Experimental Zoogeography of Islands: The Colonization of Empty Islands
Daniel S. Simberloff; Edward O. Wilson
Ecology, Vol. 50, No. 2. (Mar., 1969), pp. 278-296.
Stable URL:
http://links.jstor.org/sici?sici=0012-9658%28196903%2950%3A2%3C278%3AEZOITC%3E2.0.CO%3B2-U
Ecology is currently published by Ecological Society of America.
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained
prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in
the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/journals/esa.html.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic
journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers,
and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take
advantage of advances in technology. For more information regarding JSTOR, please contact support@jstor.org.
http://www.jstor.org
Thu Mar 20 12:54:54 2008
278 DANIEI. S . SIMBERLOFF AND EDWARD O. WILSON Ecology, Vol. 50, No. 2
ACKNOWLEDGMENTS Hermannsson, S. 1967. Introduction. Surtsey Re-

We thank the IJnited States Department of the Interior search Progress Report, 111: 1. 

for permission to fumigate the islands of this experiment. LaRue, D. C., and T. T. Muzik. 1954. Growth, reIn
particular, Jack Watson of the Great White Heron generation and precocious rooting in Rhizophora manNational
Wildlife Refuge, Roger W. Allin and William gle. Pap. Mich. Acad. Sci. Arts Lett. (1) 39: 9-29.
B. Robertson of the Everglades National Park, and Lauck, D. R. 1965. Chloropicrin for fast action with
Robert M. Linn of the National Park Service examined a Berlese funnel. Turtox News 43: 115.
the plans of the experiment and concluded, with the au- Lindroth, C. H., H. Andersen, H. Bodvarsson, and S. H.
thors, that the total mangrove faunas would be unaffected. Richter. 1967. Report on the Surtsey investigation
Steve Tendrich, Vice-President of National Extermina- in 1966. Terrestrial invertebrates. Surtsey Research
tors, Inc. (Miami) coped imaginatively and effectively Progress Report, 111: 59-67.
with the unique problems presented by the project. W. H. MacArthur, R. H., and E. 0.Wilson. 1963. An equiBossert
and P. B. Kannowski assisted in the original librium theory of insular zoogeography. Evolution
censuses. Maps and tables for all parts of this series 17: 373-387.
were done by Mrs. Helen Lyman. The work was sup- MacArthur, R. H., and E. 0.Wilson. 1967. The theory
ported by N S F grant GB-5867. of island biogeography. Princeton University Press.
203 p.
LITERATURECITED
Mainwaring, A. P. 1961. Some effects of methyl broChisholm,
R. D. 1952. Nature and uses of fumigants, mide on aphids and whitefly and their host plants.
P. 358. In Insects, The Yearbook of Agriculture 1952. ~~~~~~~l thesis, ~~~d~~ univ., ~ ~ ~~ England, d
Darlington, P. J. 1957. Zoogeography: the geographi- Monro, H. A. U.,and Delisle. F~~~~~~1943. applical
distribution of animals. Wiley, New York. 675 p,
cations of bromide as a fumigant. Sci. Agr.
Davis, J. H. 1940. The ecology and geologic role of
mangroves in Florida. Papers from Tortugas Lab., 23: 546-556.
Carnegie Inst. Wash. 32: 307-412. Page, A. B. P., and 0.F. Lubatti. 1963. Fumigation
French, R. A. 1964 (1965). Long range dispersal of insects. Ann. Rev. Entorno'. 8: 239-264.
illsects in relation to synoptic meteorology. Proc. Simberloff, D. S., and E. 0.Wilson. 1969. ~ x ~ e r i Int.
Congr. Entomol. 12th (London). 6: 418-419. mental zoogeography of islands. The colonization of
Fridriksson, S. 1967. Life and its development on the empty is1ands. Ecology 50: 278-296,
volcanic island, Surtsey. Proc. Surtsey Research Wolfenbarger, D.0. 1946. Dispersion of small orgaConf.,
1967: 7-19. nisms. Amer. Midland Naturalist 35: 1-152.
EXPERIMENTAL ZOOGEOGRAPHY OF ISLANDS : THE
COLONIZATION O F EMPTY ISLANDS
The Biological Laboratories, Harvard University, Cambridge, Massachusetts 02138
(Accepted for publication December 16, 1968)
Abstract. We report here the first evidence of faunistic equilibrium obtained through
controlled, replicated experiments, together with an analysis of the immigration and extinction
processes of animal species based on direct observations.
The colonization of six small mangrove islands in Florida Bay by terrestrial arthropods
was monitored at frequent intervals for 1 year after removal of the original fauna by methyl
bromide fumigation. Both the observed data and climatic considerations imply that seasonality
had little effect upon the basic shape of the colonization curves o i species present vs. time.
By 250 days after defamation, the faunas of all the islands except the most distant one ("El")
had regained species numbers and composition similar to those of untreated islands even though
population densities were still abnorn~allylow. Although early colonists included both weak
and strong fliers, the former, particularly psocopterans, were usually the first to produce large
populations. Among these same early invaders were the taxa displaying both the highest
extinction rates and the greatest variability in species composition on the different islands.
Ants, the ecological dominants of mangrove islands, were among the last to colonize, but they
did so with the highest degree of predictability.
The colonization curves plus static observations on untreated islands indicate strongly that 

a dynamic equilibrium number of species exists for any island. We believe the curves are 

produced by colonization involving little if any interaction, then a gradual decline as inter-

action becomes important, and finally, a lasting dynamic equilibrium. Equations are given for 

the early immigration, extinction, and colonization curves. 

