SOOTY TERN BEHAVIOR
JAMES JAY DINSMORE
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
AC K.N CII ?r=ni -T'II T -
I have benefited greatly from the counsel and interest of 'r.iny
individuals. The advice and encouragement of Dr. Oliver L. Austin, Jr.
have aided me throughout this work, both in the field and in
preparing a written report. Dr. William B. Robercson, Jr. continually
provided me with information on Sooty Terns and also arranged for my
stay on the Dry Tortugas. The comments and criticisms of Drs. George
W. Cornwell, Frank G. Nordlie, Thcnas J. Walker, and S. David Webb
bhave pointed out many weaknesses in my work and thereby have been
of great help to ne. I also have profited from discussions on Sooty
Terns with Brian Harrington, Robert W. M'Farl.ne, John C. O:ren,
Dr. Glenn E. J.oolfenden, and others, especiall-y many nenoers of
bandding parties to the Tortugas. Personnel at Fort Jefferscn
National Monument and Everglades National Park were especially
helpful in providing a place for me to live at Fort Jefferson and
then making that living enjoyable. Dr. Gary D. Schnell generously
gave me access to his unpublished data on Sooty Terin flight speeds.
The Florida State Museur provided a boat to use at Fort Jefferson.
T;o grants front the Frank M. Chapman Fund, Amrican Mluseum of naturall
History, and a Louis Agassiz Fuertes Research grant from the W;ilson
Ornithological Society helped finance the field work. A University
of Florida Graduate School Fellowship made it possible for me to spend
an extended period at the Dry Tortugas. y wi"f,?a",encouraged "m
throughout the study .nd typed the final drafts of this disSrLatiorn.
To all these individuals and organizations, an grateful.
TABLE OF CCITE:ITS
ACKNLEDG TS. ........
IST CF TABL-S3 . . . . . . . . . * *
LIST OF FIGURES. . . . . . . . * *. *
ABSTRACT . . . . . . * * *
INTRODUCTION . . . . . . . * * * i
THE D?.Y TORTUGAS . . . . . . * 2
Physical Features . . . . . ***** -
Fauna . . . . . . . 5
Clinate . . . . . . .
Pr-evious ork . .. . . . * *
THOD3 . . . . . . . . .
BREDIG ACTITES .. ........... 9
Formation of the Colony . . . . . ...* * ...
Aerial kin . . . . . . * * 10
First Lancind. . . . . ct* * t *
GOAerial o i P lc . . . .. . .. ...* ..
Establi sing Territories . . . . . . .
Incubation Period . . . . . . ............ 25
Egg-laying. .. . . . . . . . . 2 2-
Behavior Lurin^g Incubation . . ... .. .. 27
Attentiveness. . . . . . . .2
aDipping. . .. . . ......... 35
Temperature Regulation . . . . . .. 7
Nest Relief. . . . . . . . ...... 38
Length of incubation Shifts. ....... .. .. 41
Influence of Weather . . . . . . . 42
Fledging Period . . . . . .
Hatching . . . . . . . . * 43
Attentiveness. . . . . . * * * *45
Feeding. . . . . . . . * * * *
Frequency of Feeding .. . . . . . . 52
Individual Recognition . . . . . * 5
Chick Behavior . . . . . . . . . -5
Other Aerial Activities . . . . . . ... 61
Thermal Scaring . . . . . . . . 61
Fly.-ups. . .. . ... . . .... .62
Panics . . . . . . . . . . . 63
ECOLOGY . . . . . . . . ... . . .... .65
Interactions with Other Species . . . . .. 65
Enemies. . . . . . . . . ... . 65
Other Terns. . . . . . . .. . . 69
Nesting Habitat . . . . . . . ... .. 71
Site Tenacity . . . . . . . . ... . 75
DISCUSSION . . . . . . . . ... . . . 77
Behavioral Conparisons. . . . . . . ... .77
Sooties as a Pelagic Species. . . . . .... .0
LITERATUREP CITED . . . . . . . .... ..... .90
BIOGRAPHICAL SKETCH. . . . . . . . ... . 96
LIST OF TABLES
1 Activities other than incubating or shading by Sooty
Terns caring for an egg 31
2 Summary of 113 dipping flights by Sooty Terns 34
3 Time of occurrence of flights away from the egg by
adult Sooty Terns 36
4 Time of nest relief during incubation 39
5 Time of feeding of young Sooty Terns 50
6 Rate of foraging by adult Sooty Terns 53
7 Nest density and success in five 5-yard-square plots
on Bush Key in 1968 72
8 Some information on the breeding biology of several
species of terns 83
LIST OF FIGURES
1 M:ap of Dry Torugas islands 3
2 Posture of Sooty Tern in the parade display 19
3 Activities by adult Sooty Terns during incubation 29
4 Activities by adult Sooty Terns during the first
2 weeks of caring for the chick 47
Abstract of Dissertation Presented to the
Graduate Council of the University of Florida in Partial Fulfillment
of the Requirements for the Degree of doctor of Philosorhy
SOOTY TERN B3EAVIOR
Jares Jay insmrore
Chairman: Dr. Oliver L. Austin, Jr.
Major Departmentt: Zoology
A 5-year study of the breeding behavior of Socty Terns (Sterna
fuscata) was made at Bush Key, Dry Tortugas in the southeastern Gulf
of Hexyico. The results are compared wi;th the behavior of other terns
and the differences discussed, particularly in regard to the pelagic
environment the Sooty Tern inhabits.
Sooty Terns arrive at the Dry Tortugas some 2 months before eggs
are first laid. At first they circle Bush Key at night only, but
eventually land and extend the tine they spend on the ground, arriving
earlier in the evening and leaving later in the morning. After the
first eggs are laid, flocking subsides. Aerial display by Sooty Terns
consists mainly of the high_ flight in which two birds ascend, usually
by jerk-flying, circle, and then descend together in a coordinated
glide. Scoty Terns may have a poorly developed low flight display.
On the ground the major display is the parade, similar to that of other
Sterna terns. Courtship feeding is rare. Terns often interfere with
pairs attempting to copulate.
The single egg is incubated about 29.5 days with the maLe caring
for it somewhat more than the female. The birds are attentive over
95 per cent of the tine, incubating when it is cool, shading when it
is hot, and engaging in other activities such as preening, fighting, and
loafing most commonly when they change between incubating and shading.
Nest relief usually occurs in the evening after 24 or 48 hours of care.
Dipping, in which adults dip their feet, bill and/or breast feathers in
the ocean, is common during incubation and may cool the egg or provide
it with needed moisture.
Chicks are closely brooded for IL or 5 days, after which adults seem
to recognize them individually and spend progressively less time with
them, By the end of the 3rd week, chicks are alone much of the time
except when fed. Adults feed the chick by regurgitation, spend about
3.5 hours per foraging trip, and often feed a chick several times after
one trip. Males feed the chick somewhat more than females do. Vocal
exchanges between parent and chick seem most important in individual
recognition. Occasionally adults feed a chick other than their own.
Chicks first fly when about 9 weeks old and leave the colony scon
The ground and aerial displays of FSoty Tezns are similar to those
of other Sterna terns, especially the Comurcr Tern. The rarity of the
low flight and courtship feeding, both of which are common in other
terns, may be due to the different way in which Sooties carry food and
the distance they travel to forage.
Sooty Terns have a lower clutch size, longer period of development
of the chick, and first breed when older than most other terns, iiany of
which feed in marshes and coastal waters, These characteristics of
Sooty Tern breeding biology are similar to those o-' many other pelagic
birds. A distant food supply and high adult survivorship apparently
have contributed to these differences from other terns.
INTRO JCT TTCN
Sooty Terns (Sterna fuscata) occur worldwide in tropical aznd
subtropical waters, breeding mainly on low isolated islands (Ashnole,
1963). Some colonies number in the millions (Fidley and Percy, 1956)
and the species could well be one of the most numerous of the world's
birds. This tern ranges cver the open ocean to feed, thus differing
from the navn terns that feed in coastal waters. Sooties apparently
return to land only during the breeding season. Other peculiarities
of Sooty Tern behavior include a ncnannual breeding cycle at some
localities (Chapin, 1954; As ,ole, 1963), and a long migration by
juveniles (Robertson, 1969).
Although the ethology of several species of terns is well known,
no pelagic species of Sterna are adequately studied. The basic studies
of Sooty Tern behavior were made 50 years ago (Watson, 1908; Watson
and Lashley, 1915; Lashley, 1915) and certainly need to be updated.
From 1968 through 1970 I studied the behavior of Socty Terns at the Dry
Tortugas Islands, watching a group of temrs throughout the breeding
season and recording their behavior at all accessible stages. Tkis
paper presents basic information on the behavior of a tropical Sterna
ana attenots to relate t.hi behavior to the species' pelagic habits
and thus show how fun.cat differs frcm its coastal-feeding relatives.
TuB DIRY T1`TGAC-S
The Dry Tortugas presently consist of seven smnal low islands at
approximately 24038'X, 82052'W, about 70 miles west of Fey West,
Florida (Figure 1). in the past century at least four other islands,
once a part of the group, have disappeared (see Robertson, 1964, for
history of all the islands). The islands are located on a large
shallow bank, ruch of it less than 5 fathoms deep, in the scutheastern
Gulf of Me:dico. Here, near the northern lirit of their breeding range,
some 80,000 Socty Terns have nested in recent years. The earliest
definite record of S~oty Terns nesting on the Dry Tortugas is Ohat of
Audubon (1844) 'who found them nesting .ith Brown Noddies (Anou1
stolidus) on Bird Key in 1832. Both species continued to nest at Bird
Key until the early 193C's when it finally washed away. The birds then
moved to r.earby Bush Key, which has been the major breeding site ever
since, although occasionally sone have nested on Garden, Lon;, and
All the islands are l;: sand bars or piles of coral rubble. The
largest, Loggerhead Key (ca. 30 acres), has a small Coast Guard
lighthouse and soaticn and a good stand of large trees, mainly
Casuarina. Garden Key is dominated by Fort Jefferson, a forr.er arrny
outpost and later a prison. Starting with a National Audubon Society
warden stationed on the Tnruges during the 1903 nesting season, the
terns have had sone rotection from egers and other disturbatices
Map of Dry Tortitgas Islands
most subsequent years. Fort Jefferson was designated a. National
Monument in 1935, and its small permanent, human population provides
some protection for the terns in the nesting season. Middle and
Hospital Keys are presently just small bare sand bars. East Key has
some low beach strand vegetation. Long Key is a bar of rough coral
rubble with a small stand of nangroves.
I studied Sooty Terns on Bush Key, currently about 20 acres in
extent and separated from Garden Key by a 500-foot channel. It is
composed primarily of coarse light-colored sand with coral rubble on
the east end where a long narrow spit joins it to Long Key at low tide.
Maximun elevation is about A feet above r-ean high tide, Bush Key
apparently was present in the mid-1800's, disappeared, and reappeared
in the early 1900's (Robertson, 1964). In this century it probably
first appeared as a series of sand bars that eventually connected.
gained some vegetation, and gradually gre: and stabilized.
Three snall brackish pcnds in the center of the island are .rir-.ed
by mangroves (Rhizophora ..-le and L,?un._n-lariz racenosa) and
buttonwood (Cnocarus erert. s). Around then a thicket of bay cedar
(Suriana nariti.-a) 6 to S feet tall covers much of the island. Cutside
the bay cedar thicket and extending from it to the tide line, the
vegetation is relatively low and open. Most Sooty Terns nest on these
flats, although some nest in openings in and under the mangroves or the
bay cedar. The major plants on the periphery of Bush Key are sea
rocket (Cz--le lanceolata), sea oats (Uniola .r.ic~1~?ta), prickly
pear (Ounti s ), a grass (S~orojbohs s. ) and sa. nurslane (S suriir
ortul-aca-strn) .with some sea lavender (Tcurnefcr tia .inchal odes)3
around the edLes. The density of these plants varies frcn year to
year, primarily depending on the amount of rainfall.
Besides Sooty Te:ns, at least three other species of terns have
nested or the Dry Tortugas in recent times. Host n.urmerous today are
the several thousand 3rown I!oddies that build a single platform nest
of sticks and dead vegetation in the bay cedar and other low vegetation
on Bush Key. The Black Noddy (Anous tenuirostris) has been reported on
the Dry Tortugas almost yearly since 1959 (-obertson, 1964) but as yet
has not been found nesting.
Least Terns (Sterna alabifrons) formerly nested on Bush, Long, and
Loggerhead Keys but have not done sc since about 1950, while several.
hundred Roseate Terns (S. dougallii) still nest yearly on the Dry
Tortugas (Robertson, 1964). Royal Terns (Thalasseus maximus) and
Sandwich Terns (2. sandvicensis) nested on the Dry T'rtugas in the
1800's, but other than a single naxi-us egg found in 1952, no rec-n'
nesting records exist (Robertson, 1964).
Of some 2L0 species of birds reported from the Dry Tortugas, the
only others kncr;n to have bred there are so:;e species of booby (S .la)
reported by Audubon (1844) and the Mcurnmin Dove (Zen i-d.ura nmacroi.ra;).
The only nma-mals reported frcn the islands are the introduced rat
(Rattus rattus) tha-t at times has killed young birds and destroyed
eggs (Rssell, 1933) and the WesL Indian Seal (IGonachus trooicalis),
currently faced rith extinction.
Several land reptiles have been collected on the islands (Duellan
and Schw.artz, 195') but apparently nonc hac pernanenily colonized the
islands and they pose no threat to the terns. As the na:e of the
islands suggests, sea turtles formerly nested there in great numbers.
A few, probably Loggerhead Turtles (Caretta caretta), still lay their
eggs en thie Tortugas and occasionally destroy tern nests when they dig
their cun nests.
The climate at the Dry Tortugas is best described as hot and dry.
Daily temperatures often reach the low 90s (OF) and seldom droo below
the low 60s at right (Vaughan, 1915). The intense radiation of heat
from both water and the light-colored sand creates a severe
microclinate in the layer of air just above the ground, the ricro-
habitat in which Scoty Terns nest. Rainfall probabl- averages ss
than 40 inches per year, much of it coming in sudden squalls. Several
hurricanes have struck the Tortugas in recent years. On 6 June 1966
the 123 mph winds of Hurricane Alna drove water over Bush Key and
buried many young terns in the sand. iMany died, but overall mortality
was surprisingly low (Mason and Steffee, 1966). On 3 June 19'6 the
center of Hurricane Abby passed about 60 niles west of the islands and
heavy winds, rain, and waves belted the islands for several days. The
high waves washed away some eggs laid on higher parts of the beaches
but did not swamp Bush Key, and mortality was very low. Adults with
eggs or chicks sat tight on the scrape throughout the storm, and some
eggs hatched during the storm without noticeable detriment.
Tie earliest scientific studies of Sooty Terns on the Dry
Tortugas are those by Watson (1908), Lashley (1915), and Watson and
Lashley (1915). They investigated its breeding behavior and made
some preliminary tests of its honing ability. From 1936 to 1941 some
13,300 terns, mainly juveniles, were banded at the Dry Tortugas.
