Content uploaded by John C Ogden
Author content
All content in this area was uploaded by John C Ogden on Jun 26, 2015
Content may be subject to copyright.
Movements, Foraging Groups, and Diurnal Migratons of the Striped Parrotfish Scarus
Croicensis Bloch (Scaridae)
Author(s): John C. Ogden and Nancy S. Buckman
Source:
Ecology,
Vol. 54, No. 3 (May, 1973), pp. 589-596
Published by: Ecological Society of America
Stable URL: http://www.jstor.org/stable/1935344 .
Accessed: 23/05/2014 17:10
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
.
Ecological Society of America is collaborating with JSTOR to digitize, preserve and extend access to Ecology.
http://www.jstor.org
This content downloaded from 131.247.112.3 on Fri, 23 May 2014 17:10:13 PM
All use subject to JSTOR Terms and Conditions
MOVEMENTS, FORAGING GROUPS, AND DIURNAL MIGRATIONS
OF THE STRIPED PARROTFISH SCARUS CROlCENSIS
BLOCH (SCARIDAE) 1
JOHN C. OGDEN AND NANCY S. BUCKMAN2
Smithsonian Tropical Research Institute, Balboa, Canal Zone
Abstract. A total of 374 striped parrotfish (Scarus croicensis) were tagged from the reefs
surrounding Isla Pico Feo on the Caribbean coast of Panama. Many of these fish were fol-
lowed individually in the field for up to 3 months. Three different behavioral categories were
recognized: stationary, territorial, and foraging. Fish tend to aggregate in foraging groups
in a particular pattern relative to reef structure and have a predictable set of associated
species. Transfer experiments showed that striped parrotfish
have strong ties to a home reef.
Striped parrotfish migrate diurnally from shallow-water feeding areas to deeper nocturnal
resting areas along constant pathways. Direct counts of fish on migration pathways provided
information on the structure and size of the striped parrotfish population at Pico Feo. The
numbers of male striped parrotfish
may regulate the phenomenon of sex reversal in the scarids.
INTRODUCTION
Parrotfishes (Scaridae) are among the most com-
mon and colorful of fishes inhabiting Caribbean coral
reefs. They are generally classed as herbivores, biting
off algae and sediment-covered chunks of coral rock
substrate with their beak-like teeth, and grinding the
hard calcareous material into fine sand with their
pharyngeal mill (Bdhlke and Chaplin 1968, Randall
1968). Their constant rasping of the bottom greatly
influences the settlement and growth of benthic algae
and invertebrates in shallow tropical water (Bakus
1969, Bardach 1961, Cloud 1959, Hiatt and Stras-
burg 1960, Ogden 1973, Randall 1961a, 1965).
The striped parrotfish, Scarus croicensis Bloch, is
the smallest and most abundant Caribbean scarid. It
is present in virtually all reef habitats, but it is most
common in protected situations away from direct
wave action. Like other members of the genus, it is
sexually dimorphic with two color phases. These color
phases were originally described as separate species
but have since been shown to be male and female
of the same species. (Randall 1963, Reinboth 1968.
Rosenblatt and Hobson 1969, Schultz 1958, Winn
and Bardach 1957). The striped phase can be male
or female, and spawns in groups, whereas the color-
ful terminal-phase male is derived from a sexually
reversed female (proterogynous sex reversal) and
usually spawns with a single female (Randall 1963.
Randall and Randall 1963, Winn and Bardach 1957,
1960). The colored phase is the larger, attaining a
maximum length of about 20 cm. Like other scarids,
the striped parrotfish is inactive at night, resting in a
mucous cocoon within cavities in coral (Stark and
Davis 1967, Winn 1955, pers. obs.). This report tells
' Manuscript received November 11, 1971; accepted
September 7, 1972.
2 Present address: West Indies Laboratory, P. 0. An-
nex Box 4010, Christiansted,
St. Croix, U. S. Virgin Is-
lands 00820.
the results of an extensive marking and observation
program that establishes basic characteristics of the
behavior and movements of the striped parrotfish on
the reef.
STUDY AREA AND METHODS
This study was made on the reefs off Isla Pico Feo
in the San Blas archipelago (90 33' 10" N; 780 58'
30" W) from May 1970 through February 1971.
The island is on the Caribbean coast of Panama
within the Gulf of San Blas and in the lee of San
Blas Point. The protected inner reefs of this region
are subject to low wave action and high sedimenta-
tion and occasional high surface dilution during the
wet season from the influence of several large rivers
opening into the Gulf. Most of this study was done
on the reefs immediately surrounding Pico Feo, but
SAN SLAS POINT PATCH
REEF
Q PORITES
POINT
IC O UOUSE REEF
~JI RSTR
X 0 ~~~REEF MOQUITO IS.