Di~persalto these islands is predominantly through aerial transport, both active and passive.
Extinction of the earliest colonists is probably caused chiefly by such physical factors
as drowning or lack of suitable breeding sites and less commonly by competition and predation.
Present address: Department of Biological Science, Florida State University, Tallahassee, Florida 32306
Early Spring 1969 COLONIZATION O F EMPTY ISLANDS
As population sizes increase it is expected that competition and predation will become more
important. Observed turnover rates showed wide variance, with most values between 0.05 and
0.50 species/day. True turnover rates are probably much higher; with 0.67 species/day the
extreme lower limit on any island. This very high value is at least roughly consistent with
the turnover equation derived from the MacArthur-Wilson equilibrium model, which predicts
turnover rates on the order of 0.1-1.0 species/day on the experimental islands.
1NTRODUCTION
In the first article of this series (Wilson and
Simberloff 1969) we showed how the recent formulation
of mathematical biogeographic theory
has both intensified the need for studies of the
entire coloilization process and defined the measurements
required for such studies. The idea
was conceived of approaching the problem experimentally
by the removal of entire arthropod faunas
from series of small islands. Six very small mangrove
islands of the Florida Keys, each consisting
of only one to several Rhizophora mangle trees
standing in shallow water, were selected. Elimination
of the faunas("defaunationn) was achieved
through fumigation with methyl bromide; and
techniques were worked out for censusing the
arthropod species during the recolonization pro-
cess.
In the present article we discuss the criteria
we used for counting species and describe the
recolonization process on all six islands, from the
moment of defamation to the reattainment of equilibrial
numbers of species less than a year later.
For the species counts and discussion which follow
these definitions will be used:
Propagule: the minimum number of individuals
of a species capable of breeding
and population increase under ideal conditions
for that species (unlimited food
supply and proper habitat, no predators,
etc.)
Colonization : the existence of at least one
propagule of a species on an island
Extinction : the disappearance of a species
from an island
Invasion : the arrival of one or more propagules
on an islandu 

Immigration: the arrival of a propagule on
an island unoccupied by the species
The distinction between invasion and immigration
should be noted. It is incorrect to speak of an
imnligration rate for one species, since an immigration
rate for an island is in units of species/time.
A species can have an invasion rate on a given
island, however; this is simply the number of
propagules of that species landing per unit time.
In analyzing species counts made at discrete
intervals, as in our monitorings, every species
for which at least one propagule exists is designated
a colonist. It is also designated an immigrant
if it was not a colonist at the preceding
count. Every immigrant is also, by definition, a
colonist. This definition says nothing about
whether food and a breeding site exist; a species
whose propagule lands on one of our islands is a
colonist even if it is doomed to quick extinction
for purely physical reasons (e.g., the absence of
a suitable nest site in the Rhizophora for a given
species of ant).
Because of the relative nearness of our experimental
islands to source areas and, to a lesser
extent, their small size and ecological simplicity,
ambiguities concerning the state of colonization
exist that would not arise if we were dealing with
truly distant and larger islands. Except for a few
birds, any animal species for which a propagule
is recorded either avoids leaving the boundary
of the island or (much more rarely) leaves and
perishes in the sea. The island, therefdre, is not
si~nplyan extension in some sense of the mainland.
In our experiment a small percentage of the
animal species, less than 10% of all species sighted,
behave as though the distances to the experimental
islands are not qualitatively different from the
same distances overland. W e wish to discount
these species in our calculations, unless insularity
becomes important in particular instances. Two
classes of species can be recognized in this connection.
Several kinds of insects (cicadas, odonates,
foraging bees and wasps) treat small mangrove
islands as part of a fine-grained foraging
area, traveling readily and frequently among several
islands and adjacent shore regions. A species
of the wasp genus Polistes forages regularly over
small mangrove islands but rarely nests there.
When nesting does occur the wasps apparently
restrict their foraging largely to the nesting island.
Only an extant nest qualifies the Polistes as a
colonist. In similar cases actual breeding, rather
than just the presence of sufficient animals to breed
under the most favorable conditions, was employed
as the criterion for colonization. Transient adult
butterflies, particularly Ascia monuste and Plzoebis
agarithe (Pieridae), migrating over and beyond
the experimental islands, occasionally skim
briefly through but do not stop to breed. These
will not be considered colonists for reasons sirnilar
to those used to discount fine-grained foragers.
280 	 DANIEL S. SIMBERLOFF
A second difficulty associated with the nearness
of our islands concerns intermittent breeding by
strong-flying insects combined with continuous
foraging by adults. Females of the moth Automeris
io (Saturniidae) fly frequently onto small
mangrove islands and occasionally deposit eggs.
The life cycle from egg to adult of Autovneris on
Rhizophora is about 2% months, and if (as may
have happened on E3) adults breed on an island
at intervals greater than that, extinction rates
would appear to be high. For after a brood matures,
the survivors generally all disperse from
the island and no new adults may breed there for
a period. This is obviously not extinction in its
classical sense-it is not caused by competition for
food or space, predation, climatic catastrophe, etc.
-but it does accord with our strict definition given
above. This situation was fortunatelv rare. almost
entirely restricted to a few lepidopterans. Species
of this type will be considered colonists, with one
immigration only and no extinction, until a definitive
and extended absence is recorded. The behavior
of the very few nesting birds would place
them in this category but they were nevertheless
discounted in our analysis.
A few arthropods live in and among Rhizophora
roots at or below the water level, some foraging
on mud at low tide. These include the isopod
Ligia exotica, an unidentified amphipod, and three
insects : Trochopus plumbeus (Hemiptera : Veliidae)
, Axelsonia littoralis (Collembola : Isotomidae),
and Anurida maritima (Collembola: Poduridae).
These will be excluded from the
species counts because they are essentially part of
a surface marine community and apparently do
not interact significantly with the arboreal mangrove
fauna. All are ubiquitous around small
mangrove islands and cannot be eradicated with
certainty.
The impermanent mudbanks on E8 and E9
(Wilson and Simberloff 1969, Fig. 2) that remain
wet but above water for several weeks in calm
weather harbor a characteristic marine arthropod
community, listed in Table 1. Most of the species
are concentrated in washed-up and wet debris
and algae. That all but the earwig Labidura
riparia are virtually marine and do not breed on
the islands is indicated first by their never having
been collected on Rhizophora (even when the
mud is submerged and debris washed away by
wind-driven high tide), second by most species
having been observed swimming from one patch
of mud or debris to another, and finally by the
swift recolonization by large populations of most
species observed after extinction caused by extended
flooding. L, riparia is the only species
found on the islands proper, and it is probably the
AND EDWARD O. WILSON Ecology, Vol. 50, No. 2
TABLE1. Arthropod community of intermittently submerged
mudbanks on small mangrove islands
INSECTS 