During the 19401's and 1950's National Park Serv-ice personnel and others
visited the colony almost yearly and made estimates of its size in
most years. S~umaries of much of this work appear in Sprunt (1948)
and Robertson (1964).
In 1959 the National Park Service, the Florida State Museum, and
the Florida Audubon Society began a cooperative banding program on
the Dry Tortugas terns. Robertson (1964) summarizes the history of
the terns on the Dry Tortugas and later (1969) documents the
transatlantic migration by juveniles from this colony. In the 12 years
of banding scme 153,700 juvenile and 81,100 adult Sooty Terns have
been banded on Bush Key.
I lived on Garden Key from 29 March to 10 July 1968 and worked
almost daily on nearby Bush Key. I erected a blind near a snail lot
(25 x 30 feet) f-ro which I cleared nearly all vegetation. Terns
were captured in mist nets and 182 were colcx-mnarked, each with a
unique pattern of three colored leg bands and a Fish and Wildlife
Service (JS) band for individual recognition. The terns gradually
acclir-ated to ry presence in the blind, and their behavior appeared
normal once I was out of sight.
I determined the sex of the terns by noting their position in
copulation, keeping in nind that nale-~ ale nountings may occur. Once
the birds were color-marked, 1 recorded their activities and the roles
of males and females in the care of their eggs and chicks. As Sooty
Terns spend much of their time flying, I had to liiit my studies to
their activities at Bush Key and its in.ediate, vicinity After the
eggs hatched., tethered sore chicks to prevent then from hiding
throughout the day. This altered their behavior, but it w:as the only
way I could keep more than a few chicks in sight after they were about
3 weeks old.
In 1969 and 1970 I spent shorter peria.ds on the island after the
The yearly activities of Sooty Terns at the Dry Tortugas can be
su iarized as follor-s: Starting in late January or February, Sooty
Terns appear near the islands in numbers at night only, circling,
calling, and sometimes landing, but generally are absent during the
day (-obertson, 19,~4). Gradually the terns start arriving at the
Tortugas earlier each evening and also in greater numbers. Finally
in late March or April they remain in daylight and land, display,
copulate, dig a scrape, and lay a single egg that. is incubated 29 to
30 days. The chicks are closely attended for the first week or two
and then are alone much of the tine except when fed. Chicks first. fly
when about 9 weeks old and apparently leave Bush Key scon afte2zrard.
Some juveniles are flying by late June and nmst terns have left the
Tortugas by late August or early Septerber.
Little is known of the activities of Scoty Terns away fro.. the
colony. Out of some 153,700 chicks banded on Bush Key, about S0 have
been recovered away from the colony and provide some inform tio cn
the movement of juveniles. After leaving Bush Key, juveniles drift
west in the Gulf of 'Mexico and zhen south along the eastern coast of
Central America, finally moving east along northern South America and
out across the Atlantic in the vicinity of the equator. They spend
about 2 to 4 years in the Gulf of Guinea off Kest Arica, and gradually
drift back across th At.an.Aic. A fe::; 3-ear--ld birds occur at the
Tortugas lase in the breedini season, but almost cert-inly do not
breed At least a fewr 6-year-cld birds breed but sone Sooties may not
breed until older (Harrington, pers. comm.).
An~ong those banded as adults the recovery rate is louer, with
only about 25 recoveries away -ron the colony out of 81,100 banded
through 1970. Tnese recoveries suggest that the range of most adults
is the Gulf of :4e:'co throughout the year ( Robertson, 1969).
Formaticn of the Colony
As the terns assemble and reform the colony, they are very
skittish and difficult to follow. Their activities can be divided
into periods of aerial flocking and first landing.
Robertson (196L) describes a period of night flecking, starting
some 2 months before aggs are laid and during which large flocks of
terns gather near or over the breeding grounds to cll, circle, and
land, only to leave around dawn. Ashn:ole (1963) describes "night
clubs" of terns that. land on the breeding grounds in groups at nigh-
and leave before delight. Tracks in the sand on Bush Key indicate
that night groups for.-- there too, although I never saw the:i.
When I arrived at the Dry Tortugas on 29 MIarch 1968, Souty Ter.i.
had not yet laid any eggs, but they were flocking in well-developed
pazterns every evening. Few terns were present during the day. By
about 08:00 (all times are EST) most Sooties left the island and all
was quiet e-:ceot for a few Noddies. Occasionally at various times of
day, maall flocks of up to 10 Sooty Terns circled rapidly over Bush
Key, giving lo.n d "'ide.-a-w.ake" calls or sharp 'yip yip" notes as one
bird cnsscd anothe-r. The birds passed over t:he islad once or
sometimes repeatedly, flying low and often in pairs. They flew slcwly
with deep wing beats. Their flight resembled the low (fish) flight
display described for other terns (Palmer, 19t1; Cullen, 1960a), except
that they did not carry a fish in their bill.
In early evening, between 16:30 and 17:00, flocking began. The
terns approached the Dry Tortugas from the northwest, much as RPbertson
(1964) describes and gathered northwest of Bush Key, circling and
calling. As more birds joined the flocks, their calls grew louder and
the circling nore pronounced. Usually they started in a loose flock
of ,50 to 200 birds circling 50 to 100 feet above the water. They flew
with a slow, seemingly exaggerated wing beat, often gliding between
beats and drifting along in a circle 100 to 200 feet across. Eventually
a few dropped low over the water and, assuming a more rapid wing beat,
made a fast circuit over the edge of Bush Key and returned to the flock
over the water. Gradually more circled over the island. A few
started landing on the open sand beach on the north side of Bush Key.
The calling got still louder. Occasionally all the birds on the beach
flew rapidly and noisily back to the flock. There they circled and
again started moving back toward the island. Around 16:30 the number
of birds gathered on the beach increased. Fewer flew when disturbed,
and they returned sooner when disturbed. All this time more birds
continued to join the circling flocks.
Out over the water the flocking pattern changed slowly. At first
all the terns were in one staging or circling flock, from which they
flew to the island. As more birds arrived near Bush Key, more staging
flocks formed in a line strung out to the northwest. Presu:m-ably birds
entered the farthest one first and then noved from f3ock to flock as
they approached the island.
On 5 April at 17:30 I could see three such flocks, one about 1,000
feet north of Bush Key and the other two beyond. Between the flocks
I could see lines of birds flying with slow, exaggerated wing beats
about 50 feet above the water. Beyond the last staging flock a line
of terns, again about 50 feet above the water, was stretched out for
several niles. Perhaps some birds approach Bush Key low-over the
water as they normally do later in the season, but most drifted in
,with the flocks described above. The largest flocks I saw were north
of Bush Key; occasionally smaller flocks formed south of Garden Key.
After sundown more terns landed and gathered in groups. On
10 April, a night with a full noon, most Scoties were sitting quietly
along the beaches of Bush Key at 21:30, although a few still circled
and called from the air.
At sunrise no staging flocks were present, although many birds
circled and called over Bush Key or stayed on the ground. By 08:00
most birds had left the island, flying to the northwest. The exodus
was not so well defined as the approach.
Flocking by Sooties is apparently a significant social activity.
Perhaps its function is to delay breeding by early arrivals at the
colony until some critical nur:ber of terns is present. With adult
Sooties apparently spread throughout the Gulf of 1Mexico in the
nonbreeding season, such flocking might synchronize the reproductive
cycles of individual birds and insure that most Sooties breed at about
the same (most favorable?) time each year.
In 1968 the first daylight landings cf Sooty Terns a;lpeared to be
gradual extensions of ti'e spent on the ground by night groups. No
massive influx of terns with almost Irnn.ediate egg-laying was observed,
although that has been inplied for this colony (Thompson, 1903;
Sprint, 1945). Similarly on Ascension Islaand, after a period during
which Sooties are present only at night, they suddenly start to return
earlier in the evening and stay later in the morning, and egg-laying
soon follows (Ash-ole, 1963).
When I arrived on the Dry Tortugas in late March,l163, the terns
had already started staying later in the day so that some were present
until around 08:00. On 1 April I found three Scoty eggs on the west
end of Bush Key, the first of the 196S season. Within a few days
more were laid near these three, and also several hundred feet. east of
them near my blind. Egg-laying then seemed to spread out fromI these
two foci and by 21 April eggs were present virtually throughout Bush Key.
On the plot I watched from my blind, egg-laying started later
(9 April) and peaked around 15 April. Hence I followed the behavior of
these birds before they laid their eggs. At first they appeared
nervous and were easily frightened. They hovered over open ground,
landed briefly, and then flew in a panic (see p.63) alm.st immediately.
Gradually more birds landed and began to fill the available open
ground, only to fly at the slightest disturbance and then return and
go through the whole process again.
As soon as the birds landed and settled down, at least two postures
became ccnon. One of these is the parade that will be described
later. In the other the bird points its bill down, sometimes almost
burying it in its breast feathers. This posture is ccr.-on in Larids.
Harrison (1565) calls it the "stare-down' and says it seems to
"cut off" or end a low intensity agonistic or conflict situation.
Sooties often give a stare-down almost immediately after landing and
then preen their back and wings. An almost feverish amount of
displaying occurred during these early morning hours in the few days
from the time they first stayed on the island until the female laid
Little displaying occurs in the night clubs and most of the birds
just rest (Ashmole, 1963). I made only one night visit to Bush Key
during this period, and I found that most terns were resting on the
beaches or in open parts of the island with a few in the air. One
evening when eggs were present, I remained in my blind until well
after dark. Even then the birds were highly excitable, panicking
whenever I made the slightest disturbance. To prevent undue
disturbance of the colony, I avoided night work and have little
information on the bird's nocturnal activities.
All of the displays and activities leading up to egg-laying may
occur within a few days after the birds first land in daylight,
Because 1 was marking birds at this time, there were few I could
follow through the entire behavioral sequence. I also found that birds
were still moving around within the colony, since many color- iarked
birds iLmediately left Iny plot upon release and never returned. Hence
the sequence of events has been pieced together from the few pairs
that I could follow throughout this period and also by noting when
various activiti-es were most corConl on my pint.
Early in the nesting season and to a lesser exr.ent later, aerial
displays are a characteristic activity of Scoty Terns. Cullen (1 o60a)
studied aerial activities cf the Arc tic Tern (Sterns. oaradisaea) in
detail and sura-rized much of the information available on other terns.
I follow his t1rminology for acti'.-ities he describes that appear
homologous to those of the Sooty Tern.
Except for tne chases ever the colony prior to egz-lInyin;, which
may be homologous to the low (fish) flight that is cu:non anong Sterna
terns, aerial display by S-otiies seems to consist solely of a high
flight siriilar to that of the Arctic Tern (Cullen, 19.Ca).
In Sooties, the high flight is a slow steady ascent by two or
sometimes three terns to a height of several hundred feet and then a
rapid gliding descent. I noted two means of ascent, one often occurring
after the other. In one the tw-o birds ascended al:.ost vertically in
small circles, flapping their wings rapidly as the lower bird chased
and occasionally tried to grasp the tail of the upper bird. Tnis sees
comparable to the upward flutter, a hostile behavior that Cullen (1960a)
describes. I saw the upward flutter at the start of only about 30
per cent of all aerial displays, but usually it was given only briefly
and I rny have r'issed seeing t in other cases. Fro- the u':.ard
flutter, the birds changed to seemingly exaggerated deecp wing bea-ts
that Cullen calls jer-flying.. In many ins-~ances the first sign of
aerial. display was this type o flight, and it was the initial pattern
in many high flights. Flying, thusly, the birds continued to climb,
though not so steprly as with the upward flutter. Again they stayed
close together as if one w-;as chasing the other. In these chases the
lead bird gave a rattling "ka ka ka" call while the pursuer gave
upward-inflected "wek wek" notes. As they climbed, they flew in long
looping circles above the colony and adjacent water, often reaching a
height of several hundred feet. They sometimes continued to clinb
for 2 or 3 mrinu-es, although more commcnly they clinbed for abcut a
At the peace of the flight they often chased briefly and then
maneuvered close to each other by flapping their -ings in slow, almost
half beats with the wings barely moving to below the level of the body.
Then they began a rapid gliding dive. The angle of descent was
usually fairly shallow: at first and became steeper as they descended.
Several times I noted that the pursuer passed the leading bird and
took the lead at the start of the glide, and this nay be usual, much
as Cullen (1960.) reports. In the dive one bird was above and just
behind the other, the two about a foot apart. Both held their wings
rigid; the lower bird's wings were usually bent slightly at the carpals
and the upper one's wings were held almost straight out from the body.
They held the long outer rectrices scissoredd" together and in those
instances where I could see their bills, they pointed straight forward.
In this position they dove together, their movements coordinated so
that the second bird closely followed the sometimes shifting and
dodging flight of the first. Sometimes they started the glide several
times, leveled off after descending a short way, and then dove again.
At other tines they separated and each flew away alone. In a complete
display they glided together down to just above water level, swooped
up a short ways and then separated, each flying away in a normal
flight. Several times I followed both birds after the glide, but I
never saw then stay together to repeat the high flight.
I saw numerous variations of the high flight. Of 50 high flights
in which i saw all of the display, 18 started with the upward flutter
and 32 with jerk-flying. Thirteen times the birds changed from the
upward flutter to jerk-flying. Thus jerk-flying was part of the -
ascent in 45 of the 50 high flights. The other 5 times the entire
ascent was with upward flutter. In 40 high flights the two birds
circled after this ascent, mainly by jerk-flying. The birds started a
descending glide in 47 of the 50 high flights but continued to glide
down to about sea level in only 28. Thus ascending by jerk-flying
and descending in a glide are the two most conspicuous parts of the
high flight, although they do not occur in all of them.
The birds often change positions, first one leading and then the
other. These changes occurred both during the ascent and in the glide.
In the glide the change was made by the upper bird moving beneath the
lower one. I was not able to determine the sex of the birds during
the aerial display.
Except that Arctic Terns sometimes carry a fish in the high flight
(Cullen, 1960a), the high flight of Sooties seems very similar to that
described by Cullen for the .Arctic Tern.
Aerial disprlas were nost co-non front 15:00 to 19:00, but I also
saw so2m in early m-rorning. Most occurred early in the breeding
season in Aoril but I also saw sone in Iay, June, and July when they
may have been given by late arriving adults or renester-s. Unfortul.ately
I did not see the events on the ground that preceded the high flight.
The function of the high flight is not known, but Cullen (1960a)
suggests that it may be part of pair formation. If so, perhaps the
high flights laLte in the nesting season involve young birds. As Sooties
as young as 3 years old visit Bush Key late in the nesting season but
probably do not breed for several years, perhaps they pair a year or
more earlier. It is not known hsw long Sooty Tern pair bonds last
but it would not surprise me if the same birds pair in successive
years. &tch an extended "engagement" period is known for another
long-lived pelagic species, the Laysan Albatross (Dionedea i nutabilis)
(Fisher and Fisher, 1969). Moynihan (1962) also saw many .aerial displays
by Brown Noddies and the Inca Tern (Larosterna inca) late in the
nesting season, and suggests that at least in the inca Tern these may
have been young birds pairing. Brown Noddies at Bush Key also ccrronly
give aerial displays late in the nesting season.
