UC Irvine
UC Irvine Previously Published Works
Title
Female-Female Pairing in Western Gulls (Larus occidentalis) in Southern California.
Permalink
https://escholarship.org/uc/item/85c827zr
Journal
Science (New York, N.Y.), 196(4297)
ISSN
0036-8075
Authors
Hunt, GL
Hunt, MW
Publication Date
1977-06-01
DOI
10.1126/science.196.4297.1466
Copyright Information
This work is made available under the terms of a Creative Commons Attribution License,
availalbe at https://creativecommons.org/licenses/by/4.0/
Peer reviewed
eScholarship.org Powered by the California Digital Library
University of California
Female-Female Pairing in Western Gulls
(Larus occidentalis) in Southern California
Abstract. Pairs
of
females
that remain together from one year to the next are
associated with
the
presence
ofsupemormaf
clutches in western guff nests. Intervals
between laying
of
eggs in supernormal clutches are fess than those
found
in
normal
clutches, a result indicating both females
in
a pair contribute to the
cfwch.
Most
eggs
in
supernormal clutches are infertile. The pairs
of
females
occupy territories
that are not
shared with a resident male.
In
three homosexual pairs
one
of
the fe-
males exhibited behaviors normaffy ascribed only to males.
Supernormal clutches (four to six
eggs)
in
the
nests
of
western gulls (Lar-
us occidentalis)
are
of
frequent (8 to
14
percent
of
1200 pairs) and regular oc-
currence
on
Santa
Barbara
Island, Cali-
fornia
(/,
2).
Other
field studies
(3,
4)
show that supernormal clutches
occur
throughout the colonies
of
western gulls
nesting
in
the
California Channel Is-
lands. A high frequency
of
supernormal
clutches has not been reported for
other
populations
of
this species, and
the
phe-
nomenon is unusual for gulls in general
(5-8).
Our
data
suggest that supernormal
clutches
on
Santa
Barbara
Island
are
al-
most always
produced
by
pairs
of
female
western gulls
that
l
ay
eggs in
the
same
nest and defend a
shared
territory that
has no resident male.
Four
types
of
evi-
dence
support
this conclusion: (i)
the
sexing
of
adu
lt
gulls
captured
during in-
cubation, (ii) the intervals between
the
laying
of
eggs in normal and supernormal
clutches, (iii)
the
fertility
of
eggs
in
nor-
mal
and
supernormal clutches, and (iv)
observations
of
the
reproductive behav-
ior
of
bo
th types
of
pairs. Data presented
here were collected during
the
breeding
seasons
of
1972
through 1976.
On
normal clutches (one
to
three
eggs),
we
trapped 10 male and
15
female
incubating adult gulls, and
one
male and
74
females were taken on supernormal
clutches (9).
On
21
clutches containing
one
to
three
eggs,
we
caught
two
male-
female pairs. eight single males and
11
single females;
one
additional
three
-egg
clutch was incubated by two females. In
contrast.
on each
of
23
supernormal
clutches. we captured two incubating fe-
males, and
on
27 supernormal clutches,
only a single female. In
one
instance a
male-female pair was trapped incubating
a clutch
of
four fertile eggs.
Th
e
sex
of
the birds was determined by dissection
(1974)
or
laparotomy (1975 and 1976).
The
females caught incubating the
same clutch apparently form stable
pairs.
Seven
of
eight pairs
of
femal
es
lap-
arotomized
and
banded in
1975
remained
paired
on
the
same
territories in
19
76.
Members
of
homosexual pairs exhibited
al
l
courtship
and territorial behaviors
used
by
heterosexually
mated
birds, ex-
cept
that
in female-female pairs, court-
ship-feeding, mounting, and
copu
lation
were usually
absent.
In
three
female-
fe-
male pairs,
we
observed
one
member
of
the pair
exh
ibiting
be
haviors normally
Table I. Intervals between the laying
of
western gull eggs (1973
and
1975)
in
clut
ches
of
different
sizes.