Collembola 

Poduridae: gen. sp. 

Dermaptera 

Labiduridae: Labidura riparia 

Coleoptera 

Carabidae: 	 Bembidion sp. nr. contractum
Tachys occulator
Anisomeristes sp.
Actinopteryx fucicola
gen. sp.
Hemiptera
Saldidae: Pentacora sphacelata
OTHER 

Acarina 

Veigaiaidae: Veigaia sp. 

Isopoda 

Oniscoidea: Ligia exotica 

only one whose energetic interaction with the mangrove
and its arboreal fauna is significant. All
the others apparently feed on seaweed or washedup
detritus, or else prey upon those which do.
Consequently L. riparia alone is considered a
colonist.
The tree snail Littorina angulifera and tree crab
Aratus pisonii inhabit all but the upper canopy of
small mangrove islands but will not be counted
for the following two reasons. Neither can be
removed-Littorina is unaffected by 50 kg/1000
m3 of methyl bromide for 3 hr, and Aratus siillply
drops to the water and may swim under the tentand
both have planktonic larvae. Again, the interaction
of these species with the remainder of
the arboreal community appears superficially not
to be significant.
Our definition of a propagule dictates that animals
with zero reproductive value (e.g., a male
ant landing on an island after a nuptial flight)
not be considered colonists.
Although birds were not counted here, bird
parasites which establish breeding populations on
the islands rather than wholly on the birds were
listed as colonists. Specifically, the hippoboscid
flies Olfersia sordida and Lynchia albipennis
(whose puparia are con~monly found in tree
crevices) and the tick Argas radiatus (all stages
of which live under dead mangrove bark) were
counted.
Finally, except for the rare larvae and pupae,
all Diptera were excluded. Monitoring of flies
proved too difficult to warrant faith in the accuracy
of species counts, and the extreme scarcity of
immature stages indicates that small mangrove
islands rarely support breeding dipteran popula-
tions.
Deep-boring beetles were deemed valid colonists
--
Early Spring 1969 COLONIZATION O F EMPTY ISLANDS 	 281
in this series. The uncertainty of extinction discussed
in our first report (Wilson and Simberloff
1969) notwithstanding, the data imply that the
few initially surviving cerambycid larvae were
destroyed by a delayed effect, and that the weevils
may have succumbed in the same way. In any
event there are but four species involved, and
rarely were more than two found on a single
island.
Acceptable colonists were counted conservatively.
In all instances of uncertainty about the
number of species of a given taxon present, the
minimum possible number is taken. For example,
occasionally the records from one census revealed
two thrips, Neurothrips magnafemoralis and Liothrips
sp. (both Tubulifera), as well as larval
thrips identifiable only to the suborder Tubulifera.
Since the larvae could conceivably be ascribed to
one of the two species known present, only two
species are recorded for this period. Similarly
the observations on El, (May 29, 1967) of a moth
caterpillar Bema ?ydda (Phycitidae), and a small
adult moth similar to Bema but seen too briefly
to be so recorded with certainty yield a species
count of one only, since both individuals could
belong to one species.
The presence of an immature animal need not
imply breeding on an island; spiderlings balloon
more readily than adults, and any insect larva
could be blown or rafted to an island, although for
some, of course, the probability of this is quite
low. Similarly, an adult female does not constitute
a propagule or part of one if she has not been
fertilized, is not of a parthenogenetic species, and
no male is present. Nevertheless, we assume here
that an adult female, an adult of indeterminate
sex, and an immature animal each imply the presence
of a propagule. Adult males are not so
counted.
In the Appendix are given the complete recorded
histories of all of the colonists on E l , E2,
E3, ST2, E7, and E9, the six islands whose entire
faunas were removed by fumigation (see Wilson
and Simberloff 1 x 9 ) . These records are
based on direct observation in over 90% of the
cases. In certain instances (hatched bars) animals
were assumed present through one or more
monitoring cycles when not actually observed. T o
ensure consistency in such interpolations the rules
given in Table 2 were followed.
We must first discuss whether any aspects of
the colonization depicted in the Appendix (and
Figs. 1-3) are artifacts of the particular season
at which defaunation was performed. That is, if
all the islands were fumigated in September inTABLE2.
General rules applied in interpolatioil of colo-
nists
Number of
cycles
Animal interpolated Justification
Ants: 2 (previous) 	 Conservative on physiological
workers seen first 	 grounds
Ants:queen indefinite Obvious for short periods. Data 

seen first of E9 support for longer perio~s. 