MNGROVE/
MANGROVEt PT.~
REEF ~~GULF
OF
CURRENT SAN BLAS
-iw- ACROPORA,
PORITES,
AGARICIA loom
v THALASSIA
Q LOCATIONS
OF FORAGING
GROUPS
FiG. 1. Map of Isla Pico Feo, San Blas, and vicinity
showing the distribution of reefs, Thalassia grass beds,
and locations of foraging groups of striped parrotfish.
This content downloaded from 131.247.112.3 on Fri, 23 May 2014 17:10:13 PM
All use subject to JSTOR Terms and Conditions
590 J. C. OGDEN AND N. S. BUCKMAN Ecology, Vol. 54, No. 3
neighboring reefs within a 1-mile radius were also
regularly visited (Fig. 1).
The reefs of the Gulf all share certain character-
istics. The Gulf is edged with mangrove and shallow
water with a bottom covered with turtle grass (Tha-
lassia) and scattered coral heads. As the depth in-
creases, there is a definite zone composed of finger
coral (Porites furcata), staghorn coral (Acropora
cervicornis), or elkhorn coral (A. palmata). Below
this zone is a slope composed mainly of the stony
coral Agaricia agaricites (forma danae). The slope
continues to a maximum depth of 40 m in the deep-
est channels, but usually 15 m. The shallow channels
between patch reefs are covered with Thalassia grass
and silty sand.
Parrotfish were collected in fish traps made from
?.-inch mesh galvanized wire and baited with coco-
nut, broken white dinner plates, and occasionally
bread; the white material served as an attractant,
drawing many reef fishes. We tagged the trapped fish
with numbered modified Peterson disc tags sewn
through the dorsal musculature. Parrotfish scales are
large, and inevitably several would be pulled out
during tagging. Healing of these wounds was rapid
and damage to the fishes was minimal. The tags re-
mained readable in the field for up to 2 months
before they became obscured by growths of algae.
Re-trapping was then necessary to clean the tags.
As most of the activities of these fish are concen-
trated in the top 3 m of water, we needed only snor-
kel gear to follow them.
Tagging and movements
A total of 374 fish were tagged from the reef sur-
rounding Pico Feo. Of these 261 received numbered
tags and could be distinguished individually in the
field. Between 20 and 40% of the tags originally
placed were still in place after 1 month (Table 1).
Losses were due to (a) a certain mortality of fish
in the tagging process, especially among small (7 cm
and less) fish of the striped phase; (b) differential
predation of tagged fish; and (c) tag removal by the
fish scraping its sides against the bottom. Several fish
were seen after tagging with large dorsal wounds
caused by a tag pulling out. These wounds caused no
particular difficulty and served as recognition char-
acters until completely healed. Several fish, having
lost tags, were tagged repeatedly during the study.
By keeping track of the movements of individual
fish on the reef, we could distinguish several different
behavioral categories. One category, including both
striped-phase fish and terminal-phase males, was sta-
tionary, being found both singly and in small groups
in the same general area day after day. The stationary
fish did not defend the area against intruders, but
rather appeared to use the area, which varied in size
from several m2 up to 50 M2, for feeding. A second
TABLE 1. Durability of tags
Number
of fish % of tags
originally observed
tagged I week I month
2 weeks
Tag S TPM S TPM S TPM S TPM
White 20 9 75 78 55 67 40 22
Orange 27 23 44 83 30 65 22 26
S = Striped-phase fish
TPM = Terminal-phase males
category was territorial (Buckman and Ogden 1973).
The territorial forms were either striped-phase fish or
terminal-phase males and actively defended an area,
ave-aging 10 to 12 M2, against conspecifics. The ter-
ritory contained at least one dominant female with a
particular display of yellow pelvic fins used during
defense of the territory. Also usually present on the
territory was a terminal-phase male or, uncommonly,
a striped-phase male which appeared to be paired
with the female, and one to several subordinate fe-
males. A third category accumulated in foraging
groups which moved about the reef all day. These
foraging groups will be fully discussed later.
The distribution of tagged fish followed daily for
11/2 months is shown in Table 2. Most of the fish
remained in their capture area day after day. The
table does not distinguish between territorial and
stationary fish, but almost all of these are stationary.
Territorial forms must be observed directly. Table 3
shows the distribution of fish that were captured on
one occasion while in a foraging group. The majority
of these fish persisted in the group day after day.