Parade.--Once the birds landed and stopped the almost continuous
panicking, displays on the ground began. The most conspicuous ground
display of Sooty Terns is the parade in which it assumes a distinct
posture and displays to other birds by prancing rapidly with short,
quick steps. A bird may prance toward, in front of, or around another
tern; or two may prance together, moving parallel to each other.
Palmer (1941) describes a similar display in the Comron Tern (Sterna
hirndo) and says it takes two different forms, each with a different
motivation. Although I saw these two forms in Sooties, I saw much
variation in posture betwe-en the two extremes.
In the parade posture, a Sooty extends its head and neck far
forward, often seeking the feathers. ,The wings are usually held well
away front the b:.-, particularly at the carpals, and are sonetines
lowered until they drag on zhe ground (Figure 2). Occasionally the
Figure 2. Posture of Sooty Tern in the parade display (taken
front Iredale, 1914).
wings are held tight against the bird's sides. Usually the tail is
tilted upward and often the tips of the primaries cross under the
base of the tail.
If the bill is pointed upward, it is called the erect posture and
Palmer (1941) says the display shows submission. Scoties may point
the bill horizontally or slightly above horizontal but seldon point
it straight up as some other terns do. Socties also tilt their head
to the side, especially when displaying with or to another bird.
Then they tip their head aw-y front the other individual b-u- seem to
lean the body even more toward it so that the near wing often drabs
on the ground.
I saw Sooties give this di slay almost i-maediately after landing
on the ground. Often it preceded copulation but other times two
birds separated after parading. Both sexes gave the erect form of
the parade, but when one bird circled and obviously displayed to
another, whenever I could determine the sexes, it was always the male
that did so. As the erect form. of the parade was con-inon almost
iraurediately after birds landed and before they established territory,
it certainly is important in pair forration.
In the other form of the parade, the tern points the bill down,
but Sooties seldom point it directly at the ground. Palner (1941)
calls this the bent position and says it is given when a tern tries
to intimidate another tern. Sooties do not take so extreme a posture
here, usually not holding the wings so far away from the body or
stretching the neck so far forward. Several tines I saw a parade start
with the bill pointing upward and end with i: pointing toward the
ground. Thus the position of the bill varies considerably. although
the basic wing and body postures are similar in the bent and erect
forms of the parade.
Both Palmer (1941) and Cull.n (1960a) note these two bill positions
occurring in aerial displays of terns. Although I watched for this,
in all cases that I could see the bill was pointed straight fo;rard,
the aerial position apparently comparable to the erect on the ground.
In the high flight Sooties usually fly away from the island, so I may
just have been unable to see the birds well enough.
The only calls I heard associated directly -with the parade were
occasional lo.w 'wuk avuk" notes, but usually the birds were silent.
Once I noted the throat of a parading male vibrating rapidly, but
although I was only 10 feet away I heard no sound. Notes too low
for me to hear may accomr.any the display.
Several times I saw a bird on the ground raise its head and give
a loud '"ke-wat-ic" or '"an-dick" call that was answered by a second
bird in the air. After the two exchanged several calls, the second
bird landed and then the two paraded. In one exchange the male called
to the female, and after parading the two copulated. Apparently the
male called to attract a female and then paraded to her. As the
erect form of the parade was most cor.ncn when the birds first landed,
I think it functions along with the high flight in pair formation.
Courtshin feeding. --I saw courtship feeding only 17 times, mainly
from 13 to 20 April but once as late as 17 June. lot all of these
terminated with the actual transfer of food. :'here the sex was known,
the female always begged. Generally she crouched, turned her head
toward the male, and gave a rapid series of cl; chuckling "ka ka"
or "yip yip" notes, raising her open bill toward him, and biting at
his bill, much as a young tern begs for food. A few times I actually
saw the male transfer food to the female, sometimes repeatedly in one
bout of courtship feeding.
Occasionally the parade preceded courtship feeding, but the latter
is relatively rare and certainly not a regular component of the parade.
Courtship feeding is a typical precopulatory behavior in most Larids
(Cullen and Ashmole, 1963). In Sooty Terns courtship feeding seemed
to be relatively rare and only occasionally preceded copulation.
Coculation.--The parade commonly precedes mounting and copulation.
The male parades in front of and around the female, stops beside her,
and then mounts. The female crouches, holds her wings out from her
sides, raises her tail, and stretches her head forward. The male then
moves back on the female, crouches, and copulates. After coitus the
male dismounts, and the two usually fluff th3ir body feathers and then
preen. A single pair nay copulate repeatedly.
Copulation was commonest on my plot from 6 to 23 April. Of 208
times I noted a male trying to mount a female, in only 56 (26.9 per cent)
did they seem to copulate. In 68 (32.7 per cent) attempts the male
fell off the female, the female did not crouch, or he simply
dismounted without trying to copulate. More commonly (84 times,
40.4 per cent), another bird interfered and either knocked the male
off the female or the nale dismounted and fought with the intruder.
Interference from other birds became commoner as the breeding season
progressed and eventually few pairs could copulate without other
birds interfering. Several tines I saw two to six males try to mount
Sooty Terns apparently do not establish territories until after
copulation. Even then the territory is not permanent until they
finally dig a scrape and the female lays an egg. Normally Sooties
claim an area spanning little more than what they can reach while
sitting on their egg. The scrape itself is a shallow hollow sIome
3 to 4 inches in diameter and about an inch deep.
Scrace-buildi.ng. --By 10 April scrape-building was common cn my
plot and from then until 16 April it was the bird's most conspicuous
activity. Although scrape-building obviously provides a site for the
egg, the incipient scrape-building behavior associated with it often
occurs before copulation and seems to be part of courtship in Soo:ies,
much as it is in Com mon Terns (Palmer, 1941).
Pairs of Sooties usually spend some time selecting a site and
nay make several false starts. While selecting a site they walk
together, poke at the ground, pick up bits of gravel, shell, or
vegetation and then drop then; start work at one site only to abandon
it shortly, and continuously give low "puck puck" calls. Typically
two birds stand close together, point their bills toward the ground,
then lower their body, and kick dirt back with their feet. They
often give low "puck puck" notes as they poke at the ground.
Nearly always a pair works together to dig the scrape. Usually
the male does most of the digging but the female helps at least
occasionally. As they work at the scrape, they pick up small pebbles,
shells, twigs, or other small objects and sometimes drop then into the
Scrape-building generally tops once thlefale las her egg, but
the birds occasionally work on the scrape after it contains an egg.
Several pairs that lost their chick shortly after it hatched began
work on new scrapes, but I did not see any bird lay again,
Fighting. --Consoicuous fighting started around 11 April when
scrapes were being dug. In fighting, two birds face each other,
sometimes approaching each other with their heads and bodies held low
and forward or else upright, the head erect and the crown feathers
raised. The two hold their wings out from the body and usually hold
the tail up. Usually each jabs viciously at the head and bill of the
other bird or grips the other bird's bill and then both shake their
heads vigorously. Occasionally they beat their opponent with a wing,
but normally the wings are used only for balance. Sometimes they give
lcw rasping growls. Finally the two release each other and each
retreats to its own scrape.
While some fights ended with one bird clearly supplanting the
other, I often saw them end differently. The two birds stopped
scuffling, faced each other with wings against the body, bills fo-rward;
and first, one and then the other lowered its bill into the bent posture
and turned its head slowly awayZ and then broke off the altercation.
Occasionally they gaped at each other before lowering the bill. Gaping
apparently is an aggressive signal (Moynihan, 1962) as is the bent
position. Thus the aggressive display apparently ended the fight.
The stare-down posture described earlier also occurs commonly
after fights. After a paired bird drove off an intruder, it commonly
gave the stare-down upon returning to its mate. Again the aggressiveness
seemed to end with the stare-down and the bird turned its attention
bank to its mate.
Thus the sequence of events preceding egg--laying seems to be as
follows: First the terns flock and land on Bush Key every evening in
a mass social activity, perhaps synchronizing the breeding cycles.
Gradually they extend the time they spend on the ground and as flocking
decreases, they begin individual sexual activity. Males seem to land,
perhaps near where they bred in previous years (see p. 75) and call
to try to attract females to them. When a female lands, the parade
aid high flight that follow probably contribute to pair formation.
They continue courting with parading and incipient scrape-building
until they copulate. The two then dig a scrape and begin to defend
the area irmediaLely around it.
After copulating and then working on a scrape for several days,
the female Sooty Tern lays a single egg. All 14 eggs whose time of
laying I was able to determine accurately were laid in the afternoon;
five between 12:00 and 14:00, five between 14:00 and 16:00, and four
from 16:00 to 18:00. Ridley and -ercy (1958) and Ashnrole (1963)
also note that Sooty Terns usually lay their egg in the afternoon.
After the first eggs were laid on 1 April 1968, egg-laying
gradually spread through the island, The first eggs were laid on ry
plot on 9 April and the pea: of laying there was 15 April.
Usually both me-mbers of a pair are present when the egg is laid.
The female cro.uches low in the scrace while the male stands nearby,
occasionally walking around her or poking at the ground. In the first
fe-.w -utes after the egg is laid the two usually sit on it alternately
several times. The female leaves the ecg, walks around it. and the
male takes over. He pokes at the egg, rolls it into the scrape if it
was laid on the edge, and in general worries over it before finally
settling down to shade or incubate it. Usually the female flies within
about 10 minutes, apparently to drink and dip as she often returns
with her breast feathers wet. The two then stay near the e-gg, first
one and then the other shading or incubating it. Exchanges now, as
later in incubation, consist of one bird forcing the other off the
egg and then taking over care of it. Exchanges are frequent during the
first few hours after the egg is laid, but by early the following
morning the male assumes care of the egg and the female is absent,
During incubation and to some extent after hatching, besides
noting which adult cared for the egg and the adult's general behavior,
I maintained accurate records of activity patterns of 15 pairs in the
following manner. I made a complete catalog of the bird's activities,
described below. I knew the date of laying and I could distinguish the
sexes of each of these pairs. In the activity records, every 30
seconds I made a 1-second "soot" observation and record of the bird
caring for the egg, assigning the bird's activity to one of several
readily identifiable categories (e.g., incubating, shading the egg, off
the egg and preening, etc. ). This method of making 120 observations
per hour per nest was used for 1-hour periods throughout incubation.
One fault of this method is that by cataloging activities,
slightly different activities must be grouped in a single category.
The categories with the most variety are those that occur when the adult
is off the egg, and as the adults either incubate or shade the egg
more than 90 per cent of the time, this variation is only a snail
fraction of all the birds' activities. The other alternative is to
watch a few nests continuously and record all the activities. Even in
a Sooty Tern colony, where nests are close together, it is difficult
to follow all activities at a few nests, especially when several
suddenly erupt into activity with fights or other interactions.
Watching a fe;- nests involves the risk that some or all of them may
be destroyed, deserted, or otherwise disturbed.
On the practical side, continuous watching of a few nests is very
tiring. By viewing 15 nests, I kept my eyes moving and thus avoided
some of the monotony and fatigue of continuous watching.
I analyzed these activity records by sex, hour of day, and date
in the incubation cycle, excluding all 1-hour watches in which
exchanges occurred. Luring incubation, I have records for 1,632
bird-hours, each including 120 spot observations for a total of
195,840 spot observations. These include at least one hour of watching
for each sex for each of the 12 daylight hours for most days during
incubation, and for some I have many more. These were used to prepare
Figure 3 and Table 1.
Behavior Dri.ng Incubation
During incubation Sooty Terns primarily tend the egg and perform
a few maintenance activities during brief spells away from it. I
categorized their activities into the following types.
Incubation, in which the adult bird places one of Its two brood
patches directly against the egg, is obviously one of the most important
activities, Sooties incubate mainly at night, in the early morning,
and in late afternoon (Figure 3).
As the air temperature increases in the morning, adults gradually
Figure 3. Activities by adult Sooty Terns during incubation.
/ Shading .-
/ I Incubating .
0 \ Other
80 activities ........
60 / \
6o FA I
20- / D
,10 .-..* * .... ... ., ..
-., '__i ..
06:00 08:00 10:00 12 :00 14:00 16:00 18:00
Time of Day
rise up off the egg so that most days by around 10:00 they are shading
the egg rather than incubating it. They usually continue to shade the
egg until around 17:00, when they gradually change back to incubating
again (Figure 3). To shade the egg the adult merely stands over it and
keeps it in its shadow. As the sun changes position during the day
the birds turn to keep their backs toward the sun. This keeps the
maximum surface area of the body exposed to the sun and, as the upper
surface of a Sooty Tern is black, would seem to present problems of
heat load for the bird. However, keeping their back toward the sun
best insures shading of the egg. By turning one side of its body
toward the sun, the adult could probably lower its heat load, but it
also would be more likely to expose the egg to The sun. The adults
probably reduce the heat load several other ways. As it becomes hotter,
they erect their backfeathers and hold their wings out from their
sides, increasing insulation and the area for heat less. They also
gape widely and pant, apparently dissipating more heat than in normal
breathing. Another common activity is dipping as described below.
The change from incubating to shading and back again is very
gradual and proved my most difficult choice in recording activity.
Figure 3 shows clearly when this change is made. Both incubating and
shading involve direct care of the egg. All other activities, in
which the adult is not actually caring for the egg, are shown as the
bottom line in Figure 3.
Among these other activities, the most frequent is standing to
the side of the egg and preening (Table 1). Incubating adults commonly
rise up off the egg, poke at it, and apparently turn it. Other times
the adult walks around the scrape or to the ed e of its territory
Table 1. ActiLi ties other nhan incubacing or shading by Sooty Terns
caring for an egg
spot obse:rations 7er cent of all
Activity of each activity such activities
Preening 4258 48.4
Poking at egg 1540 17.5
Standing beside egg 1171 13. 3
Wa4lki]ng 886 10.1
Fighting 673 7.6
Off away from scrape 276 3.1
Total 8804 100.0
and stands. Occasionally a bird leaves its egg to spar briefly with
another tern, either a neighbor or one that has cone too close.
Generally Sooty Terns are much less bellicose daring incubation than
they are earlier in the nesting cycle or after eggs hatch. Perhaps
hone ground is permanently established and recognized, and the absence
of chicks to intrude on others' territories enhances stability.
Adult Sooties commonly defecate near their egg. They walk to the
edge of their territory and turn to face the egg before defecating.
Sometimes a stretching notion accompanies defecation. The bird
stretches both wings vertically above the body, leans forward with
head outstretched, and lets fly. thus stretching and defecating in
almost the sane notion.
Another activity of adults during incubation is suddenly flying
away leaving the egg unattended. On some occasions the birds fly
away when panicked and return almost immediately. Other times this
seems to be associated with the aerial activity called dipping.
Sooty Terns are closely attentive during incubation. The
percentage of time adults spend incubating or shading differs only
slightly between sexes. Females spend a little more of their time
incubating and males a little more shading, but with incubating and
shading combined, the total attentiveness, analyzed by hour, differs
by less than 1.2 per cent between sexes. In all, the bird caring for
the egg spends over 95 per cent of the daylight hours either
incubating or shading and only about 4.5 per cent in a variety of
other activities (Table 1). These other activities prevail from
07:00 to 10:00 and frcn 17:00 to 18:00, the times when birds shift
between incubating and shading (Figure 3). Presumably at these times
egg and air temperatures are much the same and care by the adult is
not so critical. Scoties are nost attentive from 12:00 to 15:00 and
send over 98 per cent of their time directly caring for the egg,
mostly by shading. This is certainly the time when the egg experiences
the highest environmental temperatures, and presumably when adult
care is vital to the ebioryo's survival.