~ ~ ~ p - b ~ ~ i ~ ~2+ Present in relatively low density
bet'e8 and only 10%of twigs broken at
each monitoring.
smallleaf I-2 Inconspicuous, but densities
dweUers usually incresae rapidly.
k d 1~ Often become dense quickly, and~ ~ - ~ . ~ ~ ~ ~
halitat examined completely.
Araneids 1 	 Spiderlings may be minute, but
webs are conspicuous.
Tdragnoflul 2 	 Position of webs makes less
conspicuous than araneids.
2-3 	 Furtive, often inconspicuous, and
usually present in low densities.
Anyphaenids 	 Build up relatively high densities
quickly.
l 1 ISmall 	 Usually low densities; difficult
crawlers
I 2 1to record.
Fliers 	 Frequently conspicuous, but may
be inactive because of weather.
ICaterpillars 	 No generalization possible; large
onea are conspicuous.
stead of March would the tables and derived
curves be qualitatively different ?
This would obviously be so in much of the
United States, where dispersal stages of most insects
and spiders occur at short, distinct periods
(usually in the summer). The Florida Keys are
subtropical, however, with the mean temperature
of the coldest month (20.g°C) only 7.7"C lower
than that of the warmest month. Frost has never
been recorded. The mean humidity of the driest
month is but 8% lower than that of the most
humid. Rainfall is less homogeneous, SeptemberOctober
averaging about 150 mm and DecemberJanuary
only 38 mm. The precise amount for all
months is quite variable, however, and there are
obviously no extreme dry or wet seasons. Wind
is also relatively constant over the year in both
speed and direction: it averages about 18 km/hr
from the eastern quadrant. It is not surprising,
therefore, that we found most of the mangrove
arthropod species active throughout the year. All
life stages, including dispersal forms, of many and
probably most species of insects and spiders were
present during every month. We can make no
282 DANIEL S. SIMBERLOFF
quantitative assertions, but it seemed to us that
there were no striking seasonal decreases among
the more abundant mangrove inhabitants in either
population size or activity, including flight-except
for mosquitoes, which were much more numerous
in the summer.
We had the good fortune to be present February
26, 1967,when the temperature at Key West
fell briefly to 9.5OC, the lowest reading in several
years. Our surveys during that cold spell revealed
no apparent mortality or even great lessening
of activity of mangrove inhabitants other than
a decrease in flight activity quite normal for the~ prevailing
wind speed.
Finally, and most importantly, the conclusion
that propagules were constantly hitting our islands
is incontrovertible from the data summarized in
the Appendix. Moreover, these data provide no
clear indication of seasonality in the dispersal of
DAYS
FIG. 1. The colonization curves of the experimental
islands in series 1. In each curve the last and next-tolast
census points are not connected by a line, only in
order to stress the greater period of elapsed time compared
with the times separating earlier censuses.
60 120 180 240 3W 360 420 4 8 0 540
DAYS
FIG. 2. The colonization curve of island E7.
AND EDWARD 0. WILSON Ecology, Vol. 50, No. 2
DAYS
FIG. 3. The colonization curve of island E9
any taxon. The colonization of E7, defaunated
aphroximately half a year before the other islands,
should certainly have manifested any strong seasonal
component. Yet in both form and specifics
(Fig. 2 and Appendix) it is consistent with the
colonization of the other. islands.
As would be expected in a system involving
but one plant,. succession in the usual sense, a
progression of discrete and relatively stable communities,
did not occur. This does not imply a
lack of order in the time course of colonization;
indeed, the invasion of species was remarkably
regular. But it was not accompanied by wholesale
extinction of distinct animal associations.
Several broad patterns are nevertheless evident.
First, although the earliest immigrants on all islands
included both strong fliers (especially moths
and wasps) and weak fliers or nonfliers (particularly
psocopterans, chrysopids, and spiders), the
latter built up large populations more rapidly and
became numerically dominant. That wind should
transport many of the early invaders is not surprising.
The first animal recorded on Krakatau
after its eruption was a spiderling (Cotteau 1885),
while psocopterans (including wingless nymphs)
and spiders are prominent in aerial plankton samples
(Glick 1939). But their success in colonization
deserves further comment. The food supply
for psocopterans, algal and lichen growth on the
mangrove itself, is evidently sufficient to allow
much larger populations than one normally finds
on undisturbed Rhizophora islands. This implies
that on untreated islands there may be predation
by animals not present on recently defaunated islands.
In fact, we have observed the ants Pseudomyrmex
elongatus and Pseudomyrwex "fiavidula"
(both of which colonized later) carrying appar-
Early Spring 1969 COLONIZATION O F EMPTY ISI~ANDS 283
ently freshly killed psocopterans to their nests.
Running and jumping spiders can, of course, eat
the psocopterans. Maezhia vittata (Salticidae)
was seen catching a Psocidus texanus adult on E7,
while the webs of even small araneid spiderlings
comrnonly trapped transient flies, especially tipulids
and ceratopogonids, in the same size range as
psocopterans.
Although it involves only a relatively small subset
of the entire Florida Keys fauna, colonization
by these early weak fliers was more variable than
by other classes of arthropods, both in time of
arrival on individual islands and species composition
at any given time among islands. The psocopterans
were particularly unpredictable; from
the 2d month after defaunation there were usually
1-4 species on each island at any census period,
but over the course of the experiment a total of
24 species were involved. Although certain psocopterans,