The
number
of
clutches.
number
of
intervals between
consecutive
eggs,
and
the mean
number
of
days
between
consecutive
eggs are indicated.
Inter-
Proportion
of
eggs laid
Clut
ch
Clutches
vals
D~ys
size
(No
.)
(X)
Same
Consecutive
2
days
3
days
(No.)
day
days
apart
apart
3
31
62 2.16 0
0.03 0.68 0.29
4
5
15
1.60
0,07
0.40 0.40 0.13
5 5
20 1.45 0.05
0.50 0.40 0.05
6
I 5
1.00
0
1.0 0 0
Table
2.
Frequency
of
western gull eggs showing development
on
Santa
Barbara Island. Numer-
ators
indicate
number
of
developing eggs, denominators the t
ota
l examined in
each
category.
Clutch
Total
number
of
Developing eggs
size
Clutches Eggs
1973
1974 1975
Total(%)
3
42
103
18/21
40/54 26/28 81.5
4 27
95
0/4
5120
7171
12.6
5
33
128
1/9 2/
10
16
/
109
14.8
6
5
16
0/1
0115
0
1466
restricted to males, such
as
mounting
and attempted copulation
(10). In a few
instances females in these pairs regurgi-
tated food
in
response to repeated head-
tossing
by
the
partner in a
manner
some-
what simil
ar
to courtship-feeding by
males
(I
I).
However, we
never
observed
the females offering large
amounts
of
food
as
males
do
in
normal
cour
tship-
feeding
(10).
The
close
interval between
the
laying
of
eggs in supernormal clutches (Table I)
(12)
indicates that two females
were
con-
tributing
to
these clutches. In most Jarid
species,
the
interval between the laying
of
the first and second egg and between
the
second
and third egg is simil
ar
and,
for the l
arger
species, ranges from 2 to
2.25 days (6,
13).
These
intervals
are
similar to
those
found
in
three-egg
clutches
in
this study. In
the
super-
normal clutches, intervals
between
the
laying
of
eggs were significantly
shorter
(t
= 7.69, P < .001) (Table I), and in six
instances. two eggs were laid on
the
sa
me
day.
The
infertility and failure
of
eggs from
supernorma
l clutches to develop (Table
2)
(14) also suggests female-female pair-
ing.
If
males
were
usually associated
with female-female pairs,
one
would
ex
-
pect a higher incidence
of
fertility.
The
presence
of
a small percentage of
fertile eggs
in
the nests
of
female-female
pairs
can
be
exp
lained by promiscuous
matings. During observations
of
color-
marked birds in
1975
,
we
noted
13
in-
stances
in which known male gulls copu-
lated with
or
attempted
to
copulate with
other
gulls not their
mates
(10).
To
our
knowledge this is
the
first reported in-
stance
of
promiscuous matings
in
gulls in
which the females sometimes accepted
the males and copulation was successful
(15).
There
arc
more female-female pairs
of
western gulls
on
Santa
Barbara Island
than
are
represented
by
the
number
of
supernormal clutches. We trapped
one
homosexual pair
in
1976 incubating a
clutch
of
three
infertile eggs.
Of
39 and
19 three-egg clutches examined
in
1972
and
1973
respectively, 17.9 and 21.0 per-
cent
contained
no
embryos. During ob-
servations
in
1972
we
were unable
to
dif-
ferentiate
on
the basis
of
size
between
members
of
pairs with three-egg clutches
that failed to hatch. Measurements
of
birds
sexed
in s
ub
sequen
t
years
con-
firmed that this population is
sexua
ll
y di-
morphic
(10). Had these pairs been het-
erosexual, differences in size would have
been apparent.
During hundreds
of
hours
of
observa-
tion
al
all phases
of
the
breeding cycle
SCIENCE.
VOL.
196
during five
seaso
ns (
1972
through
1976).
we
never observed three gulls defending
a territory together
or
attending the
same
ne
st. The pre
se
n
ce
of
two
females shar-
ing the
sa
me
territory cannot
be
con-
strued
as
polygamy, regardle
ss
of
wheth-
er
or
not the
egg
s produ
ced
by t
hose
fe-
mal
es
were fertilized by promiscuous
mating with a male. The finding
of
one
male-female pair incubating four fertile
eggs
docs n
ot
invalidate the conclusion
that
the
majority
of
supernormal
clutches in the colony arc the result
of
female-female pairing.
