July. At Curaçá the normal nesting season would be around December to January. A breeding attempt in synchrony with the normal timetable of wild Blue-winged Macaws in that area demonstrated how the birds had adapted to local conditions. A meeting of the Recovery Committee took place at Houston Zoo in late 1999 where the Filippino owner, Antonio de Dios, had agreed to provide five young Spix's Macaws for release into the galleried woodlands of the Melância Creek. The birds would be carefully selected so as not to jeopardise the genetic base of the captive population. There would be three females and two males. With the wild male, there was the prospect of three pairs in their natural habitat.
However, before these birds could be prepared for release, disaster again struck. The single wild bird disappeared. By early 2001 he was presumed dead. This tragic, though not entirely unexpected event, marked a watershed in the recovery effort - Spix's Macaw was from that point Extinct in the Wild. At the same time, any prospect to cross-foster captive-bred Spix's Macaw chicks into the hybrid pair's nest was ended. Also, the prospects of success for a direct soft release of captive reared birds was diminished too as the wild 'tutor' was gone.
When it was determined that only a single wild bird remained, it was clear that the only survival prospects for the species depended on the captive population. However, the challenges encountered in the captive breeding programme were considerable. These embraced legal, technical and political questions.
In 1994, some four years after the Recovery Committee was formed and after some exchanges had already take place between collections, a first meeting of a captive breeding sub-group was convened in Fort Lauderdale Florida to consider a first version of the captive breeding studbook. By that time, the captive population (following the emergence of birds in Switzerland, and because of breeding successes) had risen to 31. However, only 11 were females.
Despite the slow but steady increase in numbers, the previous year had seen the catastrophic death of two females. One was a wild-caught bird kept at Loro Parque since 1985; the other belonged to private breeder Nelson Kawall in São Paulo. The second was a particularly serious loss. She was a bird taken from the wild in the early 1980s but although she had laid eggs she had never reproduced in captivity.
The pair set up at dos Santos' collection in Recife had not bred, and neither had the birds in São Paulo Zoo. At least one pair of Spix's Macaws had been in residence there since 1969, but still no breeding had ever taken place. Other new pairings had been arranged following a build-up of numbers in de Dios and Hämmerli's aviaries. The Filipino breeder had now got 18 birds, well over half of the world population at that time, including two reproductive pairs, while the Swiss Spix's Macaw population stood at five, with a proven breeding pair among them.
Against this background of only three reproductively active couples, new matches were recommended by the meeting's participants. The overriding aim was to get more of the macaws producing, especially those that had not yet passed any of their genetic uniqueness into a new generation. At the same time the new pairings were intended to establish different lineages at the different breeding centres so that one pair of macaws would not in the future swamp the entire population with its genes.
It was proposed to shift the lone male macaw from Nelson Kawall to a facility where it could be paired with a female bird. It was proposed it be sent to the Philippines. The pair from São Paulo Zoo were to go to Loro Parque. The São Paulo macaws were the single old female kept there since the mid-1970s and a male seized in Paraguay in 1987. The Loro Parque birds were the original male and its daughter. It was decided to create two new pairs out of these four birds at Loro Parque. The remaining single male from the São Paulo Zoo would go to Switzerland and a female from the Philippines to dos Santos' facilities in Recife.
The Fort Lauderdale meeting also tried to restart previously failed efforts to get the captive birds DNA-fingerprinted. This would shed light on which birds in the breeding programme were closely related to one another and therefore give clear signals on the best genetic pairings. Such information was crucial to the recovery effort.
An attempt to gather DNA data had first been made in 1991. However, when the results were circulated in September 1992, there were 23 birds in the captive population and samples for five of these (three in Switzerland, the Walsrode bird and one of de Dios' parrots) had not worked. The data were also technically deficient and the whole process needed to be repeated - but this wasn't done.
The incomplete results meant that some transfers of birds had taken place without details on the possible genetic implications of the new pairings. This was potentially very problematic. Not only was there already breeding between relatives in the captive population, but there might have been inbreeding in the wild too. All of the wild-caught birds appeared to have come from two, and possibly only one, small population near to Curaçá, and it was quite likely that parrots kept in different parts of the world were in fact close relatives - maybe even the members of a single family! The gaps and deficiencies in the DNA testing were thus problematic.