especially species of Psocidus and
Peripsocus, were generally more prominent than
others, there was little correlation among the sets
of psocopterans found on different islands. Furthermore,
as can be seen from the Corrodentia
(= Psocoptera) sections of the Appendix, many
psocopterans persisted for less than a month. As
a group they invaded and multiplied readily, and
became extinct almost as readily. On E7, only
5 species of the 15 colonists remained as long as 2
months.
Spiders as a group were less variable than
psocopterans in their colonization pattern but much
more so than that of most later colonists. Although
the majority of the 36 spider species which
colonized the islands followed the pattern just
described for the psocopterans-that is, they immigrated
readily and were extinguished quickly,
and occurred on but 1 or 2 islands-a few species
behaved quite differently. Eustala sp., Tetragnatha
sp., Leucaztge venusta, Hentzia palwtaruwt, and
Aysha velox in particular, colonized most of the
islands and usually persisted for at least several
months. These include most of the spiders found
on the islands before defaunation.
Wasps present a similarly heterogeneous picture
of colonization, many species appearing on one
or two islands and vanishing rapidly while a few,
notably Pachodynerus nasidens, Scleroderwta wcrogaster,
and Calliephialtes ferrugineus, coloniz,ed
many islands and persisted for long intervals.
Mites did not invade as early as did wasps and
spiders, but displayed the same pattern of many
short-lived species, a few persisting and recurring.
Most of the 20-odd species of Acarina were recorded
from one or two islands only and disappeared
within a month, while A*zblyseius sp.
and Galuwtna sp. were omnipresent and their
populations long-lived. An apparent correlation
exists in the spiders, mites, and wasps between
mean length of persistence and number of islands
colonized. This relation, however, may be artifactual,
for the following reason. If two species,
the first with a very high, the second with a very
low initial probability of extinction (long expected
persistence time), invaded all six islands with
equal frequency, we would expect to see the former
on one or two islands only and the latter on most
or all of them.
Thrips, lepidopteran's, orthopterans and ants display
a regularity of colonization in sharp contradistinction
to the relatively unordered patterns of
colonization previously described. The last three
groups, particularly the ants, mount the largest
populations in undisturbed mangrove animal com-
munities.
Only five species of thrips colonized, four of
which were widespread among the islands. Almost
all invasions occurred 4-5 months after defaunation.
Most thysanopteran colonizations endured
for at least 3 months. Large populations
were occasionally produced but rarely persistedthe
late colonizing ants may have attacked thrips.
There was no consistent order of colonization:
all species commonly immigrated about the same
time.
Only 8 lepidopteran species colonized the experimental
islands, of about 30 species known
from mangrove and a few hundred from the general
Keys fauna. (This figure does not count the
2 or 3 fine-grained foragers discussed earlier.)
All eight were recorded more than once, and six
were widespread. Most colonizations were sustained
and many involved sizable populations.
There was a somewhat predictable order of invasion,
with Phocides batabano, Bewta ?ydda, and
Ecdytolopha sp. usually the first arrivals, Newtapogon
sp. appearing somewhat later, and Alarodia
slossoniae and Autonzeris io usually not seen until
about 200 days.
The orthopteroids colonized still more predictably.
Of approximately 25 species that occur
in mangrove swamps and 60 or more that occur
in the Keys as a whole, only 9 invaded the experimental
islands. If the two very near islands (E2
and E7) are discounted, only four species were
involved and all have multiple records. These
four were rarely extinguished; two, the green tree
cricket Cyrtoxipha confusa and roach Latiblattella
n. sp., produced large populations. All appeared
capable of early invasion.
The ants displayed the most orderly pattern of
colonization. These insects are also nun~erically,
------
254 DANIEL S. SIMBERLOFF AND EDWARD O. WILSON Ecology, Vol. 50, No. 2
TABLE3. Colonization of experimental islands by ants The order of colonization by ants of the experimental
islands (Table 3) provides a curious zooEl
E2 E3 E7 E9 ST2 geographic analog of Haeckel's biogenetic law.Island
Ant species
before
defaunationa
Ant
colonists
in order of
colonization
In almost every instance Crewzatogaster ashfneadi
was the first colonist, even on the two near islands
(E2 and E7) where it was not present before
defaunation. On all but the most distant island
( E l ) Pseudowyrwex elongatus was an early and
prominent colonist. Moredver, the small subset
of the mangrove ants which comprised the remainder
of the colonists was almost identical to
that found on small islands as one moves nearer to
the source areas. Only 3 extinctions were observed,
and 2 were of species believed unable to
nest in red mangrove forests. As a group, the
ants colonized later than most other taxa and at
the close of the first phase of our study (April
1968) few species had built up populations nuinerically
similar to those on untreated islands.
COLONIZATIONCURVES
Figures 1-3 show the original numbers of species
present and the colonization curves (number
of species present vs. time) for the experimental
islands.
Estimation of the number of species present before
defaunation and after regular monitoring had
ended is complicated by two factors. First, as
can be seen from the data presented in the Appendix,
a few colonists are inferred preseilt at each
regular post-defamation monitoring period without
actually having been observed because they
were recorded at both preceding and subsequent