However, because the group moved about the reef
while feeding, many of the fish were present in the
group only when it was in their particular area, and
when the group moved on, they remained behind.
This accounts for the occasional sightings (Table 2)
of stationary fish in foraging groups. The territorial
forms did not participate in foraging groups, but
attacked them vigorously whenever they were present
in the territory.
Of all of the fish tagged and released at Pico Feo,
none was ever seen off of the continuous reef plat-
form (approximately 1/2 ha.) that surrounds the
southeast portion of the island. This reef is separated
from other reefs in the vicinity by channels about
3 m deep, covered with silty sand and Thalassia.
Although striped parrotfish from reefs bordering a
channel often ventured out a short distance to feed
in the grass, they would immediately return to the
shelter of their home reef if disturbed.
To check the strength of this restriction to a home
reef, we trapped, tagged, and released fish about
100 m distant on a reef other than their home reef.
The home reef was then monitored to note the re-
turn of transferred fish. Three experiments were
This content downloaded from 131.247.112.3 on Fri, 23 May 2014 17:10:13 PM
All use subject to JSTOR Terms and Conditions
Late Spring 1973 MOVEMENTS AND FORAGING GROUPS OF SCARUS 591
TABLE 2. Distribution of tagged fish-Isla Pico Feo
Number of times
Location Term of obser- Number of times Major location observed
in
Fish number trapped vation (days) observed observed
(times) foraging
group
Striped-phase
fish
31 Rear House 21 7 Sand Trap (4) 2
33 Rear House 29 6 Rear House (6)
36 Rear House 1 3 Rear House (2)
38 Rear House 43 15 Sand Trap (10) 5
52 Rear House 33 11 Rear House (2) 9
56 Front Wall 2 2 Front Wall (2)
57 Front Wall 22 6 Front Wall (6)
Terminal-phase males
29 Rear House 23 9 Rear House (6)
30 Rear House 1 3 Rear House (3)
32 Rear House 21 7 Rear House (6)
34 Rear House 4 6 Rear House (5)
50 Rear House 8 7 Rear House (7)
55 Front Wall 6 4 Front Wall (4)
TABLE 3. Distribution of tagged fish trapped in foraging group-Isla Pico Feo
Term of obser- Number of times Number of times Other location
Fish number Location vation (days) observed observed
in group observed
(times)
Striped-phase
fish
39 Rear House 44 10 1 Cistern (7)
40 Rear House 4 5 Sand Trap (4)
41 Rear House 44 18 14 Route 3 (3)
44 Rear House 19 7 6 Kitchen Pocket (1)
45 Rear House 27 10 3 Rear House (4)
46 Rear House 31 19 10 Front Wall (7)
47 Rear House 4 5 3 Front Wall (1)
48 Rear House 22 20 8 Sand Trap (9)
Terminal-phase males
54 Rear House 36 13 7 Front Wall (3)
carried out (Table 4) but only the first two are re-
ciprocal. All of the transferred terminal-phase males
returned to their home reef, most within 1 day. The
return in most cases probably occurred within a few
minutes. On several occasions terminal-phase males
were trapped on House Reef, released on Patch Reef,
and followed. Generally, the fish swam about for a
few minutes and then swam straight out across the
Thalassia-covered channel separating the two reefs
and returned to their home reef. In the second trans-
fer experiment only 4 striped fish out of 15 survivors
returned to their home reef, but of the 11 remaining
on House Reef, 5 were regularly seen at Porites
Point, the tip of House Reef closest to Patch Reef
(Fig. 1). In the third transfer experiment from House
Reef to Airstrip Reef, terminal-phase males again
showed a strong tendency to home.
