may have been the result of local spe- 
ciation number between 21 and 55. 
The variability in these figures reflects 
uncertainties in interpreting the data. 
A conservative estimate would be that 
about 20 to 25 percent of high Andean 
species are the result of local spe- 
ciation. The rest, the overwhelming 
majority, therefore appear not to be 
the result of local speciation (unless 
their speciation is so ancient that it 
cannot be detected in present-day dis- 
tribution patterns). This breakdown of 
species suggests, then, a high influx 
of immigrants, which in time have 
become endemic to the high Andes. 
In either case — local speciation or im- 
migration — the age of the fauna re- 
mains largely unknown. 
Determining where ancestral stocks 
came from can be equally compli- 
cated. We can compare lists of species 
living in paramo or puna vegetation 
with species living in other types of 
Andean vegetation (such as cloud for- 
ests and the open woodlands of the 
intermontane arid basins) or in the 
Patagonian steppes. This comparison 
reveals that the high Andes share more 
species and genera of birds with the 
Patagonian steppes than with any 
other ecological unit. This is not sur- 
prising since these two habitats are 
structurally very similar. The Andean 
puna and the Patagonian steppes, for 
example, share 48 percent of the total 
species found in them, while only 13 
percent are shared by the puna and 
the geographically closer cloud forest. 
This would appear to vindicate F.M. 
Chapman’s view, first published in 
1917, that paramo- and pu«a-dwelling 
birds originated in southern South 
America, including Patagonia, and 
moved northward and altitudinally up- 
ward along the Andean corridor, colo- 
nizing as far as Venezuela and extreme 
northern Colombia. But Andean birds 
may also have contributed to the Pat- 
agonian fauna, and there may have 
been other routes of exchange as well. 
Birds of the puna and birds of the 
intermontane arid basins of the central 
Andes are somewhat related faunally; 
these regions share up to 30 percent 
of the species living in them, so that 
some exchange appears possible. Sev- 
eral high Andean genera and some 
species also occur in North America, 
which suggests another possible ave- 
nue of faunal flow. Examples among 
the species include the marsh hawk, 
the great horned owl, and the eastern 
meadowlark; among the genera, the 
stiff-tailed ducks, avocets, and flickers. 
What can one conclude? Recon- 
structing the avifaunal history is 
clearly not easy. Readers interested 
in cautious scientific interpretations 
can turn to my technical papers; here 
let me indulge in speculation on a 
few of the birds found in the high 
Andes. 
Once upon a time in the late Ter- 
tiary, about two to three million years 
ago, the Andes were not very tall 
mountains. In what is now Colombia 
and Ecuador to the north, their tops 
were either clad with cloud forest or, 
on the highest summits, covered with 
scrub. The high-altitude avifauna 
there included flycatchers, humming- 
birds, tanagers, cotingas, and antbirds. 
Farther south, in Bolivia, plateaus 
were covered with savannas and wood- 
lands, where caracaras, finches, and 
blackbirds lived. 
As orogeny proceeded and the 
mountains increased in altitude, local 
timberlines developed, and above 
them appeared a cold, treeless habitat 
dominated by grasslands and moor- 
lands. The formation of this habitat, 
the precursor of today’s paramo and 
puna , was the turning point in the 
faunal history of the high Andes. At 
first, few birds other than wide-rang- 
ing species that normally inhabit open 
spaces — ducks, coots, ibises, and wad- 
ers in moist areas and lagoons, hawks 
in dry places and cliffs — could live 
in these habitats. 
Some of these birds are good fliers 
and could have come from almost any- 
where: from lower down the moun- 
tainsides, provided the slopes were not 
forested; from southern South Amer- 
ica; and even from North America. 
Today’s paramo and puna are inhab- 
ited both by relicts of these early col- 
onizations and by wide-ranging species 
that have invaded the high Andes 
more recently. Single stray birds from 
the lowlands, such as the wattled 
jacana ( Jacana jacana) I saw at 
10,000 feet in northwestern Argentina, 
are unlikely candidates for coloniza- 
tion. But many birds travel in flocks 
and can thus potentially stay and 
breed in areas far from their normal 
range. 
Seed-eating finches and doves in- 
vaded the high Andes above timber- 
line soon after seed-rich grassland 
habitats opened up. Little adaptive 
modification was required of these 
birds. Siskins — finches whose flocking 
behavior and wandering tendencies 
make them excellent colonists — could 
have come from almost anywhere. 
Since they eat buds and other veg- 
55 