An activity of Sooty Terns at Bush Key that is especially noticeable
during hot midday hours, but occurs occasionally throughout the day,
entails their flying rapidly from the island, briefly dipping their
bill, feet, or breast feathers into the water, and returning to the
island. TWhen only the bill is skim ed along the water's surface, it
seems obvious that the birds are drinking, but dipping the breast
feathers requires another explanation.
Usually each bird flies directly and rapidly from the colony at
a height of about 10 feet. It then drops down, dics the bill in the
water, drags the feet in the water, and sometimes alm-ost lands
nonentarily, wetting the feathers in the process. About 100 to 300
feet offshore the bird turns abruptly and flies back to the island.
An individual nay dip from one to as many as ten times on one flight.
In 113 flights between 13 May and 7 June in which I recorded from a
boat all dipping movements. and also whether the bird landed back in
the colony or rith groups of terns sunning on the beach, over 80 per
cent of all dipping occurred on the flight out. In about. 95 per cent
of the flights the birds dipped the bill at least once (Table 2), and
in only about 17 per cent they dipped their breast feathers. As one
Table 2. Suimary of 113 dipping flights by Sooty Terns
No. of flights that
included dipping No. of times the following
following parts of body: parts of body were dipped:
Where flight Total no.
ended of flights Feet Bill Breast Feet Bill Breast Total
Cn beach 35 13 34 0 22 60 0 82
In colony 78 45 74 19 102 158 49 309
Tutals 113 58 108 19 124 218 49 391
dipping notion night result in more than one part of the body being
dipped, I recorded the number of times each part of the body actually
was dipped (Table 2). The number of times a bird dropped down to dip
is somewhat less.
None of the flights that ended with the bird landing on the beach
among birds sum--ning or resting included dipping the breast feathers,
while about 25 per cent of the flights that ended with the bird going
into the colony where eggs and young were present included dipping the
breast feathers (Table 2). This suggests that dipping the breast
feathers possibly has some relevance to nesting, something also
suggested by behavior I noted at the scrape itself.
Often du-ring the hot midday hours, I saw a single bird fly from
its egg, not in a panic, and head tow-ard the water (Table 3). After
about a ninute it returned, flying low and rapidly tcwarc the scrape.
As the bird settled back on the egg, I often could see that its bill,
feet, and sometimes the breast feathers were wet. Obviously these
direct flights from the egg are the start of the dipping flights that
I watched over the water. As dipping is mcst common when it is hottest,
it seems possible that this behavior is theinoregulatory, both for
the adult and the egg. Both sexes dip with equal frequency (139 to
137) so I have combined their records. Dipping by incubating adults
is most frequent during the first 15 days of incubation (Ts.ble 3).
Watson (1908) mentions an activity similar to dipping but says the
birds were bathing. The Ashnoles (1967: 62) note that incubating
Sooties somet-mes fly off to drink, Both Tompkins (1942) and Hardy
(1957) have seen Least Terns dip their breast feathers during
incubation and suggest that it provides water necessary for the eggs.
Table 3. Tire of occurrence of flights away from the egg by adult
Number seen during:
First 15 das Second 15 days
Time of incubation of incubation Total
06:00-107:00 4 () 1 (1) 5 (5)
07:00-0S :00 6 (3) 0 (0) 6 (3)
08:00--09:00 8 (7) 5 (4) 13(11)
09:00-10:CO 31(24) 4 (3) 35(27)
10:C0-11:00 31(25) 11 (8) 42(33)
11:00-12:00 24(18) 5 (5) 29(23)
12:00-13:00 21(19) 17(15) 38(3-)
13:00-14:00 24(20) 14(11) 33(31)
14:00-15:00 23(20) 4 (4). 27(24)
15:00-16:00 18(12) 3 (3) 21(15)
16:00-17:00 20(14) 1 (1) 21(15)
17:00-18:00 0 (0) 1 (1) 1 (1)
Totals 210(166) 66(56) 276(222)
a. Number of ties in activity records birds were away frc egg.
b. :,uber of act-ual flights represented, i.e.,one flight night
involve t 0ro or r.:ore spot cbserv tions in activity records.
Howell and Bartholonew (1962), working orn idway Island in the
Pacific, show the delicate role that parental care plays in preventing
Sooty Tern eggs from approaching the high and probably lethal
temperatures they would reach in open sunlight, 'Jn : id- ay the adult
maintained the egg temperature above that of the surrounding air and
below that of the surrounding sand. Unfortunately they do not say
whether the adult is incubating or shading the egg.
I attached thernisters to several eggs and monitored surface
temperature of the egg and air temperature at about 4 inches above the
ground for parts of several days. Unfortunately the eggs I worked
with were close to my blind, and the adults caring for then were easily
disturbed by my movements. Also the wire attached to the egg hindered
the tern in turning it and affected the bird's behavior.
I obtained useful information from a 21-dPay-old chick on 6 June
and conflicting information from the sane egg the next day. On 8 June,
in 23 temperature readings taken between 10:20 and 13:05, the egg
averaged 101.50F (range 99 to 1050F) and the air at 4 inches averaged
105.50F (range 103 to 108.50F). The highest egg temperatures occurred
when the adult left it to dip or flew off in a oanic. In both cases
the adult dipped the breast feathers before returning to the egg. The
next day in 35 readings the egg temperature averaged 105.2-F,
somewhat above the air temperature that day (1050F) and 3.7F above
the egg's temperature on the 8th. Although the adult repeatedly left
the egg to dip, it seemed unable to lower the egg's temperature. As
the egg cracked later that day and proved infertile, perhaps the heat
of decay foiled the adult's attempts to lower the egg temperature.
On the 8th the adult had been able to maintain the egg's
temperature some 4F below the air temperature. When the adult flew
off to dip, the exposed egg's surface temperature rose I to 20F while
the bird was absent but then dropped 2 to 50F within a few minutes
after the adult returned. The adult did not hold the wet breast
feathers against the egg, but held them right above the egg.
Although I have limited information, I suggest that dipping the
breast feathers by incubating terns helps regulate either the egg's
temperature or humidity. Some adults dipped often during incubation
while others did so only rarely. If dipping the breast feathers is
thernoregulatory, it is puzzling that the tern does not hold the wet,
cool feathers directly against the egg. Possibly the water dripping
off the feathers onto the egg provides moisture needed by the egg.
Late in the breeding season adults continue to dip but seldom dip
their breast feathers. Cn 27 June 1970 when most chicks were 6 to 8
weeks old and few eggs were present on Bush Key, nearly all of the
hundreds of adults flying out dipped only their bill or occasionally
Most of the 61 nest reliefs observed during incubation occurred
during early morning or late afternoon (Table 4). Generally the adult
lands near its scrape and walks up to its mate on the egg; The bird
on the egg either steps off the egg so the second bird can incubate,
or else the newcomer pushes the incubating bird off the egg and then
takes over care of it. iNo elaborate ceremony occurs. The relieved
bird generally flies away within a few minutes, presumably to drink,
but often returns to the scrape and lingers for several hours or more
Table 4. Time of rest relief during incubation
.:Lunber of :ours of Exchages/hour
Tiae exchanges Observation Observation
06:00-C7:00 13 21.52 0.60
07:00--08:00 4 27.92 0. 1
08:00--09:00 4 29.03 0.13
09:CC-10:00 3 23.50 0.12
10:cO-11:00 5 19.33 0.25
11:00-12:C0 1 13.35 0.07
12:00-13:CO 1 10.33 0.09
13:CC-14:00 3 10.25 0.29
14:0o0-5:00 i 10.97 0.09
15:00-16:00 3 14.78 0.20
16:00-17:00 12 16.93 0.70
17:00-18:00 6 17.00 0.35
18:0C--19:CC 3 4.50 0.67
19:00-20:00 2 1.17 1.70
Total 61 220.46 0.27
before leaving Bush Key. Sometimes this bird forces its way back
onto the egg so several exchanges occur before the new bird finally
takes over and the other leaves the island. I never saw an adult
feed another one at a nest exchange during incubation.
Although I saw some exchanges in midday (Table 4), most birds
return to the colony in the evening and relieve their mate then.
Instead of circling near Bush Key as they do earlier in the nesting
cycle, returning adults fly directly to the island and presumably
to their scrape. Some circle over the island itself, calling loudly
so that gradually the colony becomes noisier. The number of hours I
watched in early evening is much smaller than for other times.
Otherwise I certainly would have seen more exchanges then.
At Bush Key few adults regurgitate when mist-netted in the
morning or early afternoon, but many do so in the late afternoon,
indicating they just returned front feeding. The exchanges I saw
around sunrise may have been pairs making a final exchange after
several exchanges during the night, and the relieved bird was at last
leaving to feed. Much as when the colony is forming, the noise
gradually diminishes and by about 08:00 few birds are in the air.
Those that remain at the cclcny incubate quietly from about 08:00 to
Ashmole (1963) found nest relief most abundant from 21:00 to
06:00. He reasons that arriving at that time means that they fed during
daylight and then fle-.: some 5 to 8 hours to arrive at the colony in
the middle of the night. I have no records of changeovers from 21:00
to 06:00, but the massive influx of terns early in the evening suggests
that changeovers are commoner then than late at night as they are on
Ascension Island. If Ashmole's reasoning holds, then Bush Key Sooties
feed within a few hours flight of Bush Key at most and can easily
return when finished feeding.
Length of Incubation Shifts
To dete-rmine the length of incubation shifts (the length of time
one adult cares for the egg without relief), I recorded which adult
incubated each day for most of the nests I kept activity records on.
Usually I found that the bird present in the morning remained there
until at least late afternoon. Because most adults seer. to return
and exchange in the early evening, I assumed this was true unless I
had evidence tc the contrary. I have computed incubation shifts as
multiples of 1-day periods. I nay have missed a few exchanges but
the pattern I found is generally true of incubation shifts at Bush Key.
Watson (1908) also notes that most Dry Tortugas Sooties return in the
evening and that incubation shifts generally are 24 or 48 churs long.
Of 231 incubation shifts measured in this narner, 148 (64.1 per
cent) were 1 day in length, 70 (30.3 per cent) were 2 days long, and
only 13 (5.6 per cent) were 3 days long. Figured differently, Sootics
did 45.3 per cent of their incubating in 1-day shifts, 42.8 per cent
in 2-day shifts, and 11.9 per cent in 3-day shifts. Even though 1-day
shifts are more than twice as common as 2--day shifts, the two
contributed about equally to incubation.
Some pairs alternated daily throughout incubation, while others
alternated regularly in 2-day or, in a few cases, 3-day shifts. Otners
seemed to follow no set pattern of relief. Overall on the basis of
these records, males spent sciewhat more time (175 to 152 days) caring
for the egg than did females, but the difference is not significant
(X2=1.62, F>0.20). In the hourly activity records, males again
predominated (880 to 752 hours), the difference being statistically
significant (Z2=10.04, F<0.005). As the female left the egg shortly
after laying and the male cared for it on the first shift, some of
this difference occurred then, but even excluding the records for the
first day of incubation, the male still cared for the egg more than
W'atson (1908) shows that incubation shifts on the Tortugas average
somewhat over a day long with the longest a little over 3 days, well
within the range I found in 1968. On Ascension Island incubation
shifts average 132 hours (Ashmole. 1963), over twTice as long as those
on the Dry Tortugas. On Christmas Island in the Pacific, incubation
shifts are about 7 days long (Ashmole and Ashmole, 1967). Incubation
shifts vary from 2 hours to 3 days on the Seychelles (Ridley and Percy,
The striking differences in incubation shifts between Bush Key
Sooties and those at two other colonies are probably related to food
availability and distance from the island. Ashmole (1963) reports
many young at Ascension dying of starvation one year, apparently
because of a failure in their food supply. On the Dry Tortugas Sooty
Tern chicks have never been known to experience heavy mortality from
food shortage (Robertson, 1964). No evidence exists to tie the almost
total nesting failure at Bush Key in 1969 to food. The shorter
incubation shifts seem to indicate that the terns have an adequate
food supply near the island.
Influence of weather
The most obvious effects of weather on incubation patterns of Sooty
Terns are the shifts from incubating to shading and back to incubating,
depending on the environmental ten-erature (Figure 3). Sudden changes
in weather provided natural exper-ments on weather effects. On hot
days when clouds rapidly lower the air temperature, shading birds
quickly change to incubating or move away from the egg to preen or
perforrn other maintenance activities. Once the clouds pass and open
sunshine returns, the birds again shade the egg. On normal, clear hot
days an adult virtually never leaves its egg between 11:00 and 15:00
except to dip. The few records I nave of terns performing other
maintenance activities during those hours nearly always occurred on
Rain also changes activity patterns. At the start of a rainstorm,
swarms of Sooty Terns rise and circle over the colony, calling noisily.
As nearly all birds that are caring for an egg incubate throughout the
storm, it is primarily free birds that circle over the colony.
Apparently Sooty Tern feathers are not water repellent and are
easily soaked by rain. After rain Sooties preen their body and flap
their wings vigorously. This apparently helps them dry the feathers
to permit flight. Birds have great difficulty taking off in early
morning after a heavy dew has soaked their feathers. The flocks of
terns rising at the start of rainstorr-s are probably birds getting
airborne before they become too waterlogged. Once in the air, nany
remain there until the storn is over, the notion of their wings
preventing then front getting soaked.
Sooty Terns incubate their eggs 29 cr 30 days (nean 29 da; 12.3 -
2.4 hours, range 28 days 22 hours to 30 days for 16 eggs). This agrees
well with Ashiole (1963) who found that usually the egg is incubated
28.5 to 30 days. Watson's (1908) figure of 26 days seem somewhat
short, though Ririley and Percy (1958) say Sooties incubate for 26 to
29 days on the Seychelles.
The chick may pip the egg as much as 36 hours prior to hatching,
but usually does so only the day before hatching. For eggs that I
could detem-ine the tine of hatching exactly, most hatched around
sunrise or around ncon.
Adults attend pipped eggs very closely and seem reluctant to leave
them. The only time that Sooty Terns ever actually mobbed or struck me
in the colony was when eggs were hatching. Howell and Bartholomew (1962)
show that such an egg is particularly sensitive to heat stress and has
no effective means of dissipating excess heat. Thus care by the adult
is especially important to insure the chick's survival.
Adults appear indifferent to the presence of the empty egg shell
in the scrape and sometimes continue to sit on it after the chick has
emerged. About 20 times I saw an adult pick up and carry off a piece
of egg shell, often one from a neighboring scrape. Usually adults
allow the empty shell to roll around until it is finally broken and
Sooty Tern eggs (like those of most terns) are speckled and
camouflaged on the outside, whereas the white inside lining contrasts
sharply with the sand background of the nesting colony. Tinbergen et al.