Foraging groups
Foraging groups consisting of up to 500 striped
parrotfish and other species in lesser numbers were
characteristically present on certain reefs. Groups
formed by late morning and moved within a specific
area, averaging ?/2 ha., each day. The six foraging
groups shown in Fig. 1 were present along the mar-
TABLE 4. Summary of transfer experiments
Number of fish Number of fish Number of fish
tagged and observed after returning to
transferred tagging home reef
Transfer S TPM S TPM S TPM
House Reef to
Patch Reef 12 3 8 3 4 3
Patch Reef to
House Reef 19 5 15 5 4 5
House Reef to
Airstrip Reef 6 6 - - 1 4
S = Striped-phase fish
TPM = Terminal-phase males
gin of San Blas Point near Pico Feo throughout
most of this study. Groups were generally located in
I to 3 m of water over a bottom composed of coral
rubble, sandy sediment, and patches of Acropora,
Porites, Agaricia, and scattered coral heads, These
locations bordered a deep channel with a predom-
inating current to the open sea. Each of the locations
can be considered an "island" of coral surrounded
by open areas of sandy bottom and Thalassia which
may act as barriers to extensive interchange of fish
between separate areas of coral. This may account
for the relatively even spacing of the foraging groups
This content downloaded from 131.247.112.3 on Fri, 23 May 2014 17:10:13 PM
All use subject to JSTOR Terms and Conditions
592 J. C. OGDEN AND N. S. BUCKMAN Ecology, Vol. 54, No. 3
TABLE 5. Foraging groups of striped parrotfish
Striped Doctor- Blue Barred Spotted
Date Location parrotfish- fish tang hamlet goatfish Other
S TPM
Jan. 3 Rocky Pt. 150b -- 5 1 bluehead
June 3 Mang. Is. 150b - 5 3 foureye butterflyfish
June 8 Mang. Is. 200b 10 2 3 - foureye butterflyfish
bluehead
June 10 Pico Feo 50 3 - 1
June 11 Pico Feo 150 15 3 1
June 15 Pico Feo 300 11 5 - - 2 yellow goatfish
July 31 Pico Feo 190 10 10 2 1 1
Aug. 2 Pico Feo 250 25 15 - 3 1
Aug. 3 Pico Feo 200 21 15 2 - 1
Sept. 1 Pico Feo 100 - 3 1 1 - Bermuda chub
Oct. 1 S. Cays - 75 50
Oct. 3 Pico Feo 200 10 12 3 2 1 Sparisoma
sp.
Nov. 6 Pico Feo 70 9 - -- -
Nov. 24 Hooka Rf. 150 100" 3 - 2
Feb. 24 Mang. Is. 100 100 5 - 2
Feb. 24 Mang. Pt. - 100
Feb. 24 Hooka Rf. 300 150 10 - 3 Scarus sp.
aThese counts all approximate. bNo distinction between S and TPM. eContained many transforming terminal-phase
males.
along the reef margin. Groups were never observed
on the scattered small patch reefs surrounding Pico
Feo. There may be a minimum area necessary to
support a population of striped parrotfish large
enough to form groups.
Other species participating in foraging groups
formed a characteristic assemblage (Table 5). Be-
sides the parrotfish, the major species in the groups
were the doctorfish (Acanthurus chirurgus), the blue
tang (A. coeruleus), the spotted goatfish (Pseudu-
peneus maculatus), and the barred hamlet (Hypo-
plectrus paella). The two acanthurids are herbiv-
orous, while the goatfish and the hamlet are car-
nivorous, the former feeding on invertebrates stirred
out of the sandy bottom, and the latter feeding on
invertebrates and small fishes. All of these species
generally stayed with the group as it moved. Other
fish occasionally joined a foraging group as it moved
within their particular habitat or territory. These
include the bluehead (Thalassoma bifasciatum), the
foureye butterflyfish (Chaetodon capistratus), the
yellow goatfish (Mulloidichthys martinicus), the
night sergeant (Abudefduf taurus), the Bermuda
chub (Kyphosus sectatrix), and, less frequently, sev-
eral species of the scarid genera, Sparisoma and
Scarus.
The striped parrotfish in a foraging group were
either male or female fish of the striped phase and
usually lesser numbers of terminal-phase males (Table
5). We periodically speared fish from groups and dis-
sected them, often finding ripe males of striped phase,
but rarely any ripe females or ripe terminal-phase
males. Except for one occasion (Table 3), a group
was never trapped or poisoned, and the sex ratio of
striped-phase
fish in foraging groups must remain
speculative.
The feeding of the parrotfishes
in the group was
intense, with all members biting or stirring the sub-
strate. Although small pieces of filamentous and en-
crusting algae were taken, the main part of the diet
consisted of the sediments that covered the bottom
to a depth of several mm. Thus, the fish processed
great quantities of sediments and were partly re-
sponsible
for the scoured appearance
of open bottom
in these protected inner reefs. The group moved
within a particular
area of the reef, but no obvious
pattern of movement was noted. As it moved, the
group often hovered above the bottom until a few
individuals
developed
a common directional
tendency;
then the whole group followed, and after moving a
short distance descended
to the bottom to feed.