Despite this drawback, there was a continuing effort to match birds via the studbook, which was now managed by an employee of Houston Zoo. The captive population continued to grow by 2001 reaching some 64 birds.
However, the expansion was through offspring from the same few parents. Of the 20 birds held by the Swiss owner in 1999, at least 11 (and perhaps 16) had the same parents (note 20). These were an older male bird caught from the wild in 1974 and the young captive-bred female transferred from the Philippines.
At Antonio de Dios' facilities, where there had been a build up of numbers to 26 birds, 19 were from two pairs. One couple was the original adult birds he obtained in the late 1970s or early 1980s. They had hatched 12 young; one of these was the female half of the successful Swiss pair. The second pair (that had produced seven babies) were not only brother and sister but offspring from Antonio de Dios' first pair. Nearly half of the captive population was therefore descended from just six parents, who were in turn mainly related to one another.
The potential difficulties arising from a narrow genetic base and inbreeding were compounded by movements of birds unauthorised by the Committee. The first blow came with the departure of the Swiss owner from the Recovery Committee. He withdrew and then sold his birds to between one and three other Swiss bird keepers. The bulk (15 out of 20) went to one owner who at least had pledged his co-operation with the Recovery Committee. The other five were reported to have been split between two owners who were not members of the Committee, and who apparently had no intention of being so in the future. A further blow was dealt to the recovery effort when the Filippino owner shipped four of his birds from his facilities to the Gulf state of Qatar in early 2000. This transfer was not authorised by the Committee either but was apparently carried out with the assistance of the studbook keeper.
As a result of the changes in the ownership of the birds, which were not authorised by the Committee, tensions emerged which led in February 2001 to the Brazilian government agency - IBAMA - that chaired it, to suspend the Committee's operations. At that point, there was no functioning body able to take forward the recovery programme for the Spix's Macaw. Several birds remained unpaired in Tenerife, Switzerland and Brazil. Because there was no Committee, no transfers were able to take place. Also, following the collapse of the Committee, the Loro Parque Fundación, the institution that had funded most of the field work in the creeks where the last bird lived, decided to suspend its support for the recovery effort in the creeks. Thus the crucial link between the field and captive components of the programme were effectively severed.
In September 2001, the studbook keeper at Houston Zoo was removed from the Spix's Macaw recovery effort. Shortly after, Brazil made proposals of its own concerning the transfers necessary to maximise the effectiveness of the captive breeding effort. These were based on recommendations made by Natasha Schischakin, who although no longer working for Houston Zoo, was still regarded by the Brazilian officials as the studbook keeper. In March 2002 she was finally removed and replaced by a Brazilian scientist, Carlos Bianchi.
Achievements and disappointments
* The establishment of the Recovery Committee -Given the legal complexities and wide geographical spread of the captive birds, the fact that the Committee was established in a way that managed to bring all of the diverse interest groups together to work through one mechanism was quite an achievement. There is little doubt that without the Committee, matters would have been considerably worse.
* Captive breeding programme -Although there were frustrations in the time it took to make some pairings and a good deal more that could have been done (see below) to improve the efficiency of the effort, the fact that the population was quadrupled between 1990 and 2000 is certainly an achievement that must be recognised.
* On the ground support gained -A crucial element for the long-term success of the programme (which could still be a vital element for the future) was the extent to which local people came to support the programme to save the Spix's Macaw. The field team, principally with the financial backing of the Loro Parque Fundación, managed to generate a great deal of goodwill towards efforts to save the species locally. Experience in other parts of the world shows what a crucial dimension this might be in the future. It is of great concern that this aspect of the species recovery p)an is at present only supported with low scale funding from IBAMA
* Successful field experiments and research -A good deal of information was gathered from both observation and experiment that should be of enormous value later on. However, since most of the material gathered remains unpublished, it is at present of limited value.
* No population established around the last wild bird. The fact that the last bird lived for a decade was an amazing stroke of luck for the recovery programme. It is however disappointing that during this long period that it was not possible to establish further birds at liberty around the crucial nucleus of the last wild male.