Homo
se
xual pairing
ha
s not. to our
knowledge,
been
reported for any group
of
wild birds (16). With the exception
of
our study
it
is unknown in gull
s.
a taxon
that h
as
been expo
se
d to widespread
and
thorough scrutiny. Its existence in west-
ern gulls is apparently recent. Before
1968.
only
two
clutches greater than the
usual m
ax
imum
of
three
eggs
were re-
ported
for
western gulls
ne
sting in the
Channel Islan
ds
off
California and Mexi-
co
(17). The first records
of
substantial
percent
ages
of
~
upernorm
al
clu
tc
he
s in
this
~
pe
c
ie
s
arc from 1968 [11.3 percent
of
150 clutches examined.
San
Nico
l
as
Isl
and
(3)] and
1972
[ 1
1.0
percent
of
63
clutche examined. Santa Barbara Island
(1)).
Supernormal clutches have
bee
n
re
-
ported in colon
ies
of
ri
ng
-bill
ed
gulls (L.
delewarensi
s)
as
early
as
1942
(7). Ver-
me
er
(8) reported that in one super-
normal clutch, two
eggs
were laid on
the
s
ame
day,
but
no data were pre
se
n
ted
in
any
of
these studies on either the fertility
or
the origin
of
the large clutches in this
s
pe
cie
s.
At present we do not know whether
fe
-
male-female pairing in western
gu
lls is
pathological
or
if
it
ha
s adaptive value.
Promiscuous m
at
ings allow some homo-
sexually mated females to produce fertile
egg
s.
Without being paired, they would
not
be
able to incubate the
se
eggs
or
rai
se
chicks (18).
If
there were an excess
of
females in the population (yet
lo
be
determined), then homosexual pairing
would raise from zero the probability
that the excess females would raise
offspring.
Although female-female pa
ir
s may
produce fertile
eggs
fr
om
promiscuous
mating
s,
these pa
ir
s may s
till
l
ac
k other
contributions to chick production pro-
vided
by
the male. One difference be-
tween pair types is the lack in female-fe-
male
pa
irs
of
net energy input to females
by males
thr
ough courtship-feeding.
Eggs laid by the stable homosexual pa
ir
s
are
sma
ller
than those laid by hetero-
sexually paired femal
es
(19). Chicks
24 JU
NE
1977
from small
eggs
have lower
po
sthatching
survival
(20). This study may provide a
unique opportunity to
te
st the adaptive
value
of
energy provided
by
the males to
the
femal
es
for
egg
pr
oduction.
GEORGE
L.
HUN
'I
',
JR.
M
OLLY
WARN
ER H
UNT
Departme
nt
of
Ecology a
nd
Evolutionary Biology. University
of
California. Irvine
92717
Rer
er
ences and Notes
I.
G.
L.
Hunt
and M. W. Hunt, Condor 15. 483
(1973).
2.
In 1973. 7.7 percent
of
104
clutc
he
s: in 1974.
13.8
percent o
f6
5 clutches:
in
1975, 10.3 percent
of
126
clutc
he
s: and in 1976,
8.6
per
ce
nt of
162
clutches on Santa Barbara I
sla
nd
co
ntain
ed
more than three eggs [G.
L.
Hunt
and
M.
W.
Hunt. unpublis
hed
field notesJ.
3. R.
W.
Sc
hreiber, Condor 12.
133
(1970).
4. F.
Gress.
unpublished field
notes;
G. Hunt. un-
publish
ed
field
note
s.
5.
Clutch
,izc
in large
Ltmts
gulls usually ranges
between
two
a
nd
th
ree
eggs
per
clutch,
and
clutches of four eggs
arc
very rare I K. Paludan,
Vidensk. Medd. Dan.