As the foraging group passed through the terri-
tories of the dusky damselfish (Eupomacentrus
fus-
cus) and the striped parrotfish,
it was attacked,
with
the attacking fish frequently making contact with
members
of the group. Often a single dusky damsel-
fish turned a whole group by advancing on it side-
ways while blanching about the head. This continual
harassment
did not scatter the group, but tended to
keep it moving. With the close approach of larger
fishes, the group moved more rapidly;
if attacked, all
the fishes raced for cover. We observed predatory
attacks by Nassau grouper (Epinephelus striatus),
cero (Scomberomorus regalis), and bar jacks (Ca-
ranx ruber).
The cohesion of the group correlated with light
level. On bright days with clear water the group was
closely organized in a tight ball. On cloudy days or
when the water was turbid,
the group was more scat-
This content downloaded from 131.247.112.3 on Fri, 23 May 2014 17:10:13 PM
All use subject to JSTOR Terms and Conditions
Late Spring 1973 MOVEMENTS AND FORAGING GROUPS OF SCAR US 593
tered and often did not form at all. As night fell the
group fragmented little by little and individuals
from
groups joined with other fish and migrated off the
shallow reef platform
to deeper water.
Diurnal migration
Diurnal migration occurred along specific path-
ways. At Pico Feo, where we studied migration in
greatest detail, we called the pathways Routes 1, 2,
and 3 (Fig. 2). There was a characteristic and con-
sistent segregation of types of fishes on these routes.
Route 1 was used mainly by striped-phase
fish of
adult size (7 cm and larger), with a few terminal-
phase males. Near Route 1, Route 2 was used by
juvenile fish (less than 7cm). Route 2 joined with
Route 1 at a depth of about 4 m and the fish ap-
peared to share the same resting area at about 15 m.
Route 3 was separated from the first two and was
used predominately by terminal-phase males, with
some striped-phase fish. All of the routes followed
coral cover and none went across open sand or Tha-
lassia grass. When a fish stopped feeding and started
to migrate, it swam fairly directly to a pathway and
off the shallow reef platform stopping on occasion
to feed or at well-established
cleaning stations to be
cleaned by juvenile blueheads (Thalassoma bifasci-
atum) or cleaning gobies (Gobiosoma sp.).
Because the pathways were consistent we could
get an accurate count of the numbers
of fish migrat-
ing by floating on the surface and counting passing
fish (Table 6). Morning and evening migration lasted
about 2 hours, correlated with light level. The eve-
ning migration
ended approximately with sunset and
the morning migration began with sunrise. On dark
U
133P.. S \ I
_~~~~~~~1 =
FIG. 2. Schematic diagram of the reef (to 2 m con-
tour) at Isla Pico Feo showing diurnal migration routes
of striped parrotfish.
TABLE 6. Diurnal Migration counts-Isla Pico Feo
Total Total
Total striped- terminal-
Date Route Time fish phase fish phase males
July 26 1 1645-1820 1440 1374 66
July 28 1 0600-0800 2443a 2380 75
Aug. 12 1 1645-1830 1145 1090 55
Sept. 2 1 1630-1825 1109 1032 77
Feb. 20 1 1630-1830 847 812 35
Aug. 9 3 1630-1830 706 436 270
Aug. 10 3 1645-1830 699 414 285
Sept. 1 3 1635-1820 626 344 282
Oct. 28 3 1615-1755 651 382 269
Nov. 8 3 1600-1800 462 216 246
Feb. 19 3 1600-1830 333 207 126
Feb. 22 3 1600-1830 356 189 167
Sept. 8 2 1615-1810 1472 1472 --
aThis high count is due to the joining of Routes 1 and 2 during the
morning migration.
days the two migrations began later and earlier re-
spectively. On Routes 1 and 3 the striped fish were
the first to migrate
in the evening, with the terminal-
phase males at the rear. In the morning the situation
was reversed, with the terminal-phase
males migrat-
ing first. Over the short term the number of fish using
a particular
route was quite consistent, especially in
the case of terminal-phase males on Route 1 and 3.
The consistency of the counts and the separation of
the color phases suggested that the same fish were
using the same route in the evening and returning
on that route in the morning. For this reason we
kept a close watch for tagged fish during migration
counts. It was often difficult to see the tags because
the fish remained in sight for only a few seconds
before disappearing into deeper water, but at least
some of the same tagged fish were observed on the
same routes on consecutive migrations.
After migration, fishes accumulated in deeper
water over patches of Agaricia agaricites (forma
danae), which grew in convoluted masses over the
silty bottom. Several night dives were made into this
area to look for parrotfish.
In spite of careful search-
ing, none were found, but this was not surprising,
in view of the many cavities in this type of coral
growth, and the stirring of the silt when the coral
was disturbed.