(1962) show that rapid removal of the empty shell has definite survival
value for Black-headed Gull (ari ridibundus) chicks. in a colonial
species like uscata., removing the egg shell probably makes little
difference to chick survival. Any predator that reached the colony
would have little trouble finding a chick, whether the ~white inside
lining of an empty egg marked the scrape or not.
My data on parental care of chicks are less complete than those
for care of the egg. Every time 1 entered my blind, the chicks more
than a few days old scattered and hid under the nearest cover. It
sometimes took several hours for them to return to their scrapes where
I could watch then again, and some never returned.
The records I do have, suniarized for the first 2 weeks of the
chick's life, indicate that the adults are most attentive during the
midday hours, much as they were when caring for the egg (Figure 4). As
in incubation, adults rise up off the chick to shade it in hotter
hours of the day and brood it when it is cool. Fenales care for the
chick somewhat more than nales (157 bird-hours to 132 bird-hours), but
statistically these are not significantly different (C2=C2.16, P>0.10).
I found little difference between nales and females in care of chicks
and I have combined the records in Figure 4.
Generally adults are less attentive when caring for the chick than
when caring for the egg. They often stand to the side of the chick
for an hour or more, and their attentiveness decreases as the chick
grows older. After the chick is about 3 weeks old, the adult spends
very little time actually caring for it other than feeding it, although
they may stand near it during the day. From then on chicks even seem.
to spend the night unbrocded by an adult.
Figure 4. Activities by adult Sooty Terns during the first
2 weeks of caring for the chick.
~J ~. _I LL
Time of Day
__ __ (_I_ __ _
Chick care centers around the provision of food. Sooty Terns
apparently feed almost exclusively on fish and squid (Ashmole, 1963;
Ashmole and Ashnole, 1967). Much of this food is caught when schools
of tuna, mackerel, or other large predacious fish drive smaller fish
to the ocean surface. There the Sooties dip down to seize food from
the surface or in the air above it, but seldom dive headlong into the
water for their prey as most other terns do (see Ash:ole and Ashmole,
In the detailed study by the Ashmoles (1967) on Christmas Island
in the Pacific, fish nake up about 60 per cent of the food items and
squid 40 per cent, with tha figures reversed when volume is measured.
Flying fish (Exocetidae) and tuna (Sconbridae) are two of the most
important fish along with snake mackerel (Gempylidae) and
On the Dry Tortugas Sooties have been reported eating fish of the
families Carangidae and Clupeidae (Watson, 1908). A variety of
regurgitated fish have been found in the tern colony (Longley, 1929;
Longley and Hildebrand, 1941; see Erdman, 1967) but none of these lists
separates what was taken specifically by Sooties. In recent years
Robertson has collected many fish and squid that Sooties and Noddies
regurgitated when mist-netted. At present only the Scombrids have
been analyzed (?oth"off and Richards, 1970).
I saw one chick fed about 4 hours after hatching, and they may be
fed even earlier. The adult stands near the chick and points its bill
down. The chick then either grips and bites the adult's bill or else
pecks at it. The chick nay also beg with some rapid "cheep" calls.
This action seems to release regurgitation by the adult. The adult
stretches its neck arid head upward, tips the bill down, gags, and then
brings up a fish. The adult may hold the fish in its till before
feeding the chick, or it may slide Lhe fish directly down the bill and
into the chick's mouth. Sooty Terns were never sesn returning to the
colony with fish in their bill, but always carried the fish in their
esophagus and regurgitated it for the chicks.
Older chicks seemed to beg more by opening their bills and giving
rapid "cheep" calls than by pecking at the adult's bill. They also
grabbed at the fish while the adult was still holding it, sometimes
leading to tugs-of-war between the two.
In nearly every case where I could see clearly the exchange of
food from adult to chick, the fish was regurgitated and passed to the
chick tail-first. Iany of the fish appeared to be scaleless, and in
some cases they were partly digested or fragmented, especially the
last ones passed to the chick (presumably the first ones the adult
caught). The regurgitated food is often coated with nucous. This may
retard digestion of the food as the adult carries it back to the colony
(Askhole and AsLhmole, 1967).
A chick might get six or more fish in one feeding, depending on
the size and degree of digestion of the food. Several tines I saw an
adult feed a chick six times in 3 to 4 minutes, each time passing one
fish. Other times a whole meal might be one large bolus containing
Of 349 occasions when I definitely saw a chick fed, the majority
occurred from 06:00 to 10:00 and from 16:00 to 18:00 with many of the
rest occurring just before or after those periods (Table 5). Passing
Table 5. Time of feeding of young Sooty Terns
Adult feeding young
of times Undet. adult Hours of Feedings/
Time fed Male Female Both sex present obs. hour
06:00-08:00 74 22 21 9 22 16 54.78 1.35
08:00-10:00 58 27 23 1 7 27 34. 33 1.68
10:00-12:00 26 13 8 0 5 5 16.82 1.54
12:00-14:00 12 5 6 1 0 5 14, 42 0.83
14:00-16:00 44 18 6 1 19 8 22.88 1.92
16:00-18:00 85 34 34 4 13 29 33.48 2.53
18:00-20:00 50 19 4 1 26 1 20..83 2.40
Totals 349 138 102 17 92 91 197.54 1.76
several fish to a chick in the soace of a few minutes is counted as
one feeding. As the number of observation hours varied during the
day, the number of feedings per hour of observation (Table 5, last
column) is a more valid comparison. This shows that peak chick feeding
occurred in late afternoon and early evening with a smaller peak early
in the morning. These observations were made from 16 May to 10 July,
the tine when chicks were present on my plot.
In about 5 per cent of the feedings, both adults fed.the chick; far
more commonly two adults were present but only one actually fed it.
Of 240 cases in which I knew the sex of the bird feeding the chick, the
male did so more often than the female (138 to 102) and the difference
is statistically significant (.2=5.40, P<0.025).
In addition I recorded some 240 occasions where a chick begged
and an adult tried to feed it without actually doing so. Sometimes the
adult regurgitated and held food in its bill and then reswallowed it,
even when a chick begged loudly. The adult night do this several ties,
but usually eventually fed the chick. Other times an adult tried to
regurgitate but was unable to bring up any food.
The adult places all food directly in the nouth of very small
chicks. Any that drops on the ground remains there unless an adult
picks it up to eat or feed to the chick. The youngest chick to pick
up food from the ground was about a month old, but they may do so when
younger. These older chicks oick up food from the ground on scrapes
other than their own. Once as an adult held a fish in front of a chick,
I saw another adult seize tne fish and swallow it.
Freauency of Feeding
Sooty Terns on Bush Key feed their young infrequently enough to
make it difficult to obtain good information on the rate of feeding.
Although I watched many chicks continuously for long periods, I have
few good records of an adult feeding it, leaving it to forage, and
returning to feed it again. A major complication is that rather than
immediately giving a chick all the food it has in its esophagus, an
adult may feed the chick several times over a period of several hours.
I saw one adult regurgitate and feed its chick some 5.5 hours after it
returned to the colony. Thus if an adult fed a chick, flew away and
returned in 4 or 5 hours and fed it again, I could not be sure if the
adult had left the island to forage in that time or if it had just
loafed elsewhere on Bush Key before returning to the chick. Consequently
I have determined the rate of feeding two ways.
I have five good records of an adult leaving the chick and
returning later to feed it. These were all made on 26 May with chicks
2 to 8 days old. The length of time the adult was absent (foraging
time) averaged 3.87 hours and ranged from 2.63 to 5.30 hours.
I also tabulated the number of times chicks were fed during long
periods of continuous observation, separating these into chicks up to
15 days old and ones 15 to 27 days old (Table 6). After listing the
total number of times the chick was fed, I subtracted those extra
feedings that occurred when an adult fed a chick more than once after
returning to Bush Key. This gives the probable number of foraging
trips made per bird-hour of observation and gives a rough estimate of
the time an adult spent foraging, about 3.44 hours. Apparently
foraging trips are about an hour longer when the chicks are older,
Table 6. Rate of foraging by adult Sooty Tens
i 'uber probable number of
Sc number of Bird-hours hours per
Age of chick feedings foraging trips of observation foraging trip
1-15 days 14i 113 369. 42 3.27
16-27 days 20 19 84.58 4.45
Totals 164. 132 454.00 3.14
but the sample size is small and the adults may just loaf part of the
time. Note that here time is in bird-hours of observation whereas in
Table 5 it is hours of observation, during each of which I watched
about 15 chicks.
The two estimates are close enough to indicate that Sooty Terns
at Bush Key probably forage 2 to 5 hours when they are feeding their
chicks. Adults seem to pass food to older chicks nore rapidly after
they return to Bush Key. That is, rather than standing with the chick
and feeding it several times in a few hours, they seem to feed it all
the food they have and then leave it.
Often I saw exchanges that appeared to be after foraging trips of
4 to 8 hours but the birds nay have exchanged in short intervals when
I was not watching. Thus my records of foraging time are biased for
shorter periods. Watson (1908) gives 4' to 7 hours as the interval
between feedings by Dry Tortugas Sooties, consistent with my findings.
Sooty Tern chicks on Bush Key are fed much more frequently than
those in other colonies that have been studied. On Ascension Island
chicks are probably fed about once a day (Ashmiole, 1963). On Christmas
Island chicks are probably fed every 2 or 3 days (AshnMole and Ashmole,
1967) but good information is lacking.
As the chicks grow, the adults become less attentive and often
gather by the hundreds on the Bush Key beaches and sun much of the day.
Other adults soar over the island during the hotter tines of day (see
p. 6! ). Perhaps adults away from their chicks spend part of their time
in these groups. They must spend some of their time foraging but
exactly how much is difficult to determine.
At Bush Key more adults are present and the colony is noisier in
the evening than at any other time of the day. Hence I think that both
adults are probably at the colony at night, though both may not be
with the chick at all times. Elsewhere Sooty Terns feed at night at
least occasionally (Bruyns and Voous, 1965; Gould, 1967), but at Bush
Key, with the noticeable influx of birds in the evening and exodus in
the morning, I doubt that many adults feed at night. As during
incubation, mist-netted birds regurgitate far more frequently in the
evening than at other times.
The average flight speed of Sooty Terns is about 27 mph (Schnell,
unpublished). Thus on longer trips they may forage as much as 100
miles from Bush Key, but the average foraging trip of about 3.5 hours
gives them a range of about 47 miles. Sooties from Bush Key usually
fly toward the west when they leave the colony and return from that
direction. How far they go is unknown but they easily could fly to
the nearby Florida Current and feed there.
At about 3 weeks of age, the chicks begin to wander from their
scrape. They may spend much of the day alone, being accompanied by an
adult mainly when being fed. Some adults seen to check on their chick
periodically during the day, landing beside it, possibly feeding it,
and then flying off only to return in an hour or so. These adults
probably spend much of their time loafing on the beaches or soaring.
I was not able to visit Bush Key during the height of Hurricane
Abby (3 June) but on 4 June it was obvious that the tern's ncrmal
schedule had been disrupted as few adults were present until that
evening when they started to streak into Bush Key and continued to do
so until the evening of the 5th. They probably had been un.b.'e to
feed during the storm and had left as soon as possible to obtain fcod
for themselves and their chick. Mason and Steffee (1966) noted a
similar disruption after Hurricane Alma.
As the chick grows, its physical appearance changes and it may
wander farther from the scrape. As the colony has thousands of chicks,
adults undoubtedly have probleris locating and feeding their own (or
adopted) offspring. Sooty Tern chicks and adults evidently learn to
recognize each other so the chick can be located and fed.
Much as Lashley (1915) and Burckhalter (1969) report, I found
that parents apparently do not recognize their chicks individually at
first. In the first 4 or 5 days chicks often return to the wrong scrape
and are readily accepted and reared by foster parents. After that they
are pecked savagely if they intrude near another scrape, and some are
killed. Thus adults seem to recognize their chicks when they are about
4 to 5 days old. As Davies and Carrick (1962) and Hutchison et al. (1968)
suspect for other terns, this recognition probably is based on calls
between the adults and chicks.
After a disturbance, typically the adult tries to attract its
chick back to the home scrape with a combination of bill novements and
low calls. The call is a low two syllable "kraa-unk" note accompanied
by bowing movements of the bill and head, ending with the bill
pointing down toward the breast feathers. Often an adult used such
behavior to lure the chick back to the scrape, backing away from the
chick and toward the scrape until the two were finally back at their
own scrape. Once I watched an adult hold a fish in its bill and back
away from its chick, apparently using the fish to lure it back to the
By 3 weeks of age nost of the chicks on my plot spent nuch of
their tine hiding under the nearby bay cedar. By removing most of
the vegetation front the nesting plot I nay have forced them to nove
to this cover. Typically these older chicks emerged from the bay
cedar late in the afternoon and stood at its edge or ran out to their
scrape. The chicks ran swiftly to a spot and stopped, seemingly
knowing where they could stop and not be attacked by other chicks
and adults. Once on the scrape, they stood and waited for an adult to
come and feed then. Whenever caught one of these chicks, it was
always at the scrape where it had been reared and thus had "homed"
correctly. Also when an adult fed the chick, normally the adult was
one of the pair that had used that scrape. Hence the chicks apparently
knew their home scrapes and returned there to be fed by a parent.
Sometimes the adult landed at the scrape before the chick arrived.
Then the adult r,ypically circled 10 to 15 feet over the scrape and
delivered loud '"wid-ik" or "ka-wid-ik" notes from the air until it
was answered by a loud piercing "che-up" call front a chick hidden under
vegetation. The two birds exchanged calls several tines and the adult
landed. The chick then ran out to the adult, begged, and was fed. As
the chick approached the adult, the adult often gave what appeared to
be a greeting, flying straight up 2 to 5 feet in the air, giving a
loud "wide-a-w-ake" call at the peak, and dropping back down to the
ground to feed the chick. These "fly-ups" occurred in other situations
and are discussed later. Occasionally the adult pecked the begging
chick sharply. I was not able to determine the relationship of the two
in these cases, but I suspect that the tw.o had nade a mistake in
recognition and the adult did not realize it until the chick cane close.
Although chicks are usually fed by their parents (or by adults that
adopt then), at least eight times I saw a tern feed a chick other than
its own. In five of these instances an adult that had lost its chick
or whose egg had failed to hatch fed a chick, usually at an adjacent
scrape. Three times an unmarked adult fed a chick whose parents were
both marked and thus recognizable.
I also saw some chicks that seemed to approach and beg to any
adult that cane near. Although these chicks were not marked, I am
fairly sure that some were fed by adults other than their parents.
Often the adult pecked the chick and drove it away, but often the adult
tried to regurgitate. Occasionally it was successful in bringing up
some food that it then fed the chick. Possibly these chicks had lost
their parents in the shuffle of chicks in the first few days of life,
and no adult recognized them as its own. The adults that fed then or
tried to feed them may have been their lost parents, or other adults
that had lost their chick or egg.
Thus older chicks seem to find their parent both by knowing where
the home scrape is and by recognizing the adult's voice. Although I
think Sooty Tern adults and chicks do recognize each other
individually, I believe that nuch of this recognition is done by the
adult, while the chick may try to get food from almost any adult that
Sooty Tern chicks can walk almost immediately after hatching and
stand and beg for fecd within 4 hours. For the first few days they are
closely brooded by one of the parents, the chick resting either
between the parent's feet or crossways in front of them. At first
they walk by half crouching with the body low and almost on the ground,
but they soon walk upright like adults. After a disturbance these
small chicks frequently end up in the wrong scrape. I often saw two
small chicks under one adult that seemed to accept both of them. The
extra chick eventually returned to its own scrape, usually when
called by an adult. A common alarm response of these small chicks was
to lie flat on the ground with the head and bill extended forward.
