In long-lived birds, monogamy is thought to enable breeders to retain the s
ame mate from year to year, but exceptions occur For example, King Penguins
(Aptenodytes patagonicus) and Emperor Penguins (A, forsteri) have much low
er mate fidelity than do smaller species of penguins, despite their high ra
tes of survival. Recently Olsson (1998) suggested that divorce in King Peng
uins could be adaptive. Although Olsson was the first to propose an adaptiv
e function for divorce in this species, he was unable to assess the relatio
nships among individual quality, date of arrival, mate choice, and breeding
success. Accordingly, we studied King Penguins and Emperor Penguins to fur
ther examine the determinants and consequences of divorce. Mate retention w
as not affected by breeding performance in the previous year or by experien
ce, and neither mate retention nor divorce had significant consequences on
chick production the following lear. King Penguins were more likely to divo
rce as arrival asynchrony of previous partners increased. In Emperor Pengui
ns, this tendency to divorce occurred only when females returned earlier th
an their previous mates, Most Emperor Penguin pairs formed within 24 hours
after the arrival of malts, which were outnumbered by females. King Penguin
s that had nested successfully in their next to last attempt were favored a
s mates for ones that had been unsuccessful, and individuals of both specie
s probably chose the best mates available. Most of our results for King Pen
guins and Emperor Penguins supported Olsson's (1998) conclusions in that di
vorce appears to be adaptive, Mate retention in the absence of a true nesti
ng site (neither species builds a nest) would be maladaptive for these spec
ies, which face strong time constraints for breeding. Therefore, divorce oc
curs because costs of mate retention are high. Aptenodytes penguins appear
to have adopted an optimal divorce strategy in order to adapt to their long
breeding cycle in a demanding environment.