Dives made during the evening migra-
tion disclosed many parrotfish hovering above the
coral before entering for the night. Fish kept in
aquaria formed mucous cocoons and rested very
close together against
the glass wall or under a piece
of rock.
DIsCUSSION
Several investigators
have followed the movements
of fishes on coral reefs by trapping and tagging, but
these studies have depended upon re-trapping
to as-
say movements (Bardach 1958, Randall 1961b,
Springer
and McErlean 1962). Striped parrotfish
are
easily tagged and small enough to allow the observer
This content downloaded from 131.247.112.3 on Fri, 23 May 2014 17:10:13 PM
All use subject to JSTOR Terms and Conditions
594 J. C. OGDEN AND N. S. BUCKMAN Ecology, Vol. 54, No. 3
to follow them closely in the field. Thus, day-to-day
records of the movement and behavior of individual
fish can be compiled.
Three behavioral categories of striped parrotfish
have been recognized: stationary, foraging, and ter-
ritorial. The tagging work reported in this paper in-
dicated that fish retain the characteristics of a cat-
egory for at least 1 to 2 months, and in several cases
territorial fish were followed almost daily for 4
months (Buckman and Ogden 1973). The presence
of three distinct behavioral categories within a single
species is unusual and difficult to explain. The be-
havioral category of stationary is relatively undiffer-
entiated and involves males and females of all phases,
while that of territorial is highly differentiated and
the dominant territorial fish is a female. In foraging
groups primary phase and terminal-phase males ap-
pear to predominate but unripe females are some-
times also present. On several occasions (Table 5),
we observed groups composed entirely of terminal-
phase males; foraging groups may be a primary char-
acteristic of males. These categories could be part of
an ontogenetic series in the striped parrotfish, but no
correlation with size was ever noted; indeed, every
category had some fully adult sized fish. More likely,
the categories are determined by such factors as
aggressiveness and other behavior correlates which
could be tied intimately to sex reversal in the scarids.
Until more behavioral and ecological information is
obtained concerning sex reversal, the origin and de-
velopment of various behavioral categories must re-
main speculative.
In foraging groups of striped parrotfish there is a
consistent mixed-species association. The four asso-
ciated species (Table 5) are not always present in
any single group at any given time, but they do ap-
pear in groups regularly. There may be at least two
reasons for this assemblage. First, the acanthurids
and scarids are herbivorous and their association
appears to be one of several species doing the same
thing-substrate feeding. Acanthurids of larger size
frequently form large groups in outer reef areas,
often accompanied by a smaller number of scarids.
The group affords some protection from predators,
as it has been observed to behave like a school when
under attack. The two other species often seen in
groups, the spotted goatfish and the barred hamlet,
are carnivorous and associate with the group prob-
ably for the additional benefit of feeding. The ham-
lets adopt a characteristic nose-up attitude near the
center of a milling group, and their gut contents in-
clude larger invertebrates and small fishes, which are
picked off as they are stirred out of hiding places.
The spotted goatfish probe the bottom near the group
with their barbels, picking up smaller invertebrates
living within the sediment layer. The other species
(Table 5) are mainly carnivorous and associate
loosely with foraging groups for a similar combined
benefit of predator
protection and feeding (see Hob-
son 1968 for further discussion of associations of
fishes).
Striped
parrotfish that are scattered about the reef
by day migrate
off of the shallow reef platform each
evening and return along the same pathways at sun-
rise the next day. For some fish, migration
involves a
swim of several hundred m from feeding areas to
sleeping areas. Other reef fishes also migrate. These
include common representatives
of hamlets (Serrani-
dae), goatfishes
(Mullidae), butterflyfishes and angel-
fishes (Chaetodontidae), damselfishes (Pomacentri-
dae), parrotfishes (Scaridae), surgeonfishes (Acan-
thuridae), and probably others. There are several
possible reasons for diurnal migration. Many of the
suitable sleeping sites in shallow water are defended
by the small but highly aggressive dusky damselfish
(Eupomacentrus fuscus), which we have often ob-
served driving off other species seeking shelter in
coral areas which it defends. The heavy growths of
the coral Agaricia agaricites (forma danae) provid-
ing excellent cover are concentrated off the reef plat-
form in deeper water. Finally, predators
working the
shallow water at dusk and sunrise
may also influence
the movements
of diurnally
feeding reef fishes (Hob-
son 1965, 1968).
Migration usually takes place with falling light
level at the end of the day, but fish do move off of
the reef early during the afternoon for spawning.