N11111rhis1
Foren. Khoben-
havn
11
4. I (1951): R. A. For
dham
.Notornis
1
1.
3
(1964);
J. Keith,
J.
Appl. Ecol. 3 (Suppl.),
(1966); H. Klo
mp
, Ardea 58. I (
19
70); G. L.
Hunt
and
M. W.
Hunt.
unpub
li
shed
observa-
tions:
(I.
6)1
.
6. K.
Verm
eer.
Occas. Pap. B. C. Prov. Mus. 13. I
(1963).
7. Supernormal
clu
tches
of
four
to
six eggs have
occurred
more
frequently in ring-billed gull
(lams
delell'aren.sis)
colo
ni
es
than
in
other
large
larus
gulls [J. Moffitl. Condor 44.
10
5 (1942): D.
W.
Joh
nston and M. E. Fos
ter
, ibid. 56. 38
(
1954)1
.
8. K. V
ermeer.
Can. Wild/. Serv. Rep.
12
(1970).
9. In
cubating
adult gulls were
trapped
with eith
er
walk-in t
raps
placed
over
their
ne
sts
[D. K.
We
aver
and
J.
A. Kadl
ec,
Bird-Banding
41
, 28
(
1970)
)
or
monofilament snares placed around
the
edge
of the n
est.
which
ca
ught
the
birds
around
the feet when the line
wa
s pulled.
10.
Details
of
these
behavior pa
tl
erns
will
be
pub-
lished
elsew
here.
11.
R.
G.
B. Brown. Behaviour 29.
122
(1967).
12.
We
determined
laying
patterns
by
m
ar
king
nests
with
numb
ered wooden stakes
and
visiting them
daily
(
197
3 and
197
5)
or
every oth
er
day ( 1975)
a1
approximately the
sa
me ho
ur
in
order
to
re-
cor
d n
est
conten
ts
and
individually
mark
newly
laid eggs.
13.
lnt
ervab
range
from 48.7 to 54.8 h
ours
(J.
Bur
-
gcr.
Anim.
Belwv. 22. 521 (1974
)1:
R.
Drcnt
. Be-
lwv
.
Suppl.
1, I (1967).
14.
In 1973
and
1974.
we
examined inc
uba
t
ed
eggs
for
evidence
of
development. In 1975.
we
distin-
guished fertile from nonfcrtilc eggs on lhe
day
they
were
laid
by
exa
mining
the
blastodi
sc
(B.
M
arquez
an
d K. Ogasawara. Poult. Sci.
SJ
, 1607
(1974)).
15.
M. Mac Roberts [Z.
Ti
erp
syclw
l. 32. 62 (
1973)1
reviewed t
he
literature on
"extrnmaiitar·
com1-
ing in gulls
and
presented detailed information
on
thi5
behavior
for lesser black-backed
(Lams
fi
1
sc11s)
and herring gulls (LllrttS
arge111a111s).
In
all
instances. M
ac
R
oberts
reported
that male
gulls
attempted to rape femal
es
and that
the
f
e-
males rejected these advances. No succe
ss
ful
co
pulations
were
see
n.
Vermeer (
6)
reported
in-
sta
nces
of
rnpe in glaucous-winged gulls (
Lam
.v
gttwcesce
n
s).
which
apparently
resulted in suc-
cessful
co
pulation
s.
J.
Burger
and
C.
G. Be
er
JBehaviour
SS
.
301
(1975))
reported
rape
of
in-
cuba
ting female laughing gulls (Lllrus lllricilla)
by intruding mal
es.
16
. Male-
male
co
urt
ship
bu
t n
ot
pa
i
iing
ha
s
been
rc-
po
11cd
in
wi
ld ostriches
(S
trtuhill came/us
m1s-
1
m/is)
JG.
Sauer.Auk
89, 7
17
(1972)1.
Homo
sex-
ual pairings of birds
in
ca
ptivity h
ave
been
re-
ported
by
N. Collias and
L.
John
(A11k
76. 478
(1959)1.