Chicks apparently did this to avoid being pecked by adults, as adults
pecked viciously and sometimes killed strange chicks that came close.
Chicks 5 days old preened and voided with movements typical of adults.
For the first 2 weeks, one adult nearly always stays at the
scrape with the chick. DIring the 3rd week this attention gradually
diminishes, and by the end of the 3rd week the chick is often alone.
Chicks defend the scrape, pecking at and driving off other chicks and
Brown Noddies that intrude.
Most chicks stay close to their scrape until they can fly, but
older chicks that wander or are moved can find their way back to the
scrape when displaced several hundred feet (Burckhalter, 1969). Chicks
from scrapes close Lo the beach may sun and rest with adults in large
flocks on the beaches during the day, and then disperse in late
afternoon. Generally the chicks congregate just above water level and
the adults assemble higher on the beach. Perhaps the sand is cooler
there than higher on the beach. I did not see adults feed chicks
in such flocks. Apparently chicks return to their home scrape or
some other nearby place to be fed.
1 observed 6-week-cld chicks jumping into the air and flapping
their wings vigorously. I lack exact records of when they begin to
fly, but one chick flew several hundred feet on 4 July and others
were doing so within a week. If that chick had hatched from one of the
eggs laid on 1 April (it was in that part of the colony), it would
have been about 9 weeks old, the best estimate I have for the age
Sooty Terns begin flying at Bush Key. As noted by Ashjnole (1963),
chick development varies with the food supply. One year he found
chicks at Ascension flying when about 8 weeks old and the next year,
when food apparently was scarce, birds around 9 weeks old.still could
not fly. Burckhalter (1969) saw 8- eek-old chicks flying on the
Hawaiian Islands, but he says they stayed at the island another 2 to
Host of the chick's activities before it starts flying seem to
be related to feeding and temperature maintenance. Obviously the adults
help with the latter when the chicks are snail, but 3- to 4;-week-old
chicks start assuming their juvenal plurage and are largely independent.
When placed in open-sunlight, their body temperatures rise to the
lower range of black bulb temperature and then level off (Howell and
Bartholonew, 1962). Chicks nay help stabilize body temperature by
panting, facing away from the sun, drooping the wings, and erecting
the back feathers, much as adults do when it is hot. A few of r.y
tethered chicks died in open sun when about this age, apparently from
heat stress. Thus body temperature must closely approach lethality
when chicks are exposed to open sun.
Juveniles seen to leave the colony soon after they learn to fly,
as there seldon are many flying juveniles at Bush Key at any given
tine. As they do not feed near Bush Key, little is known of their
activities. In all probability, the adults stay with the young and
continue to feed it until it can capture food by itself (Robertson,
1964; Burckhalter, 1969) as do adult Royal Terns (Asnzcle and Tovar,
1968), Caspian Terns (.ydrcr;rocne casoia) (Jozefik, 1969), and Elegant
Terns (Thalasseus ele-ans) (Monroe, 1956). Bush Key adults cannot
feed flying Sooty Tern young nore than about 2 months though, as the
juveniles leave the range of adults by about October (Robertson, 1969).
This certainly nust be a difficult time for young Sooties as they must
learn to capture food while on the wing, a skill requiring good
eyesight, coordination, and timing,
Other Aerial Activities
Sooty Terns at Bush Key commonly soar in dense circling cclu,:s
during midday hours on hot days. Possibly at these times updraits or
thermals develoD near Bush Key. The columns approach 200 feet in
diameter and sometimes go from just above sea level to several hundred
feet in altitude, although most birds are usually below 200 feAt.
Within the columns the birds circle as they climb and then soar off to
join another column. Most cormon late in the breeding season, these
columns of soaring birds are taller than the circling flocks seen
earlier, and the birds also appear to fly faster than in the early
season flocks. Usually the terns are silent as they soar but
occasionally one gives a "id-ik" call.
Soaring often continues for several hours although the composition
of the flock changes as birds join or leave it. The cclunns often
move laterally and two nay merge into a single colu-mn. MTagnificent
Frigatebirds (Frgat-, rna.ificens) and occasionally Brown Pelicans
(Pelecanus occdentalis) and Laughing Gulls (Larus atricilla)
(Harrington, pers. comm.) join these flocks, but I did not see Brown
Noddies do so.
A rather common activity in Sooty Tern colonies, especially when
chicks are present, is for one or more adults to fly straight up 1 to
5 or more feet, give a loud "wide-a--ake" call at the peak, and drop
back to the ground. I call these flights "fly-ups," and I believe they
are a type of social behavior, often used in greeting, Besides the
birds that actually fly, other birds around then often raise their
wings vertically over the body in a flight intention movement, but do
not fly. These fly-ups usually involve a small group of terns. During
a fly-up, other nearby birds become very active and noisy for 15 to
60 seconds and then gradually quiet down.
During incubation, I saw few fly-ups. They usually occurred
when an adult arrived at the colony to assume care of the egg. Other
adults near the arriving tern's scrape might fly up, seemingly in
excitement over the arrival of another bird at the colony. Burckhalter
(1969) also saw this group activity when an adult arrived at the colony.
Once the chicks were present, I often saw fly-ups when an adult
landed to feed a chick or had started feeding it. Again several adults
flew up in the air in a sudden burst of activity and then quieted
Fly-ups were rather common when the chicks got fairly large. Then
I often saw adults fly up when a chick ran out from cover and approached
them. As the chick approached, the adult flew up, called and then
landed to feed the chick. In these instances the chick seemed to
stimulate the fjy--ip. Several times I saw a chick run past several
adults, each one flying up in turn as the chick approached, until the
chick finally cane up to an adult that fed it. Thus the adults
seemed to react as if the chick was their own while the chick went by
and approached another adult, presumably its parent, to be fed.
Sooty Terns exhibit two distinct types of disturbance flights,
cormcnly called panics. Of these, dreads apparently include the flights
Palmer (1941) calls dreams and panics for Commcon Terns w-'le those
termed alarms are similar in both species.
Alarn.--Wrhen an intruder openly approaches the colony, the birds
stand upright with the neck and head stretched vertically. As the
intruder cones closer they fly, calling as they do so, and then circle
and hover over the intruder until it leaves the vicinity of the
scrape. Unlike many other terns, Sooties seldom actually strike an
intruder, but rather hover near it or dive at it without striking it.
I did see Scoties attack Cattle Egrets (Eubulcus ibis) and once a
Purple Gallinule (Porohyrala nartinica) that had broken a tern egg.
Tnen the Sooty landed on the gallinule and jabbed it. The fewtie
a Sooty Tern struck me occurred about the time the eggs were hatching.
A Icud, long alarm call, usually a do-;nward inflected "kee aa"
or "kerr aa" often precedes the alarms. Host alarms are auite local,
involving relatively few birds in the colony. Occasionally though,
they spread throughout the colony and virtually all of the terns
fly. Alarms seem to start front an intrusion that is not sudden,
but anticipated for a short tine. Th'e circling and calling by Scoties
during rain are probably just a form of the alanm. Other disturbances
are caused b1y los tangible factors such as a ioud sound (e.g., sonic
boom, boat whistle) cr a sudden nover.ent, These elicited a different
response the dread.
Dread.--In dreads the terns suddenly become silent and fly rapidly
from the colony to the water, darting and swooping as a unit silently
down low over the water. At the end of the swoop, they rise up,
start calling loudly, and gradually drift back to the colony. Thus if
a bird is on the ground when the dread starts, it flies rapidly out
over the water, and those birds that are already in the air suddenly
swoop out over the water. This seens to be a high intensity form of
One final, poorly defined alarm reaction is that in which virtually
the entire colony flies up from the ground, either at once or, more
often, in a gradually spreading group from one end of the island to
the other. In the air the birds call loudly and drift out over the
water and then gradually move back to the colony. This delayed alarm
seens to start as an alan_ in one part of the colony, and as those
birds fly, they scare up birds near them and so on until the whole
colony is in the air.
Interactions with Other Scecies
As they nest on isolated islands, Sooty Terns come in contact with
relatively few vertebrate predators. Those predators that reach the
nesting colony find eggs and chicks plentiful and vulnerable.
Adult Sooty Terns are fast and shifty in flight, and probably
few other birds are capable of capturing a flying adult. Certainly the
most serious natural predator on Bush Key is the Peregrine Falcon
(Falco nererinus). Cne and occasionally two peregrines were present
at the Dry Tortugas from 8 to 11 A-ril 1963. I saw a peregrine saoop
at Sooty Terns 23 tines without capturing a bird, but I did find the
decapitated remains of four Sooties that the falcon had almost
certainly killed. r.ave seen peregrines capture incubating Sooty
Terns on Little Tobago in the southern 'iest Indies.
On 15 May 1968 a Purple Gallinule cracked an egg and ate the
contents until a Sooty attacked and drove it away. In June, 1969,
Ruddy Turnstones (A~raia i. rrt ) broke open and ate the contents
of several unattended eggs front which I had frightened the adult Sooty
Tern. allyly the adult terns sit tight if a turrstone approaches,
but turnstones can cause considerable mortality a:ong unattended eggs.
The Magnificent Frigatebird is perhaps the greatest threat to
Sooty Tern chicks. Frigatebird predation must vary considerably fro.
year to year as in 1968 1 saw one capture a Sooty chick only once
(3 June) while other years they have taken many chicks (Beard, 1939;
Sprunt, 1948). 'When frigatebirds do prey on tern chicks, it sees to
be a prey preference of only a few individuals rather than of all the
frigatebirds present (Robertson, pers. commn.). On 24 May 1968 a
frigatebird chased a Sooty Tern adult. They may occasionally steal
food front the terns. Ashnole (1963) reports ?reqata acuila taking many
young Sooty Terrs on Ascension Island and F. mincr does the same on
Christmas Island (Ashrnole and Ashnole, 1967).
Frigatebirds usually take only small chicks, mainly in open parts
of the island where they can not find escape cover. Thus the relatively
thick vegetation on Bush Key in 1968 probably discouraged frigatebird
Predation by rats (Rattus rattus) also must vary from year to
year. I saw no evidence of predation by them in 1968 but in other
years they have been a serious predator (Russell, 1938) on Bush Key.
Several times an Osprey (Pandion haliaetus) circled over Bush Key
but usually the terns ignored it. Once several terns mobbed an Osprey.
A few other times the birds panicked, but it may not have been due to
the Osprey's presence. The only other hawk I have seen cause any alarm
was a Broad-6inged Hawk (Bute-o rlaty/rteris) that circled Bush Key on
16 June 1969 and apparently started several panics. Other hawks
migrate through the Dry Tortugas and occasionally may disturb the terns.
Cattle Egrets occasionally take a Sooty Tern egg or chick on
Bush Key (Robertson, pers. cor n.) but more commonly just disturb them.
Ridley and Percy (1958) report that Cattle Egrets often take Sooty Tern
eggs and chicks on the Seychelles. Of the Cattle Egrets that migrate
through the Dry Tortugas, those that linger generally are unable to
find enough food to maintain their flying strength and eventually
starve to death. This shortage of food apparently forces them to
feed on almost anything they can find as they often take small migrant
passerines (Cunnincghan, 1965; pers. obs. ). They also feed on fish the
terns drop and on insects, disturbing the terns as they feed.
Typically as an egret approaches a tern with an egg or chick, the
tern faces the egret and gives a series of low, hoarse "wuk wuk" notes.
As the egret comes closer the tern raises up off its egg or chick,
erects the feathers on the top of the head and continues to direct
"wuk" calls at the egret, the calls becoming lcuder and nore rapid and
often changing to a rapid "ka ka ka" call. If the egret comes within
a few feet of the tern, the tern flies from the scrape and circles
over the egret until it departs. A few tines a tern stretched its
head and bill forward, raised the wings almost vertically over the
back, and ran at the egret. Sooties scmetines dive at an egret but
they seldom actually strike it.
Cattle Egrets often fly low over nesting terns, eliciting a long,
drawn out "kaaa" note from the terns and sometimes starting a panic.
The Sooties often chase egrets in the air and dive at then, mobbing
them more corronly ir flight than when the egret is on the ground.
Brown : oddies frequently join the Sooties in these attacks on Cattle
Egrets and the I.oddies are much more aggressive, striking then on the
ground and in the air, and chasing then farther than the Sooties do.
Cattle Egrets harn Sooty Terns mainly by disturbing then and
keeping the adults away from the egg and chick, which can be fatal to
the young tern. The Great white e Heron (Ardea cccidentlis) has been
reported preying on Sooty Terns on the Tortug.s (Robertson, 1962).
A few Herring (Larus argentatus), Ring-billed (L. delawarensis),
and Laughing Gulls frequently linger around Bush Key. Gulls cclmmonly
prey on terns (Hatch, 1970), but I never saw one enter the colony at
Bush Key to prey on tern chicks or eggs, although they could do so
easily. Watson and Lashley (1915) saw Laughing Gulls prey on Sooties
on Bird Key.
Adult Sooties nay peck and kill chicks that wander away from
their scrape and can be a serious cause of mortality.
Ghost crabs (Ocnode cuadrata) occasionally prey on tern chicks.
A few times a tern gave a "wuk" note as a crab scuttled by or pecked at
it but usually they ignored the crab. They also seemed to ignore the
hermit crabs (Coenobita clvneatus) that probably feed only on dead
chicks and broken eggs.
Although most Sooty Tern colonies are on relatively isolated
islands, man reaches many colonies at least occasionally and adversely
affects nesting success. Tern eggs are still collected for human
consumption at some colonies (Cott, 1954; Ridley and Percy, 1958), but
National Park Service protection has curtailed egging at Bush Key.
It is much more difficult to measure mortality caused by human
activity in the colony, but at times such activity must be very
harmful. Every time I entered or left my blind, I disturbed the birds.
Although they soon habituated to this and rapidly returned to the scrape
once I was out of sight, I still caused mortality, both by stepping
on eggs and by separating chicks from their parents. Such mortality
is inevitable whenever a human works extensively in a Sooty Tern colony.
Other disturbances such as sonic boons, boat whistles, and low
flying airplanes may cause the birds suddenly to leave the island in a
dread. However I did not see any of these disturbances keen the birds
away from their scrape for more than a few minutes unless they were
The interrelations of Sooty Terns and Brown Noddies on Bush Key
are interesting. These two species are abundant and widespread in the
tropics and nest together on many islands. On Euzh Key the Brown
Noddies nest prin rily along the edge of the bay cedar, but also in
other low vegetation and occasionally on the ground. Although some
Sooties nest under the bay cedar, they always nest on the ground and
hence are usually vertically segregated from zhe ihoddy nests. Most
Sooties nest in the open flat parts of the island outside the bay
cedar, again separating then from the Noddies. Some Noddies nest in
low vegetation within inches of Sooty scrapes. I often saw young of
both species in such situations, so apparently they can breed
successfully in close proximity.