Group spawning and pair spawning were seen often
during the study. We found no evidence to support
the contention of Randall and Randall (1963) that
most of the spawning of the striped parrotfish
is pair
spawning. The trapping program indicated fluctu-
ations in the presence of ripe females on the reef.
Very likely spawning occurs throughout
the year with
peak periods each month. A similar
pattern
has been
found in the yellowtail parrotfish (Sparisoma rubri-
pinne) (Randall and Randall 1963).
One of the most interesting
aspects of the migra-
tion is its possible use by population biologists as an
accurate census tool to detect changes in population
size and structure over time. Almost all of the striped
parrotfish
migrate. The exceptions are a few of the
territorial forms that rest in the territory
if a suitable
cavity is available and very small juvenile fish that
rest in Agaricia coral near the edge of the reef plat-
form. By summing the counts made on 3 consecu-
tive days, an accurate total can be obtained (Table
7). Note that of the total about 10% are terminal-
phase males and close to 50% are juveniles (less than
7 cm long). Given the difficulty
in obtaining a total
of very small fish, and the fact that some striped-
phase fish rest in their territories,
this total is most
accurate
for terminal-phase
males.
This content downloaded from 131.247.112.3 on Fri, 23 May 2014 17:10:13 PM
All use subject to JSTOR Terms and Conditions
Late Spring 1973 MOVEMENTS AND FORAGING GROUPS OF SCARUS 595
TABLE 7. Total number of striped parrotfish-Isla Pico
Feo
Total Total
Date of Total striped- terminal-
count Route fish phase fish phase males
Sept. 1 3 626 344 282
Sept. 2 1 1109 1032 77
Sept. 8 2 1472 1472
Total 3207 2848 359
In the migration counts made at Pico Feo (Table
6), the numbers of terminal-phase males did not
fluctuate as greatly as did those of striped-phase fish.
Terminal-phase males are larger, and probably live
longer than striped fish, but the consistency of the
counts is a salient point. Perhaps there is a tightly
regulated carrying capacity for terminal-phase males
on a given reef, and the removal of one will lead to
the rapid transformation of a female to a terminal-
phase male. The stimulus regulating transformation
is not known, but transformation of coloration from
the striped-phase to the colored terminal-phase male
can occur within 5 days when a female is injected
intramuscularly every other day with one mg of tes-
tosterone propionate. The relative ease with which
counts can be made of migrations in this species
opens the way for field experimentation on the ecol-
ogy of sex reversal in fishes. This problem has
received considerable attention recently, but is diffi-
cult to approach experimentally (Ghiselin 1969,
Reinboth 1962, 1968, Robertson 1972, Smith 1967).
Many terminal-phase males could be removed from
a reef by trapping or spearing, and subsequent events
could be watched closely in the field. Alternately,
large-scale artificial transformations of females could
be undertaken using testosterone and these fish could
be followed individually in the field.
The numbers of fish on the reef at Pico Feo (Table
6) decreased strikingly after January 1971. The most
dramatic environmental change during this period
was the onset of the dry season. There is some ev-
idence that populations of striped parrotfish within
the Gulf of San Blas are sediment limited. The dry
season brings a much lowered level of sedimentation
on the reefs of the Gulf, and the bottom in shallow
feeding areas during this season appears much more
scoured than during the wet season. The lack of food
may be one direct cause of the population decline.
ACKNOWLEDGMENTS
This study was supported by a Visiting Research Fel-
lowship from the Smithsonian Tropical Research Insti-
tute (S.T.R.I.). The latter 3 months of the study were
supported by a grant from the American Philosophical
Society. We are indebted to Dr. Ira Rubinoff and Dr.
Martin Moynihan of S.T.R.I. for their hospitality and
encouragement. Dr. Peter Glynn of S.T.R.I. and Dr.
John McCosker aided in many aspects of the field work
and provided stimulating discussion. Mr. John Mann
encouraged our field work on Isla Pico Feo in many
ways and was patient in showing us the delightful tempo
of life in the San Blas. Sr. Jose Gracia kept our outboard
motor in running condition and provided us with essen-
tial services. Drs. Jeff Graham and Eric Todd read and
criticized the manuscript. This paper is Contribution
Number 13 of the West Indies Laboratory, Fairleigh
Dickinson University, St. Croix, U.S. Virgin Islands.
LITERATURE CITED
Bakus, G. J. 1969. Energetics and feeding in shallow
marine waters. Int. Rev. Gen. Exp. Zool. 4: 275-369.
Bardach, J. E. 1958. On the movements of certain Ber-
muda reef fishes. Ecology 39: 139-146.