W. Dilger (Z. Tierpsychol.
11
. 649
(1960)].
and
D. Jefferies (Ibis 109. 266 (1%7)].
17. L. Kiff, unpublished
da
ta
on
museum
co
ll
ec-
tions
of
eggs
taken
from
the
Channel Isl
ands
of
California and Mexico (337 clutch
es.
1891
to
1969).
18.
In 1975.
we
compare
d chick survival
an
d growth
rates
of 37
young
produced by 20 normal pairs
with
41
foster young given
as
pipping eggs to 22
pairs
tha
t
had
laid supernormal clu tches. Si
nce
growth
rate
s and fledging su
ccess
in
the
two
groups
were
nearly identical (10). female-female
pairs
ca
n
be
consid
ere
d
capable
parents
under
the
conditions extant at
Santa
Barbara
Island.
19.
G.
L.
Hunt, M. W.
Hunt
. R. W. Risebrough,
in
preparation.
20. J .
Parsons.
N1
1111r
e
(Lo
ndon) 228,
1221
(1970).
21. We
thank
the
per
so
nnel
of
the Channel Islands
National
Mo
num
ent for
their
invaluable help
and
coopera
ti
on.
C.
Green
(1973);
S.
Anthony.
Z.
Ep
pley, P. Ewald.
L.
Holmgren. D.
Schwartz
(1975); A. Brand. D. Kna
pp
,
G.
Kunz
. M.
Naughton, F. Picrotli, and K. Winn
etl
(1976)
have
all helped with the fieldwork. Dr. J. Os-
borne instructed
M.W.H.
in
laparotomy
tech
-
niques. Financial support
wa
s
pr
ovided
by
two
grants
to
G.L.H
. from
the
Frank M.
Cha
pm
an
Memorial
Fund
of
the
American Museum
of
Natural Hist
ory
an
d
by
the
Sc
hool of Biological
Scien
ces.
Univ
ers
ity of California. Irvine. D.
O'Dowd
and
Or
s.
D. An
ders
on, R. Schreiber.
R. Ri
sc
brough, and K. Verme
er
provided help-
ful
co
mments on various versions of this manu-
script.
16
November
1976; revised 4 February 1977
Somatostatio: Analogs with Selected Biological Activities
Abstract. [o-Cys
11
}-So
mato
stMin is the first analog
of
somatostati11
found
to
be
more
potent
in inhibiting glucagon
and
gr
ow
th h
or
mon
e secretion than it is in inhib-
iting insulin secretion. [
r>-TrpRj-Soma1ostati11
is eig
ht
to t
e11
tim
es
more
potent
tha11
so
mato
statin in inhibiting insulin , gluca
go
n,
and
growth hormone
sec
retion. (D-
Trp
8,
o-Cys
11
}-Somatostatin is
mor
e potent than [ o-Cys
14
}-so111atostati11,
but
retains
its relative selectivity
i11
being a
mor
e
potent
inhibitor
of
the secretion
of
glu
cago
n
and
growth h
ormone
than
of
insulin.
Somatostatin is now recognized
as
a
supp
res
so
r
of
the secretion
of
various pi-
tuitary. pancreatic, and gastrointestinal
hormone
s(
/ ). In addition,
so
mat
os
tatin
h
as
been sh
ow
n to influence an array
of
neurochemical, neurophysical. pharma-
cological,
and
be
havioral parameters
(I,
2).
The plethora
of
actions of
so
mat
os
tatin
and the demonstrations
of
its anatomic
dist1
ibut
ion throughout the
cen
tral ner-
vous system
(J.
4).
gastrointestinal tract
(4, 5), and
pa
ncreas (4, 6,
7)
have sug-
gested that
so
matostatin might play sev-
eral physiological roles
as
a local extra-
cellular
me
sse
n
ger(/)
.
Somatostatin when given continuously
intr
avenously (8)
or
subcutaneously (9)
decreases the
gl
ucose intolerance
of
hu
-
m
an
juvenile diabetes mellitu
s.
Thi
s ac-
tion is
at
least partially secondary to the
loweri
ng
of
the
ab
normal eleva
ti
on
of
1467