stick or twist? | 135between one breeding attempt and the next. In a nutshell, the alba-
trosses occupy those spare months with one of four travel patterns; stay
in local waters in the Southern Ocean, stay locally but occasionally ven-
ture downwind towards Australia, migrate as far east as New Zealand,
or embark on a grand circumpolar tour. Amazingly the Crozet birds
mostly adopt the stay- near- home option while the Kerguelen birds are
the Wanderers that truly wander on circumpolar journeys. Incidentally,
since the birds on the two archipelagos are known to be genetically
alike, this argues against consistent patterns being the result of genetic
inheritance. And these travel patterns are consistent. Once a bird from
either archipelago has reached adulthood, it will persist with the same
migratory habit, sedentary or migratory, for many years.^5
Migratory consistency extends to the Atlantic and doubtless beyond.
Few visitors to the subtropical jewel that is Madeira will take much note
of the Ilhas Desertas, a barren string of islets to the east of the principal
island. One of the islets, Bugio, is home to a very rare gadfly petrel, the
Desertas Petrel, with a world population of just 200 breeding pairs.
When not breeding in winter, each petrel heads to one, and only one, of
five distinct areas across the ocean. Two of the areas are north of the
Equator, in the Gulf Stream off the south- eastern United States, and
around the Cape Verdes. Three are south of the Equator, off northern
and southern Brazil, and in the central South Atlantic. When 26 birds
were tracked over two or three winter periods, every one remained faith-
ful to its chosen ocean region.^6
Not only is there evidence of fidelity to migratory routes, there is also
evidence that the timing of migration can be characteristic of the indi-
vidual. For example, the Black- browed Albatrosses nesting on tussac-
clad Bird Island, South Georgia, have been a source of inspiring infor-
mation. When not facing breeding responsibilities, most individuals
head eastward to the Benguela Current off South Africa. Their sojourn
there is characterised by remarkable individual consistency in timing.
For example, at the end of the winter, the dates at which individuals
leave south-west Africa, arrive in the south-west Atlantic, and arrive near
to Bird Island are strongly correlated from one year to the next. One
bird bucked the geographical trend, wintering off Australia instead of
Africa, but it did not buck the timing trend. Its dates of departure from