Anim, Behav.

, 1995, 49, 855-856

Reciprocal altruism between male vampire bats, Desmohs votundus

LISA K. DENAULT & DONALD A. McFARLANE
Joint Science Department, The Claremont Colleges, 925 N. Mills Avenue, Claremont. C,4 91711,
U.S.A.

(Received 31 May 1994: initiui acceptance 6 riugust 1994;

jkal acceptance 21 Novet?lher 1994: MS. wnlber: as-l I II)

Reciprocal altruism is an example of social behav- counter-altruistic aggressive behaviour betw-een

iour that has generated much interest among potential reciprocators). Males are more transient
evolutionary theorists, but relatively few well- members of vampire colonies than are females
documented case studies. Among mammals, (Wilkinson 1988), and strong dominance hier-
reciprocal altruism has been reported for the archies have been reported among adult males
dwarf mongoose, Helogale par&a (Rood 1983), (Wilkinson 1984; Park 1991). These behaviour
naked mole rats, Heterocephalus glaber (Jarvis patterns should undermine the stability of
1978), impala, Aepyceros melampus (Hart & Hart reciprocal altruism between male D. rotundus.
1992) and a few other species, but the best known We conducted a 7-month behavioural study of
and most intensively studied example is the regur- a captive colony of vampire bats and documented
gitation of blood-meals by successfully foraging that reciprocal altruism between adult males of
adult vampire bats to unsuccessful individuals. low relatedness did occur, was relatively common,
This behaviour has apparently evolved in and was associated with a very low-intensity,
response to the finely balanced energy budget of non-linear dominance relationship between males.
vampires, which can result in starvation following The experimental group consisted of two male
as little as 48-72 h of food deprivation (McNab and four female adult vampire bats of four inde-
1973). A foraging situation that results in an pendent maternal lines. We calculated relatedness
average of -8% of adults failing to feed success- between members of the colony from genealogies
fully on any given night (Wilkinson 1984), com- (Krebs & Davies 1978) using data extending back
bined with the low fecundity of vampire bats, three generations. We determined maternal con-
makes food sharing an essential element of tributions directly, and determined paternal con-
survival. tributions probabilistically using the number of
The evolutionary stability of reciprocal altruism available potential sires at the times of conception.
was studied by Trivers (197 1), who concluded that Relatedness between the two males (32L and 35R)
the behaviour depends on several critical factors. was 0.26 (x41 s~=O.l5 & 0.06 for the whole
Particularly important are the ability of altruists colony). The males were 51 and 42 months of age,
to detect and refuse cheaters, and a high prob- respectively, at the conclusion of the study. Bats
ability of reciprocation in the future. Both criteria were housed in a 0.4-m3 plywood roost box
require frequent and intimate social interactions designed to simulate a hollow tree and provide a
with a high degree of individual recognition. variety of roosting options in all three dimensions.
Although reciprocal altruism between adult vam- We subjected the colony to experimental manipu-
pires has been well studied, only female-female lations (unpublished data) intended to promote
and occasional male-female regurgitations have blood-sharing behaviour and test the extent of
been reported (Wilkinson 1988). Furthermore, reciprocity and effects of ‘cheating’ (i.e. non-
reciprocal altruism between adult males has reciprocation).
not been predicted, since several studies (e.g. We observed blood-sharing regurgitations
Joermann 1984; Wilkinson 1985) indicate that between males on three occasions, twice from 35R
males violate two of Trivers’ (1971) five criteria to 32L and once from 32L to 35R. 35R never
for the evolution and maintenance of reciprocal regurgitated to the females in the colony. even
altruism: low dispersal rates (which maximize the when solicited, and 32L regurgitated only once to
opportunities for donors to receive reciprocations) a female (47R; relatedness=O.ll). The total data
and weak dominance hierarchies (which minimize set (N= 10 regurgitations; 15 1 solicitations)

0003-3472/95/030855+02. $08.00/O c 1995 The Association for the Stud) of Animal Behaviour
855
856 Animal Behaviour, 49, 3

revealed no significant correlation between relat- male dominance hierarchies may need to be
edness and regurgitation for the colony (r=0.288, re-examined.
Pr0.05). Association time between the two males, This research was supported by the Joint
defined as the total time a pair of bats were ScienceDepartment of the Claremont Colleges ir,
observed in close proximity (less than one body partial fulfilment of the undergraduate research
length) as a percentage of the total observation thesis requirement for L.K.D.
time, was high: 28% of 32L’s total association
time was with 35R, compared with as little as 1% REFERENCES
for some female pairs. Mutual social grooming
between males, not previously reported, was rela- Hart, B. L. & Hart, L. A. 1992. Reciprocal allogrooming
in impala, Aepyceros melampus. Anim. Behav., 44,
tively common. Also, male-female association 1073-1083.
and non-aggressive social interactions were much Jarvis, J. U. M. 1978. Energetics of survival in
higher than previously reported. We observed Heterocephalus glaber (Ruppell), the naked mole rat
aggressive interactions between the two males on (Rodentia: Bathyergidae). Bull. Carn. Mus. nut. Hist.,
only four occasions, resulting in two ‘wins’ 6, 81-87.
Joermann, G. 1984. Echoortung bei Vampir.
(defined as an unopposed or unreturned aggres- fledermausen (Desmodus rotundus) in Freiland. 2.
sive act) each. Aggressive interactions between Siiugetierk.. 49, 221-226.
male-female pairs were much more common. Krebs, J. R. & Davies, N. B. 1978. Behavioural Ecology:
Blood regurgitation from adult females to adult An Evolutionary Approach. Oxford: Blackwell
Scientific Publications.
males, also previously unreported, accounted for
McNab, B. K. 1973. Energetics and the distribution of
three of the 10 observed events, with one female vampires. J. Mammal.. 54, 131-144.
(41R) regurgitating to both males. We interpret Park, S. 1991. Development of social structure in a
these results as indicating that the dominance captive colony of the common vampire bat, Desmodur
relationships between both male and female rotundus. Ethology, 89, 335-341.
Rood, J. P. 1983. The social system of the dwarf
colony members were weak. mongoose. In: Advances in Mammalian Behavior (Ed,
We conclude that social interactions between by J. P. Eisenberg & D. G. Kleiman). Spec. Pubis Am.
captive, socially acclimatized adult male vampire Sot. Mammal., 7, 454458.
bats in non-crowded conditions are much more Trivers, R. L. 1971. The evolution of reciprocal altruism.
Q. Rev. Biol., 46, 35-37.
frequent and varied than previously reported or
Wilkinson, G. S. 1984. Reciprocal food sharing in the
supposed, and that adult males participate in vampire bat. Nature, Lond., 308, 181-184.
reciprocal altruistic food sharing. Indeed, the Wilkinson. G. S. 1985. The social organization of the
energetic constraints imposed by foraging failure common vampire bat: I. Pattern and cause of associ-
that favour reciprocal altruism among females ation. Behav. Ecol. Sociobiol.. 17, 111-121.
Wilkinson, G. S. 1988. Social organization and behav-
also apply to males. The prevailing paradigm in ior. In: Natural History of Vampire Bats (Ed. by A. M.
which vampire bat society is dominated by strong Greenhall & U. Schmidt), pp. 85-97. Boca Raton,
resource defence polygyny and strong male- Florida: CRC Press.