periods. Since there was but one pre-defaunation
survey, its total species number must be increased
by the mean number of species inferred present
without having been observed for all regular postdefaunation
monitorings after an approximate
equilibrium 3 had been reached. The last four
regular monitorings were used for this purpose,
since the population structure then was probably
closest to that before defaunation. A similar correction
must of course be made for the final census,
after regular monitorings had ended. In
addition to this correction, the pre-defaunation
surveys of E l , E2, E3, E6, and E7 were believed
deficient for leaf fauna. Since this habitat normally
harbors 3 to 5 species on all islands, it was
assumed that a total of 4 such species were present
on each of these islands, though the actual number
seen was 1 to 4. Figures 1-3 do not include
a correction for unseen short-lived species, estimated
at about 2 species per monitoring period but
with very high variance (Simberloff 1969).
The numbers of species recorded on the control
islands before and after the experiment were:
5 3 5 1 4 5
6 8 6 2 8 6
11
11 11 11 11 11
l3 l3 l4 l2 l3
14 13 14
6
1 l? 11.
[
{7 b
11 14 1 l1
11
{ {14 14 13
12
9 = Paracryptocerus varzans
10 = Paratrechina bourbonica
11 = Pseudomyrmes elongat~s
12 = Pseudomyrmez"fiavzdula'la
13 = Tapznoma lzttorale
14 = Xenomyrmezflondanus
15 = Brachymyrmez sp.
16 = Hypoponera opaczor
*Species are coded as follous
1= Camponotusfiorzdanus
2 = Camponotus p1anatt.s
3 = Camponotus tortuganus
4 = Camponotus sp.
5 = Camponotus (Colobopszs)sp.
6 = Crematogaster ashmeadz
7 = Crematogaster atkznsonz
8 = Monomorzum fiorzcola
'1 ~nd~cateslater extinction.
and probably energetically, the dominant animals
on all small mangrove islands. Of the more than
50 species found in the Keys, 20 species inhabit red
mangrove swamps and about 12 of these normally
occur on small islands. The pre-defamation surveys
revealed a highly ordered fall-off of ant species
in two directions. First, on islands of equal
size but varying distance from source area, the
most distant islands contain Crewatogaster ashmeadi;
those somewhat nearer, both the Crewtogaster
and Pseud~wyrvnex elongatus; those
nearer still, these two species plus Paracryptocerus
varians, Tapinovlza littorale, and Camponotus (Colobofisis)
sp.; and on islands near shore, most or
all of the above species plus one or more species
of Camponotus, Pseudomyrwex "flavidula," Monomoriuuuz
floricola, and Xenouuzyrvlcex floridanus.
If instead one fixes a distance (usually small)
from the source area and examines islands of
increasing size, he generally finds on the smallest
bush (ca. 1 m high) Crematogaster ashmeadi;
on slightly larger bushes, C. ash~vzeadiand/or
Pseudomyrwex elongatus; on small trees these
two with perhaps two species drawn from among
Paracryptocerus, Tapinowa, Colobopsis, Xenomyrvnex,
and Monomorium; and on islands the
size of our experimental ones, the full complement
expected on an island of the appropriate
distance from source. In short, the ability to
colonize increasingly smaller islands parallels
closely the ability to colonize increasingly distant
ones.
Early Spring 1969 COLONIZATION O F EMPTY ISLANDS 	 285
before after
E6 30 28"
E l 0 20 23
* plus 9 spp. in bird nests only
These cannot be corrected for unseen colonists,
but show conclusively that no important long-term
effect was operating in the Keys that might have
distorted the results from the defaunated islands.
Although the numbers of species on the control
islands did not change significantly, the species
colllposition varied considerably, implying that the
number of species, S, approaches a dynamic equilibrium
value, 3.
The most apparent implication of the colonization
curves and other information presented so
far is that this equilibrium S does exist. Three
lines of evidence are relevant. First is the fact,
iust mentioned. that S on the control islands did
not change greatly from the beginning to the end
of the year-long period in which the nearby experittlental
islands were being colonized. Second
is the static observation that untreated islands
with similar area and distance from source have
similar 3 (note E3 and ST2 before defaunation).
Perhaps the most convincing argument. for an
equilibrium S, however, is the increase of species
present on all our islands to approxin~atelythe
same number as before defaunation, and then
rough oscillation about this number. This S may
be only a quasi-equilibrium-that is, the curve of
S versus time after S is reached may not be truly
stationary-because of two long-term processes.
The first is that when an approximate .$ is first
reached the population structure of the particular
set ot' species on the island is still changing rapidly;
some species are represented by few individuals
and may have large r (intrinsic rate of
increase). Others may have abnormally large
populations. ~ o ~ u l a t i o hsizes are generally fluctuating
much more rapidly than on untreated
islands. W e have already indicated one manifestation
of this process in our experiment: psocopterans
colonized early and built up immense
populations before presumed ant predators appeared
in number. The effect of this extreme
population fluctuation on the colonization curve
is that s~ecieslnav be eliminated and added at a
rate systematically different from that on an untreated
island. The concept of a dual dynamic
ecluilibrium-of species number and population
structure-will be mentioned briefly in this paper
and described more completely and formally by
Sin~berloff(1969).
The joint evolution of the particular constellation
of species on an island ought logically to
raise S systematically over very long periods of
time (Wilson and Taylor 1%7), but we will neglect
this effect because the time course of such
a change is obviously beyond that of this experiment.
Also, the invasion rates for most species