* No recovery programme agreed -In spite of a great deal of effort placed in convening meetings (albeit insufficiently frequently to make smooth and efficient progress), captive breeding and fieldwork, it is remarkable that there is still no agreed overall recovery programme for the Spix s Macaw. This is an especially notable outcome given that the generation of such a plan was one of the main aims of the Committee at its inception. There is still no official policy or programme backed by agreed budgets, responsibilities or timetables.
* Habitat remains without formal protection -IBAMA initially signed an agreement with the landowner responsible for the area where the last bird was found, but this expired after five years. A proposal to purchase the most important areas was drafted in that period but was not acted upon as of mid 2001. Some fencing of important areas of woodland took place but there is still no long-term plan to combat grazing pressure and its effects on the Spix s Macaw's unique habitat. * No husbandry guide written -Despite commitments made back in 1994, there is still no single guide for the husbandry of the species in captivity. Therefore, much of the experience gained on the veterinary care, general well being and breeding of this species in captivity remains scattered and unpublished.
* Absence of comprehensive genetic analysis on which to make pairings. Despite the vital need to avoid inbreeding, there is still no comprehensive genetic analysis of the captive birds. This remains a handicap for effective captive breeding. This work, and that concerned with the production of a husbandry guide, were responsibilities delegated to the studbook keeper. Clearly it will be necessary in future to ensure that the studbook keeper has both the commitment and capacity to deliver of such important items of work.
* Difficulties created by transfers and sales -The fact that individual owners decided to sell and transfer birds out with decisions made by the Committee is a source of serious concern. The fact that these individuals chose to behave in this way is ultimately a matter for their conscience. That there was no sanction that could be applied to them highlights the centrality of the question of who owns the birds.
Never let it happen againIt is quite clear that the extinction of the Spix's Macaw in the wild should not have been permitted to occur. The last wild population should have been protected and zero tolerance shown towards those trapping and trading the species. However, it must also be recognised that the fortunes of endangered species, like the Spix's Macaw, cannot be disentangled from longer term economic trends that shape the loss and degradation of habitat and efforts to conserve critically endangered creatures must been placed in a realistic context in this regard.
Ownership questions must be resolved
A central issue that dogged the recovery programme from the start has been the matter of who owns the birds. Without clear agreement on ownership, control and coordination of the captive birds became quite impossible. The Loro Parque Fundación honourably (and uniquely among the non-Brazilian keepers) returned the ownership of the birds to the State of Brazil. Had there been overall control of the captive birds, it is more likely that decisions could have been taken faster, funds more easily orchestrated and the practical efforts to save the species rendered considerably more efficient.
The Spix's Macaw can still be saved
The key question, of course, is can the species still be saved? Yes it can. There is ample experience from around the world to encourage an optimistic outlook. However, it will require a more determined and systematic approach to that seen to date and will rely on the clarification of central questions including the matter of ownership.
The future should be based on an independent view of the recovery effort to date. Many important lessons should be drawn out of the first 12 years of the official recovery programme, and these should be used to inform the coming phases. In order to command credibility and respect from the many parties involved, such a review should be conducted by a truly independent agency that has had no direct role in the recovery effort to date. A clear process must be conceived whereby all parties can contribute their perspectives to the review.
Return control to Brazil
The recovery effort should in future be under clearer Brazilian control. This should include not only a resolution of the ownership issues in a manner that puts more influence in the hands of the birds' native country, but also the physical relocation of at least some birds to a breeding facility in the north-east of that country - preferably in the locality where the last birds were known to live. Certainly there are problems arising from the remoteness of the area and the poor infrastructure locally, but this should nonetheless not be a reason for not fully scooping and costing such an option. The studbook should be under Brazilian control and any future Committee based more firmly around Brazilian sovereignty over the species. Such a shift will only be viable, however, if the Government of Brazil more firmly and credibly commits itself to the rescue of this species. That will require both political and financial capital being raised by politicians and officials in that country. Until that happens, many will remain sceptical about Brazil's ability to play a full role in the rescue of this species.
