The Indigo Macaw (Anodorhynchus leari) is found locally in the state of Bahia, Brazil, in the region of the Vaza-Barris river (Sick, Alauda 47, 59-60, 1979). The species was described from captive specimens and nothing was known about its range and habits until its discovery in the wild by Sick (1979). Here I present information on the roosting and feeding behaviour of the species and assess the likelihood of its survival. Field work was conducted from 12th to 30th July, 1983.

Roosting sites and behaviour: The range of the Indigo macaw is within the " caatinga " region of Brazil, which is dominated by thorny scrub vegetation. Altitude in the region varies from 380 to 800 metres (Ed: 1,300-2,700 ft) with daily temperatures varying between 15° and 45° C. Macaws roost in the cliffs or canyons (locally known as "Talhados" or "Serras") that vary in height from 30 to 60 metres (Ed: 100-200 ft).

Macaws roost in burrows in the top third of the cliff faces in sedimentary holes created by weathering of the sandstone. Burrow tunnels are fairly narrow, allowing passage of only one individual at a time. As many as four birds used a single burrow. Burrows were often within half a metre (Ed: just under 2 ft) of one another.

Some macaws roosted outside the burrows, clinging to the cliff or shallow shelves. The Indigo Macaw is very shy and quite different from other macaws with an extensive social organisation. Individuals leave their roosting cliffs before dawn for the feeding grounds and return after sunset. Just after sunset 2 or 3 individuals return to the roost area, flying over the canyon and calling out. They then sit quietly on the tallest tree chewing the tips of the branches. Trees used by these individuals are easily recognised by the presence of broken branch tips. After about 10 minutes these " scouts " begin to call out loudly and the rest of the flock approaches. The numbers of individuals in the flocks vary. Groups of 22, 23, 19 and 23 were seen on successive days in the same canyon. The birds fly over the canyon screeching, then land near the scout birds and sit quietly. By then it is already dark. The birds then begin to screech again as they fly directly to the cliff edge where they sit before quietly entering their roost/nest holes. The " scout " birds roost in an adjoining canyon. The latter awake before sunrise and fly screeching over the canyon harbouring the flock's nest holes. All the macaws then drop out of their holes and the whole group flies out of sight.

Feeding habitat and behaviour: The Indigo macaw, like other macaws, feeds largely on hard nuts, which it obtains from palm trees found in the thorn scrub and in pastures cleared for cattle grazing. Its most important food source is the " licuri " palm (Syagrus coronata), which grows on top of the plateau (locally known as " Raso da Catarina") and on crystalline soils in the surrounding lowlands. This palm ranges in height from about 700 cm to 2 metres (Ed: approx. 30" to 6ft ).

I saw flocks of Indigo Macaws feeding on four occasions. The birds feed on fruits on palm trees in small subgroups of 2 to 3 individuals at a distance of 5 to 30 metres (Ed: 161/2 to 100 ft ) from one another. They also search the ground (Sick, Int. Counc. Bird. Pres. Bull. 1:439-444, 1980) for palm nuts. After they seize a nut with their bill, they fly to a palm tree, where they manipulate the nut with their feet, rolling it to remove the pericarpel skin. They then make two transverse cuts before extracting the kernel. They can open " licuri " nuts (approx. 30 x 20 mm or 1" x 3/4") easily with perfect transverse cuts. There is always a sentinel keeping watch on flocks as they feed. The birds alternate their watching and feeding activities.

Conservation: The distribution of A.leari is located in a compact region of crystalline rocks and sandstone, occupying an extremely small area of only 15,000 sq. km (Ed: approx. 6,000 sq. miles) of which 60% has been thoroughly surveyed to date. Only two colonies were located in the area covered and the total number of individuals is 60. Taking an optimistic view and assuming the remaining 40% of the species' range consists entirely of optimum habitat (which is not very likely), the total population will consist of far fewer than 200 individuals.

Two heavily used paved roads cross this area. The area has been densely populated since the late 1880s and there are many foot and donkey trails in the area. Hunting is a serious problem. The local human population is very poor and hunters either eat their catch or sell both live and dead wildlife products in regional markets. The local economy depends on subsistence agriculture and free range cattle and goat ranching. The local farmers clear the " caatinga", but leave " licuri ", which in the dry season provides an important food supply for cattle, which eat its racemes and young leaves. Cattle consumption of racemes and unripe fruit may limit the abundance of ripe nuts for the macaws. Although macaws prefer mature fruit and cattle green fruit, many farmers believe that the macaws compete with the cattle for food.

Although many cliffs are seemingly available in the range of A. leari, macaws roost in only a few canyons. The reason for this is unknown. The present sandstone cliffs that are used for roosting sites are fractured. High thermal variation from day to night produces slides, and a slide occurring at night or during the breeding season could reduce substantially the population of this rare macaw..

Unfortunately A.leari is not protected by any Public Reserve or National Park. The Ecological Station of Raso da Catarina (SEMA - Secretaria Especial do Meio Ambiente) has no resident group of this species. Although A.leari sporadically feeds in this " protected " ecological reserve, cattle feed on unripe racemes at the station and the area furnishes little food.

It is difficult to be optimistic about A.leari's survival prospects as both roosting and feeding areas are privately owned, highly vulnerable and already subject to considerable human pressure.

Acknowledgments: I thank J.Hart, S.Lindbergh, P.Antas, S.Charity and G.Rebello for their comments. R.Cavalcanti made several suggestions and helped revise the manuscript. (Manuscript received 23 May 1985 and accepted 4 September 1986)

(Website editor: I visited the area in March of this year (1998) and spoke to Carlos in Sao Paulo afterwards. The total population is now believed to be some 90 individuals. Two years ago there were some 20 or 30 more, but these were trapped and many exported illegally from Brazil. At the end of May a rancher in Rondônia in the far west of Brazil was found to have 7 Lear's Macaws on his property as well as other rare Brazilian fauna. Other Lear's Macaws elsewhere are being traced and steps taken to recover and repatriate them.)