on our islands are probably too high to allow significant
genetic alteration to occur in populations
on individual islands; they are not isolated in an
evolutionary sense.
The curves of Figs. 1-3, except for that of E l ,
are believed best explained by the following equation,
based on a model devised by W. H. Bossert
and P. N. Holland:
P
ia
EIS(t)l = C i a + e a - - ( i a + ea)t)
a = l
where E [ S ( t ) ] =	the expected number of colonists
present at time t
P =number of species in the pool
i, =invasion rate of species a
g,= intrinsic probability of extinction for species
a
The derivation of this equation and a discussion
of the concepts involved will be presented elsewhere
(Simberloff 1969). The important aspects
of this theory for our immediate purpose, however,
are as follows :
The variance of S ( t ) is high:
P
var [S(t)]= C E[Sa (t)] { 1 - E [ S a (t)]
a = l
S , is a species indicator variable, which equals 1
when species a is present and 0 when species a
is absent. Also,
Once the number of species on these particular
islands is between about 75% and 90% of the
equilibrium value of
the major premise of this stochastic version of the
MacArthur-Wilson equilibrium theory, namely
non-interaction of species, would be invalidated.
From this point onward S ( t ) declines slightly, at
a slow rate which cannot yet be predicted well
stochastically. It equilibrates ultimately at an enduring
S partly determined by interaction and
close to the number that existed before defaunation.
The decrease on all islands in number of
species present for the final census (after regular
286 DANIEL S. SIMBERLOFF
monitoring had ended) is believed to be a manifestation
of this decline.
El was so distant from its presumed source area
that a few early invaders were able to build up
large populations before their probable competitors
arrived. Interactions thus became important
before even a small fraction of the non-interaction
S was achieved. We predict that on El the colonization
curve will ascend slowly and irregularly
to an equilibrium near the pre-defamation 3. This
enduring equilibrium will probably be the same
one to which the colonization curve of El would
ultimately have descended had interactions not
become important until a large fraction of had
been achieved, as on the nearer islands.
As the distance of the island from the faunal
source increases, the non-interaction 3 should decrease
because of decreases in the i ,. Furthermore,
the time necessary to reach any given percentage
of the non-interaction 3,though not readily
expressed mathematically, can be shown to increase
with increasing distance from source. If
this time is sufficiently long the few early colonists
are able to produce large enough populations to
interact significantly with later immigrants. This
in fact is what happened on El.
The colonization curve of E7 (Fig. 2) must
be considered in light of the fact that 85% of the
tree was killed by the fumigation (Wilson and
Simberloff 1969). The dead portion did not disappear,
but rather deteriorated until by the end of
one year the wood was brown and dry and much
of the bark was peeling. Thus the island was
not a constant factor, and the relative proportions
of the various microhabitats changed drastically
until ultiinately there was far more dead wood and
bark and far less leafy canopy than on an untreated
island. If a consistent measure of area existed
for these islands it would probably have remained
unchanged on E7, but the expected number of
species would not because the different microhabitats
normally support different numbers of
species. In particular, dead bark shelters numerous
species while mangrove leaves rarely support
more than four species on a single island. Before
its gradual decline, the colonization curve of E 7
rose to a far higher percentage of the original S
(135%) than did that of any other island. In
addition, it was still rising a year after defaunation,
when those of all other islands but El had
leveled off. Arachnids (excluding orb-weavers)
and psocopterans were the animals largely responsible
for the higher S near the end of the experiment.
Both groups are primarily bark dwellers.
All the above considerations imply that one of
the determinants of the shape of the colonization
curve on this island alone was variation of the
AND EDWARD O. WILSON Ecology, Vol. 50, No. 2
island habitat, and that the enduring S which will
ultimately be achieved on E 7 may be very different
from the pre-defamation figure.
The agents of dispersal for specific immigrations
can rarely be given with assurance, but the
evidence implies that aerial transport, passive and
active, is the major mode of invasion.
For the several parasites of vertebrates that
breed on the islands proper, zoochorous transport
is certain. These arthropods include the hippoboscid
flies Olfersia sordida and Lynchia nlbipennis
and tick Argus rndiatus on El and E9. All
three species parasitize the cormorants and pelicaris
which roost on these islands. A number of
mite colonists could also have arrived on transient
larger animals. This appears to he so for Entonyssus
sp. on E2; it is an obligate parasite in
snake lungs, a::d a Natrix was recorded sloughing
on E2. The inl-asions of El,. E3, and E7 bv the,
mite Ornithonyssos bursa, an avian parasite, were
probably ornithochorous.
A more interesting ;~ossibilityinvolves phoresy
on birds, particularly by soiile of the psocopterans
prominent among early immigrants. Mockford
(1967) has hypothesized that phoresy may be a
more efficient means of insular invasion than wind
transport, especially for smaller species. Anlong
four species found on Asian birds he lists Ectopsocopsis
crypto?neriae, which has also colonized
E3 and E7. IYhether phoresy is a significa~ltphenomenon
in the Keys remains to be determined.
Other arthropods that may utilize phoresy are
the pseudoscorpion TyrannocheliJer sp. found on
E7, and free-living mites on all islands. These