Early in the nesting season Noddies often searched for nest
material on my- study plot. In all 35 instances of direct supplanting
between these species that I saw on r.y study plot: the Sooty drove off
the Noddy. In two of these a Sooty chick drove an adult Noddy off the
plot. Generally the Sooty stretched its head and neck forward, held the
wings either tight against its sides or slightly away from the body,
and advanced toward the Noddy until the lioddy retreated or flew. I
never saw a oddy], fight back or withstand the attack long. Cnce I
heard a Sooty give a low growling "urr" note as it advanced at a Noddy
but usually both were silent. The few tines I watched for the reverse
encounters. I saw hoddies drive off Sooties that had come too close
to NIoddy nests.
Noddies arrive at Bush Key at about the same time as Sooties,
but begin work on their nest almost at once rather than having a
pronounced period of flocking as Sooties.do. In 1968 Noddies started
laying somewhat after Sooties (10 April) and as their incubation
period is longer than the Sooties' (35 to 36 days, Thompson, 1903),
Noddy eggs hatched later than the Sooties' eggs. Young Noddies
mature faster than Sooties and many were flying by late June, well
before most Sooty chicks.
Both species feed in a similar manner, dipping down to capture
food at the ocean's surface, although :Noddies occasionally plunge
to the surface or rest on the water to capture food. The Ashmoles
(1967) show that Sooties and Hoddies have very similar diets on
Christmas Island, but the Noddies seem to feed much closer to the
island. They also note that the wing of the Noddy, broader than
that of Sooties, perhaps gives it greater maneuverability and allows
it to forage more efficiently near the island, whereas the narrower
wing of Sooties enables them to forage farther from their colonies.
At the Tortu.as Noddies often feed within sight of Garden Key,
dipping down and capturing food at the ocean's surface. Noddies
have incubation shifts of 30 minutes to 5 hours, much shorter than
Scoties. They feed their chick every 2 to 4 hours (Tjatson, 1908),
slightly less often than Sooties. The fact that Sooties virtually
never feed within sight of Bush Key and the somewhat different foraging
times, especially during incubation, suggests that the two feed in
different areas, much as they do on Christmas Island. The two species
apparently feed in different zones at other colonies in the Pacific
and Indian Oceans (Ashmole and Ashmole, 1967: 65-66).
The few hundred Roseate Terns at the Dry Tortugas generally nest
in habitat quite different from that used by the Sooties. In recent
years they have nested in the coral rubble on Long Key and on the open
sand on some of the other islands. When they have nested on Bush Key,
it usually has been at the far east end away from most Sooties.
Roseates usually arrive later than Sooties and in 1968 did not start
nesting until micd-ay when Sooty eggs were hatching. The incubation
and fledging periods of Roseates are much shorter than those of Sooties,
so young Roseate Terns were flying well before any Sooty Tern chicks.
Roseate Terns often forage within sight of Fort Jefferson and probably
avoid competition with Scoties by foraging in different areas. They
dive from the air and plunge into the water, thus capturing food that
is unavailable to the Scoties.
Although Sooty Terns nest throughout Bush Key, they nest in
greatest concentration outside the bay cedar thickets and mangroves
of the central cart of the island. I set up five plots outside the
bay cedar, each 5 yards square, and at 2- to 6-day intervals marked
all the eggs laid on each. I usually visited the plots in late
afternoon. As the eggs hatched, I banded the chicks in each. To
avoid mortality, I did not fence in the plots, and thus I missed many
chicks that fled as I approached. Hurricane Abby caused no appreciable
mortality on these plots.
The density of eggs on each plot is summarized in Table 7.
?lot 1 was close to the beach and had a thick cover of Osuntia, 'Uniola.
and Cakile. Plot 2 was mainly low grass with some Oruntia on one
Table 7. .,est density and success in fiva 5-yard-square 1p ts on
3ush Key in 1968
No. eggs iinim no.
Mo. eggs/ broken or chicks known Dead chicks
Plct no. *o. eggs sq yd infertile to hatch found
1 75 3.00 18 29 1
2 161 6. 21 91 5
3 62 2.8 13 33 1
4 28 1.12 2 13 2
5 131 5.21 18 90 20
Totals 457 3.66 72 256
side and about 30 per cent bare ground. The first eggs on Bush Key
in 1968 were laid on this plot. Plot 3 had a thick cover of Soorobolus
and spurge (E' rba bxi fol gi ) 6 to 12 inches tall. The main cover
on plot 4, 4- to 6-foot tall Uniola, probably made it difficult for
terns to drop to the scattered openings on the ground. Plot 5 had
scattered Euchcrbia and Cakile and numerous bare spots. The highest
nest densities were on plots 2 and 5, both of which had scattered bare
areas. This seems to be a major factor in determining nest sites for
Sooty Terns, as bare sand is avcided. The terns apparently require
some features such as plants or rocks as reference points for locating
the scrape and. as cover for the -hick. Ashmlole (1963) noted the same
thing on Ascension Island. The other three lots all. had a very thick
ground cover in which adults probably could not dig an adequate scrape.
Mry data on density of eggs are somewhat higher than those used by
others for determining Sooty population densities. For instance,
Sprunt (1948) used densities ranging from 0.5 to 5 per square yard
depending on the thickness of the vegetation, but he counted birds
only once. Sooties arrive over an extended period and fill in space
as some eggs are broken or deserted. My counts, made over an extended
period, included these additional eggs and thus increased the estimate.
Ashrmole (1963) also used 5-yard squares and counted over an extended
period. His maximum density of 132 eggs in a plot is somewhat less
than my maximuon of 161.
My data provide a rough estimate of the nesting success of Sooties
at Bush Key. it. least 256 of 457 eggs hatched, and probably nearly
385 hatched (457 minus 72 known broken or infertile). Thus somewhere
between 56.9 and ?4.2 per cent of the eggs laid hatched. As scme pairs
undoubtedly relaid (I often found a fresh egg next to one of the
deserted ones), perhaps 85 to 90 per cent of the breeding pairs on
Bush Key produced a chick. Rid-Ley and Percy (1958) report that
82.5 per cent of the eggs at the Seychelles hatch.
Chick mortality is much harder to measure. Some chicks ran into
nearby thick clumps of ODuntia and Suriana and I was unable to band
many of them. Of the 256 that I know hatched, I found 29 dead, a
known mortality of 11.3 per cent. Chick mortality was undoubtedly
higher as I could not follow chicks until they were flying and some
older, more mobile chicks certainly died without my knowing it. As
a rough estimate of total mortality, if 385 of the 457 eggs hatched
and,of these, 11.3 per cent died, 341 out of 457 would be the maximum
number surviving, a survivorship of about 75 per cent. This of course
assumes that all eggs that I did not find broken or deserted did
hatch and also underestimates mortality of chicks. This is not far
from the 70 per cent survivorship Robertson (1964) gives as typical
for Bush Key for summers without hurricanes.
My presence on the plots undoubtedly increased mortality of eggs
and of chicks, and I see no way that it can be avoided if one works
with the birds intensively. Some chicks that ran from me probably did
not find their way back to their parent and starved or were necked
severely by adults and died from the blows. Repeated disturbance
also caused desertion of some eggs.
For example, I have good records of the number of eggs laid on
my study plot and the number of these that hatched. Of 122 eggs,
93 (76.2 per cent) hatched, somewhat less than the max:dium figure
arrived at on the other five less frequently visited plots, but
certainly not inconsistent with it in consideration of the variables
in that information. Pally half the eggs that did not hatch were those
closest to my blind, the ones that I disturbed repeatedly.
Ste tenacity, the tendency for a bird to return to the sane nest
site year after year, occurs in the Co.-mon Tern (Austin, 19;9). As I
color-banded 182 Sooties at one location on Bush Key in 1968, later
sightings of these provide sone information on site tenacity. I later
saw 106 of the 182 (58.2 per cent) in the general area of banding,
and most of then probably nested in that general part of the colony.
From 28 April to 2 May 1970, 0. L. Austin, Jr. sat for several
hours each day near the plot where the terns had been banded to watch
for and record color-banded birds. He saw at least 100 birds with
color bands on the plot or in.-ediately adjacent to it. I saw three
more in June and July, and three others were caught in other parts of
Bush Key. Of the 103 birds on the plot, 76 could be identified
individually; the others had lost some of the bands and were not
caught to read the FJS band. Of the 76 individually recognizable
Scoties, 53 (69.7 per cent) were among the 106 that had been seen on
the plot in 1963 after banding. Thus, of 106 color-narked birds that
probably nested on or near the study plot in 1968, at least half cf
them (53 of 106) were on the sc.ae plot 2 years later, The 30 birds
that could not be identified individually may well have included birds
that nested on the plot in 1968. Thus 50 per cent is a mniinun
figure and perhaps 70 per cent is a more meaningful estimate of site
tenacity. These data suggest that individual birds had a strong
tendency to nest in the same parzs of Bush Key in 1965 and 1970. As
ny plot and the ground innediately around it covered much less than
1 per cent of the habitat suitable for Sooties on Bush Key, it seems
unlikely that by chance alone so many color-banded birds would return
to this same area in 1970.
Site tenacity implies individual attachment to a specific locale
in the colony. Group adherence implies that subgroups exist within the
colony and these subgroups stay together and individually recognize
other members of the group. Austin (1951) describes such. subgroups in
Cormaon Terns and they nay well exist in Sooty Tern colonies. Young
Sooties, if not allowed to return to their scrape, establish a spatial
arrangement with respect to other chicks similar to the pattern that
existed when they were at the home scrape (Burckhalter, 1969). This
implies that they recognize other chicks around them and act as a group.
The fly-ups already described also suggest that subgroups exist in
Sooty colonies. In these, adjacent birds react to the activities of
one bird, seemingly as if they recognized one another.
Subgroups that move together within the colony could also explain
the clustered nesting of color-banded birds seen at the plot in 1970,
but this implies a fairly sizeable group of perhaps several hundred
birds functioning as a subgroup. Although I definitely think subgroups
exist in Sooty colonies, site tenacity remains the more likely
explanation for birds nesting in the same general part of the colony in
two different years. If individuals repeatedly nested in the sane
part of the colony, they wculd of necessity contact the same birds
year after year and then possibly forn subgroups.
Two broader aspects of Sooty Tern behavior are particularly
important and warrant further discussion. First is the comparison of
the behavioral repertoire of fuscata to that of other species of terns,
especially those within the genus Sterna. This should clarify the
a-ffirities Sooties have with other terns. Fortunately, several other
terns have been studied carefully enough to provide a basis for
Equally important are the adaptations of fuscata that have allowed
it to survive and thrive in a pelagic environment. 3y feeding in
pelagic wa-ers, Sooties differ considerably from most other terns and
in many ways closely resemble pelagic species of birds in other orders
such as the Frocellariifornes.
Behavioral comparisons can be made with several other er.merate
Sterna terns, the White Tern (Gvis alba), the two Anous. several
species of Thalasseus, and the Black Tern (Chlidcnias nier).
The high flights of the Cormron, Arctic, Roseate, and Sandwich
Terns are all quite similar (Cullen, 1960a). The high flights of the
Caspian Tern and Black Tern resemble those of the others, although in
niger many birds may ascend together. The circling ascent and gliding
descent of the Sooty Tern high flight are almost identical to thosa of
the Common and Arctic Tern described by Cullen (196Ca). In the high
flight of the Brown Noddy the ascent is somewhat like the upward
flutter cf Sooties but the birds descend in a circling glide rather
than a long, essentially straight glide as the Sterna terns do
(Moynihan, 1962; pers. obs.). The high flight of GW-gis is similar
to that of the dark Noddies (Moynihan, 1962).
The low flight is poorly developed in fuscata and not nearly so
common as in other terns except Gygis (Dorward, 1963). Cullen (1960a)
notes that the low flight has fewer special features and may have
developed independently in each species. The rarity of the low flight
by fuasata may be due to its different manner of carrying food.
Cullen (1960a) notes that food also has little importance in
advertising by unrated birds of both species of Anous, both of which
car r food in the esophagus as Sooties do.
Perhaps carrying food in the esophagus allows Sooties to transport
more food (Ashumole and Ashnole, 1967) than if it were carried in the
bill. Also it :ay prevent food desiccation during the of-en lengthy
return flight to the colony. The food is the major water source for
the chicks, so noist food may be necessary for the chick's survival.
Moist food should be easier to handle and swallow than dried-out food.
Besides Sooties, _. an.thetus and both Anous carry fish in their
The ground displays of Sooties seen similar to those reported for
other Sterna terns, all having some form of a parade display. In
particular the parade of Sooties seems to resemble that of the Corzon
Tern described by Palmer (1941). Incipient scrape-building activities
are part of courtship in Sooties, just as they are in Common Terns
The most conspicuous ground display of the dark Noddies is nodding,
and generally their behavior is gull-like (Moynihan, 1962). Nodding
emphasizes head movements, whereas the parade emphasizes the position
of the wings and head, providing a clear contrast between the Noddies
and the Sterna terns. The displays of i are similar to those of
the dark Noddies, but are very simplified (Moynihan, 1962).
Comparisons of behavior help clarify the relationships of various
terns. Moynihan (1959) used comparative behavior as a basis for
classifying the Laridae and put all terns into three genera. Larosterra
remains monotypic, Anous includes GYRgs and Procelsterna, and the rest
are placed in Sterna. I prefer the older terxiinclog, remaining
numerous genera, but nonetheless agree with the group limits he has
established on the basis of behavior. In Moynihan's classification, the
typical black-capped terns include fuscata, hir-undo, oaradisaea, and
others. He puts the large Thalasseus terns and the small Sterna terns
like albifrons in separate groups within his large genus Sterna and
also keeps the marsh terns (Chlidonias) together as a separate group
This grouping agrees well with my observations. Thre high flight
of fuscata is almost identical to that of hirundo and caradisaea. The
ground displays of fuscata are very similar to those of hirundo and on
behavioral grounds the two species clearly are closely related. In
contrast the displays of G~yis and both species of Anous, three species
that share the tropical oceans with Sooties, are clearly different from
those of Scoties.
Moynihan (1959) points out, and I agree, that a most profitable
area for future work is with the many old world and tropical forns that
are only slightly known. Two other tropical terns, Sterna lunata and
anaethetus, resemble fuscata in their feeding habits and probably their
breeding biology, although they may feed more offshore than pelagically.
As yet these two are poorly kncwn and further wcrk on them certainly
would be of interest.
Sooties as a Pelagic Soecies
Of the 40 species of terns (Sterninae), most feed either on
freshwater marshes, lakes, and streams, or along marine coasts and
estuaries. A few venture into offshore waters (on continental shelves
or near islands), especially in the nonbreeding season. Sooty Terns
seem to be the most pelagic (beyond the continental shelf) of terns,
although several other Sterna, both species of Anous, and 'yis feed
in offshore and pelagic waters. Apparently by exploiting pelagic
waters and breeding on isolated islands, Sooty Terns utilize a niche
available to few other species. Sooties are highly adapted to exploit
this niche, particularly in regard to flying ability, population
structure, and breeding biology.