A great many people have contributed to my understanding and knowledge of the Spix's Macaw and efforts to save it. They are too many to mention in a paper of this length but even in this limited space I would like to acknowledge the particularly helpful assistance of Dr Nigel Collar of Birdlife International, macaw ecologist Carlos Yamashita and of the Loro Parque Foundation, especially its Director, Yves de Soye.
1 Spix and Martius recorded their expedition in three volumes published as An account of travels in Brazil at the command of his Majesty, Maximilian Joseph I, King of Bavaria, in the years 1817 to 1820. The original work was published in German. An English translation of volumes one and two was completed by H. E. Lloyd and published in 1824.2 This name was attached to Spix's description of the small blue caatinga macaw in his volume called Avium Brasiliensium Species Novae published in 1824-25.3 Reiser, O ( 1926).Liste de Vogelarten welche auf der von der Kaiserl. Akademie der Wissenschaften 1903 nach Nordostbrasilien entsendeten Expedition unter Leitung des Herrn Hofrates Dr F. Steindachner gesammelt wurden. Denkschr. Akad. Wiss. Wien, math.naturwiss. Kl.4 See Juniper, T. (2002). The Last Spix. Fourth Estate: London.5 Keller, C. (1987). Up-to-date information on the Spix's Macaw (Cyanopsitta spixii [Wagler]) and the environmental situation in Brazil. Unpublished.6 Lei No. 5197 was passed in that year by the Government's forestry institute and ended the legal export of native Brazilian wildlife.7 Readers should see Collar et al 1992, Threatened Birds of the Americas, ICBP: Cambridge. This author provides for a very thorough and closely referenced account of material available on the Spix's Macaw in nature up to this time to the early 1990s.8 See for example Sick, H. (1989). Ornitologia brasileira, uma introdução. Brasilia: Editora Universidada de Brasilia.9 See Ridgely, R. S. (1981). The current distribution and status of mainland neotropical parrots. In R.F Pasquier, ed. Conservation of New World Parrots. Technical publication No 1. International Council for Bird Preservation, Cambridge, UK.10 ICBP was renamed Birdlife International in 1993.11 It was later to emerge that two birds were exported from Brazil that year via a trader in Paraguay and were sent to Europe accompanied by a forged export permit. It is possible that these two were the individuals in question.12 For example Roth, P (1989). Spix's Macaw: population, actual status and biology of Spix's Macaw Cyanopsitta spixii. Unpublished.13 Data on the creekside vegetation and the apparent reasons for its scarcity was published by Juniper, A. T. and Yamashita, C. ( 1991 ) in the Habitat and Status of Spix's Macaw, Cyanopsitta spixii. Bird Conservation International 1 (1).14 This figure was suggested to Roth by locals in the area of Curaçá.15 See Juniper, T. (2002). The Last Spix. 16 The Permanent Committee for the Recovery of Spix's Macaw was formally established under Brazilian law (Portaria no. 330) in March 1990. Its statutes were set out in a separate decree (Portaria no 331). This was nearly four years after the first attempt to agree a plan to save the species. The Committee adopted the acronym CPRAA (derived from the Portuguese title for the institution: Comitê Permanente para a Recuperação da Arahinha Azul) but for the sake of simplicity and brevity is referred to here as the `Recovery Committee' or simply `Committee'.17 Thomsen, J. (1991), provides an excellent summary of the evolution of the Committee in an unpublished memorandum prepared for TRAFFIC. A fuller account of the recovery committee and its activities is also provided by Juniper, T. (2002), The Last Spix.18 These details are according to the minutes of the first meeting of the Committee.19 Further details concerning the hybrid pair and the release of the Blue-winged Macaws n be found in various issues of Cyanopsitta, the bulletin of the Loro Parque Fundación. Juniper, T. (2002), also provides a fuller account in The Last Spix.20 The uncertainty about the parentage of five birds arises from the loss of records available to the author following the transfer of five birds outside the control of the Recovery Committee.
" Naturam expellas furca, tamen usque recurret "
( If you drive out nature with a pitchfork, she will soon find a way back)
Horace (65-8 BC)