. 1961. Transport of calcareous fragments by reef
fishes. Science 133: 98-99.
Bbhlke, J. E., and C. G. Chaplin. 1968. Fishes of the
Bahamas. Livingston Publ. Co., Wynnewood, Pa.
771 p.
Buckman, N. S., and J. C. Ogden. 1973. Territorial be-
havior of the striped parrotfish
Scarus croicensis Bloch
(Scaridae). Ecology (in press).
Cloud, P. E., Jr. 1959. Geology of Saipan, Marianas Is.
Part 4. Submarine topography and shoal water ecology.
U.S. Dept. Interior, Geol. Survey Professional Paper
280-K: 361-445.
Ghiselin, M. T. 1969. The evolution of hermaphroditism
among animals. Quart. Rev. Biol. 44: 189-208.
Hiatt, R. W., and D. W. Strasburg. 1960. Ecological re-
lationships of the fish fauna on coral reefs on the
Marshall Is. Ecol. Monogr. 30: 65-127.
Hobson, E. S. 1965. Diurnal-nocturnal activity of some
inshore fishes in the Gulf of California. Copeia 3:
291-302.
. 1968. Predatory behavior of some shore fishes
in the Gulf of California. Bur. Sport Fisheries and
Wildlife, Research Rep. 73: 92 p.
Ogden, J. C. 1973. Behavior and ecology of the striped
parrotfish Scarus croicensis (Bloch) Am. Phil. Soc.
Year Book (in press).
Randall, J. E. 1961a. Overgrazing of algae by herbiv-
orous marine fishes. Ecology 42: 812.
. 1961 b. Tagging reef fishes in the Virgin Is.
Proc. Gulf Caribb. Fish. Inst. (14th session): 201-
241. . 1963. Notes on the systematics of parrotfishes
(Scaridae) with special emphasis on sexual dichroma-
tism. Copeia 2: 225-237.
. 1965. Grazing effects on sea grasses by herbiv-
orous reef fishes in the West Indies. Ecology 46: 255-
260. . 1968. Caribbean reef fishes. T.F.H. Publications,
318 p.
Randall, J. E., and H. A. Randall. 1963. The spawning
and early development of the Atlantic parrotfish Spa-
risoma rubripinne, with notes on other scarid and la-
brid fishes. Zoologica (N.Y.) 48: 49-60.
Reinboth, R. 1962. Morphologische und funktionelle
Zweigeschlechtlichkeit bei marinen Teleostiern (Ser-
ranidae, Sparidae, Centricanthidae, Labridae). Zool.
Jahrb., Abt. Allg. Zool. Physiol. Tiere 69: 405-480.
. 1968. Protogynie bei papageifischen (Scaridae).
Z. Naturforsch. Teil B 23b (6): 852-855.
Robertson, D. R. 1972. Social control of sex reversal in
a coral reef fish. Science 177: 1007-1009.
Rosenblatt, R. H., and E. S. Hobson. 1969. Parrotfishes
This content downloaded from 131.247.112.3 on Fri, 23 May 2014 17:10:13 PM
All use subject to JSTOR Terms and Conditions
596 J. C. OGDEN AND N. S. BUCKMAN Ecology, Vol. 54, No. 3
(Scaridae) of the Eastern Pacific, with a generic re-
arrangement of the Scarinae. Copeia 3: 434-453.
Schultz, L. P. 1958. Review of the parrotfishes, family
Scaridae. U.S. Nat. Mus. Bull. 214: 1-143.
Smith, C. L. 1967. Contribution
to a theory of hermaph-
roditism. J. Theoret. Biol. 17: 76-90.
Springer, V. G., and A. J. McErlean. 1962. A study of
the behavior of some tagged South Florida coral reef
fishes. Am. Midl. Nat. 67: 386-397.
Stark, W. A. II, and W. P. Davis. 1967. Night habits of
fishes of Alligator Reef, Florida. Ichthyologica 38:
313-356.
Winn, H. E. 1955. Formation of a mucous envelope at
night by parrotfishes. Zoologica (N.Y.) 40: 145-148.
Winn, H. E., and J. E. Bardach. 1957. Behavior, sexual
dichromatism
and species of parrotfishes.
Science 125:
885-886.
. 1960. Some aspects of the comparative biology
of parrotfishes at Bermuda. Zoologica (N.Y.) 45:
29-34.
This content downloaded from 131.247.112.3 on Fri, 23 May 2014 17:10:13 PM
All use subject to JSTOR Terms and Conditions