Probably foremost among their adaptations for exploiting pelagic
waters is flying ability. Sooty Terns rapidly become waterlogged when
placed on water (Watson and Lashley, 1915) and apparently seldom rest
on water. As Socties inhabit the open oceans during the nonbreeding
season, birds from some colonies must fly continuously for 6 months
or more. Virtually nothing is known of the physiological adaptations
involved in flying for such long periods. For example, do they sleep
on the wing as the wift (Anus a/t) apparently does (Lack, 1956), or
do they go without any sleep at all?
Their flying ability allows Scoties to exploit food in a large
area far from land, reached by few other birds. Several factors
restrict Sooty Tern food availability. Except for some areas where
upw-ellings or convergences occur, tropical pelagic waters are poor
in nutrients and plankton that forn the base of their food chain
(Paymont, 1963). Sooties apparently are restricted to feeding on food
items that occur aT, the very surface of the ocean, or in the air above
it. This food is driven to the surface mainly by large predacious
fish (Ashnole and Ashmole, 1967: 58) and thus is not evenly distributed
throughout the ocean.
The length of incubation shifts and brooding spells varies
considerably between different Sooty Tern colonies. Adults usually
shift at 1 or 2 day intervals on Bush Key whereas they vary from
2 hours to 3 days in the Seychelles (Ridley and Percy, 1958), and are
as long as 6.5 days on Ascension Island (Ashmole, 1963), and 7 days on
Christmas Island (Ashnole and Ashmole, 1967). The duration of the
shift must depend on the time required to fly to the feeding ground,
find enough food to last until the adult can feed again, and then return
to the colony. Unfortunately feeding rates and exact locations of
feeding grounds for various colonies are not known. The most
reasonable explanation for the great variation between colonies in
time spent foraging is that food is farther away or scarcer for some
colonies than others.
More puzzling is the change in length of spells once the egg
hatches and the chick must be fed. At Ascension Island adults change
from 5.5-day-incubation shifts to feeding their chick about once a
day (Ashnole, 1963). As the breeding cycle at that colony is nonannual
(9.5 months), it seems unlikely that a regular change in abundance of
food could occur and thus account for the differences in foraging
between incubation and fledging periods. On Bush Key adults change
from 1 cr 2 day incubation shifts to feeding the chick several times
a day. As the breeding cycle there is annual, perhaps food is more
abundant, closer to the island, or the adults forage more diligently
when chicks are present. The meager evidence available indicates that
terns do not tend to capture smaller fish when they are feeding chicks
as might be expected (Potthoff and Richards, 1970) but little is known
about the food supply. As the parent forages not only for itself
but also for the chick in a shorter period of time, it probably works
harder when feeding the chick than during incubation. Put another
way, after sitting quietly on the egg for several days, adults probably
loaf at least part of the time during breaks from incubation.
Table 8 gives some information on the breeding biology of several
representative tern species. I have selected one species each from the
genera Chlidonias, Thalasseus, and Anous and only a few from Serna,
picking species for which information is available and that seem to be
representative of other similar species. This table is similar to one
in Lack (1968: 262). The patterns in Table 8 vary both between various
colonies of a species and from year to year within a colony, but this
does not negate the comparisons to be made here. The purpose of the
table is to try to compare some general features of the breeding
biology of marsh and coastal feeding terns with those that feed
offshore or pelagically.
On Bush Key as at other Sooty Tern colonies, occasionally a scrape
contains tw-o eg.s, but by far the usual clutch is one egg. Single egg
clutches are ccmnon among marine birds (Lack, 1968) and seem to imply
Table 8. Some information on the breeding biology of several species of terns
Feeding Clutch Incubation Fledging carried
Species habitat size period period exposed? Reference
Black Tern Marsh 2-3 20-22 20-24 Yes Goodwin, 1960
Commion Tern Coastal 2-4 21-26 30 Yes i.almer, 1941
S:oty Tern Pelagic 1 29-30 ca. 60 No This study
Least Torn Coastal 2-- 19-22 28 Yes Witherby et al., 1941; Hardy, 1957
Sani(.rich Tern Coastal 1-2 20-24 35 Yes VWitherby et al., 1941
Brc n iJoddy Offshore, pelagic 1 35-36 42+ No Thompson, 1903; Docuard and
W ite Tern Offshore, pelagic 1 36 60-95 Yesb 3Drward, 1963; Ashmole, 1968
a. In days
b. Also carries food in esophagus
that something about the oceanic habitat limits their clutch size.
In contrast, coastal and freshwater feeding terns commonly lay
more than one egg (Table 8). Besides fuscgaa, only Anous, Gynis, and
a few other Sterna terns (e.g. S. anaethetus and lunata) have a normal
clutch of one. A clutch of two or more presumably is primitive in
Sooties and Brown Noddies, as they both have two brood patches. Gygis
lays one egg and has only a single brood patch, but its unusual
nesting habits may explain this (Dorward, 1963).
With the smaller clutch size, pelagic terns must have either a
high nesting success, a long life span, or some combination of the two
to maintain their populations. With a clutch of one, partial nesting
success does not exist. Either they fledge a young or they fail.
mkong Sooties, a naximnm of 50 per cent renest if the first attempt
fails and the percentage declines as the breeding season advances
(Ridley and Percy, 1958; Ashnole, 1963).
The single egg of fuscata is incubated longer and the chick is
cared for longer than are those of inshore feeding terns (Table 8).
As Sooty Tern chicks may not be fed at regular intervals, a lengthened
period of development probably allows them to withstand the occasional
short periods of fasting (Lack, 1968). Chicks grow rapidly when food
is available, but can survive long periods without food if the adults
cannot find any for several days (Ashmole, 1963). Storms and a distant
food supply guarantee that short fasts occur regularly at some colonies.
Feeding rates also differ between Sooties and coastal feeding
terns. The latter generally forage close to their colony and, although
they may raise more than one chick, they feed their young frequently.
On the Farne Islands, Cormon, Arctic, and Sandwich Terns all feed each
of their young about once ar hour (Pearson, 1968) and Black Terns
feed young as often as 15 times in an hour (Goodwin, 1960). As a
result their young fledge much faster than do young Sooties. The
differences between Sooties and ioddies, both of which raise one chick
per pair, have already been discussed. Noddies forage closer to the
colony, feed their young somewhat more frequently, and their young
mature faster than the farther ranging Sooties (Table 8).
An important factor favoring the slow development of young Sooties
is the scarcity of predators on islands where they breed. They can
"afford" long incubation and fledging periods without drastically
raising the risk of predation. Common and Least Terns nesting on or
close to the mainland are constantly exposed to predation (Austin, 1948;
Hardy, 1957) so it is probably advantageous for them to raise their
young as quickly as possible. In turn rapid growth of their young is
made possible by a food source close to the colony that permits more
feeding trips in a day. Cat predation on Ascension Island shows
vividly what can happen to seabirds, including Sooties, when a land
predator is introduced (Stonehouse, 1962; Ashnole, 1963).
Sooty Terns defecate near their egg, do not always remove the empty
egg shell, and nest in dense colonies. These habits seem related to
the isolated and usually land--maial-free islands where they nest.
Sandwich Terns have similar habits, but frequently nest with gulls that
help drives off intruders (Cullen, 1960b). In contrast, Arctic and
Common Terns carry off the empty egg shell, defecate away from the
nest, space their nests farther apart, and vigorously attack intruders,
all of which should help reduce predation. Common and Arctic Terns
also have much shorter incubatLon and fledging periods than Sooties,
again reducing the time they are vulnerable to predation.
With all the problems that Sooties encounter in feeding far from
their nesting colony, they still are very plentiful. Obviously the
number of islands, suitable for nesting, within their range is finite.
On some islands nan has greatly reduced tern numbers, either by
egging (Ridley and Percy, 1958), or by introducing predators (Ashrtole,
Sooty Terns probably do not first breed until 6 years old (Robertson,
1969) whereas most Conmon Terns first breed when 3 or 4 years old
(Austin and Austin, 1956). Delayed maturity occurs in many other
seabirds, including many Larids (Lack, 1968). Robertson (1969)
suggests that the transatlantic migration by juvenile Sooties from
Bush Key, a colony relatively stable in numbers, keeps them from
competing with older and presumably more efficient breeders for several
years. The 3-year-old birds that visit Bush Key late in the breeding
season have little chance of finding a suitable nest site and breeding
successfully. By returning to the colony earlier in successive years,
eventually they should be able to arrive early enough to find a
suitable nest site and breed successfully. Thus delayed maturity
keeps younger birds from competing with older, established breeders
for nest sites, and also keeps the younger birds frcm undergoing the
rigors of reproduction until they have a fairly high probability nf
success. The Fishers (1969) note a staggered arrival cf age groups
of Laysan Albatrosses, with the young birds not breeding for several
For delayed maturity to persist, individuals with s:ch maturity
must raise nore offspring in the long ran than if they first bred
when younger. As sore 3G-year-old- Scoty Terns are presently on Bush
Key, the adults there clearly have many breeding opportunities. Still
Sooty Terns, by delaying breeding for several years and then by
raising at most a single young per breeding season, have a lc-wer
madxinu natality than other similar terns that feed close to their
Lack (1968) proposes that birds raise as many offspring as possible
with the usual clutch size being that that results in the nost
breeding adults in the next generation. According to his ideas, Sooty
Terns with a clutch size of one canr.ct raise more than one young per
breeding season. Presumably the limiting factor is their distant
and sometimes unreliable food supply.
Certainly the hypothesis fits the information from Ascension
Island where an apparent failure of the food supply one season resulted
in few chicks surviving (Ashnole, 1963). This and the long spells
that Ascension adults spend foraging suggest that at nost a pair could
raise one chick. Other colonies that have been studied have not had
Except for the cats on Ascension Island, adult Sooty Terns suffer
little predation and have a low annual adult mortality, perhaps
around 12 to 18 per cent at Bush Key (Austin, pers. ccnm, ), versus
25 per cent in adult Comrmon Terns (Austin and Austin, 1956). Although
the food supply ay limit the numbers of young they can raise and hence
determine the usual clutch size of Sooty Terns, perhaps delayed
maturity is a function of the low predation on and high survivorship of
the adults. High survivorship of adults could result in many birds
competing for a limited number cf nest sites, but delayed maturity
could reduce this competition as has already been discussed.
Sone pelagic species can raise more than one young (see Harris,
1970), indicating that food is not entirely limiting, but so far the
problem has been following the survival of these "twins" until they
return to breed.
Thus Sooty Terns differ fron most other terns. Many of these
differences derive from their adaptations to a pelagic environment.
The basic behavioral patterns in rair formation and courtship show
clearly that Sooties are closely related to typical Sterna terns such
I believe several behavioral differences between Sooties and most
other species of Sterna are due to their different way of carrying
food. Sooties do not have a well-developed low flight display, a
display in w-hich other terns often prominently carry a fish in their
bill. Other terns also may carry a fish openly in the bill during the
high flight while Sooties do not, although they do have a well-
developed high flight.
The high cost of obtaining food, compared to that of coastal
feeding terns, perhaps explains why courtship feeding is relatively
rare in Sooty Terns. A coastal feeding tern can easily replace the
food it feeds to another bird in courtship. For a Sooty Tern,
replacement would require a long flight.
Comparison of the breeding biolong of Sooties and coastal feeding
terns shows other differences. The development of the Sooty Tern egg
and chick is much slower than that of most other terns, probably sc
their young can' survive lengthy periods when the parents are unable
to find food. The high survivorship of adults and the distant food
supply have led to their delayed natur.ty and small clutch size.
Thus in their breeding biology, with a long incubation and
fledging period, delayed maturity, and a one-egg-clutch, Sooty Terns
resemble other pelagic birds such as the albatrosses, shearwaters,
and tropicbirds more closely than most other terns.
Ash.role, L/. 1963. The biology of the Wideawak-e or Sooty Tern
Sterns. iuscata on Ascension Island. Ibis, 13b'o: 297-360 .
Ashmole, :,'. 1968. Breeding and molt in the ie it Tern (C/vr^is al-'a)
on Christmas Island, Pacific Ocean. Condor, 70: 35-55.
Ashr:cle, .?',, and h.J. Ashmole. 1967. Comparative fee dng ecology
of sea birds of a tropical oceanic island. Bull. Peabody .Ius.
Nat. Hrfi. 24: 1-131.
Ashrocle, e ?, and H. Tcvar 3. 1968, Prolonged parental care in
Royal Trms and other birds. Auk, 85: 90-100.
Audubon, J.cJ. 1841t. The birds of America, vol. 7. 1967 reprint.
ow Yor-k, Dover Publ., Inc.
Austin, O.L. 19h8. Predation by the co.-.on rat (3jitt ncrvei.cus)
in the Cape Cod colonies of nesting terns. ird-Banding,
Austin, O.L, 1949. Site tenacity, a behaviour trai t of the Ccrnon
Tern (Sterna hir-.no Linn.). Bird-3anding, 20: 1-39.
Austin, 0.L, 1951. Group adherence in the Common Tern.
Bird-?3~ading, 22: 1-15.
Austin, O.L., ,and C.L. Austin, Jr. 1956. Some demographic aspects
of the Care Cod pcpulaticn of Common Terns (Stena hl r:r,).
Bir-ar.ding, 27: 55-66.
3eard, D.B. 1939. ..a!-o'-'.:ai~--birds prey on eastern S.ot.y ern .
Auk, 576 32?-32r.
Bryns, W.F.J.:M., and K. H. Vocus. 1965. Uight-feeding by; Sooty
Terns (Sterna f'iscata). Ardea, 53: 79.
3urckhalter, D.L. 196. Orientation and communication of Sooty Tern
chicks. H. S. thesis. Tuscon, Arizona, Univ. of Arizcna.
Chapin, J. P. 1954. The calendar of wi deaake fair. Auk, 71: 1-1.5.
Cott, H.3. 1954. The exploitation of wild birds for their eggs.
Ibis, 96: 129-149.
Cullen, J.M. 1960a. The aerial display of the Arctic Tern and other
species. Ardea, 48: 1-37.
Cullen, J.M. .1960b. Some adaptations in the nesting behaviour of
terns. Proc. 12th Intern. Ornithol. Congr., 153-157.
Cullen, J.iM., and N.P. Ashnole. 1963. The Black 'Icddy Anous
tenuiroQjsj on Ascension Island. Part 2. Behavicur.
Ibis, 103b: 4i23-24O6.
Cunningham, R. L. 1965. Predation on birds by the Cattle Egret.
Auk, 82: 502--503.
Davies, S.J.J..F., and R. Carrick. 1962. On the ability of Crested
Terns, Ste.ja berfii, to recognize their own chicks. Australian
J. Zool., 10: 171-177.
Do.rard, D. F. 1963. The Fair- Tern Gis _nba on Ascension Island.
Ibis, 103b: 365-37.
Dorwnard, D. ., and H. P. .sh-:cle. 1963. Notes on the biolovg of the
Brown ,oddy Anous stolidus on Ascension Island. Ibis, 103b:
Duellrm.n, .i.3,, and A. Schwartz. 1958. Amphibians and reptiles of
Southern Florida. Eull. Florida State :Ius., 3: 179-324.
2rdar, D.S. 1967. Sea birds in relate n to gane fish schools off