could have been transported by larger iilsects as
well as by birds.
A related transport mechanism utilizes nesting
material carried by birds. W e suspect without
proof that several mites in our experiment arrived
in this fashion. Two beetle larvae were found in
deep excavations in Green Heron nest twigs soon
after the nests were built: Chrysobotlzris tranquebarica
on E 3 and ?Sapintusfulvipes on ST2 (the
latter in a twig not of Rhizoplzora). It seems
certain that both were in the twigs when the birds
constructed their nests. Meyerriecks (1960) has
observed green herons using twigs from other
trees and from the nest tree itself. It is our impression
that most nesting material that goes into
new nests on small mangrove islands is brought
from elsewhere. This was assuredly so for the
non-limngrove twig on ST2.
Hydrochorous transport, either free or on rafts,
was not as important as aerial transport in the
invasion of our islands. Many mangrove colonists
287Early Spririg 1969 COLONIZATION O F EMPTY ISLANDS
remain afloat almost indefinitely in salt water, but
only the orthopteroids seem able to achieve significant
independent, oriented motion. Furthermore,
most floating insects and spiders of all sizes
are rapidly devoured by fingerling fishes which are
immensely numerous about all mangrove islands.
Even actively swimming crickets and earwigs fre~
quently meet this fate, attracting more attackers
by the very vigor of their efforts.
Nevertheless, at least one invasion (though not
immigration, since the species was already present)
occurred this way, and there are other pertinent
observations. An adult female Cyrtoxipha cricket
was seen floating from Upper Snipe Key to E2,
a distance of only about 2 m. She successfully
climbed a root and disappeared into the lower
canopy. That a winged individual fully capable
of oriented flight should actively choose to disperse
by water is dubious, but at least over this short a
distance the method is feasible in case the animal
accidentally lands in the water.
Four Azttowzeris io caterpillars were seen to
float from one end of E 9 to the other, a distance
of about 8 m. Three of them lodged on roots and
eventually climbed up out of the water. The weakening
of leaves attacked simultaneously by several
large Automeris caterpillars (a common occurrence)
might cause them to fall into water occasionally.
Also, the caterpillars release their hold
readily when the branch is shaken, apparently as
a defensive maneuver. It seems doubtful, however,
that a high proportion of those that land
in the water could successfully travel between islands.
Aside from fish attack, they would be
plagued by their inability to direct their motion
and to climb readily from the water.
Occasionally earwigs (Labidwra riparia) were
seen swimming from one root- to another within
an island. W e do not know how well this s~ecies
flies, but its swimming ability seems adequate for
aquatic travel of considerable distance, and it
readily climbs out of the water. Fishes would
still be a hazard, however.
Our experiments resulted in no unequivocal evidence
of invasions by rafting. In fact, several
considerations imply that rafting must play a
minor role in inangrove colonization. There is
rarely land on which rafts can lodge, since Rhizophora
islands generally have no supratidal ground,
Drifting wood usually hits an island, gets trapped
temporarily among roots, and eventually floats
away. Even more importantly, green Rhizophora
wood sinks immediately and many dead twigs also
fail to float, thus precluding rafting by much of
the wood-boring fraction of the fauna. Finally,
except during hurricanes, there is very little floating
debris in Florida Bay-far less than the
amount found at river mouths.
The preceding considerations lead, by elimination,
to the inference that aerial transport must
be an important means of dispersal to our islands.
Several observations of actively flying propagules
support this hypothesis. Buprestid and cerambycid
beetles, lepidopterans, wasps, and a lacewing
have all been seen to land on the islands after flight
from an outside source. Many other mangrove
colonists are evidently carried passively by wind,
especially psocopterans, thrips, neuropterans, and
most spiders. Many of these more or less passively
dispersed organisms are minute, and direct
evidence on their invasion method is therefore
scarce. Occasionally spiderlings and psocopterans
were found in the air around our islands. In
general, they are usually important components
of aerial plankton (Glick 1939), and it is known
from numerous anecdotal records (e.g., Bristowe
1958) that ballooning for the distances involved in
our experiment is regularly achieved by some species
of spiders. So far, the small sizes of most of
these animals has made it impossible to follow a
flight visually from source to island. One suggestive
record, however, is that of a spider dragline
stretching the 2 m between Upper Snipe Key
and E2.
An attempt to correlate Figures 1-3 with hourly
wind data of the U. S. Weather Bureau station
at Key West was inconclusive, but this piece of
negative evidence is hardly damaging to the thesis
of dominance of aerial transport. For such data
give only timed readings at one nearby point,
while dispersal depends largely on specific gusts
at odd times in highly circumscribed areas about
the islands. Furthermore the published wind data
are from a single fixed anemometer, while the
entire wind profile would be necessary to describe
dispersal.
The intermittent nature of the monitoring had
the unavoidable result that many immigrations and
extinctions (perhaps two-thirds) were not observed,
the species involved being obligate or nearobligate
transients which immigrate and are extinguished
all within one interval between two
monitoring periods. For this reason absolute
immigration and extinction curves cannot be derived
from the observational data.
The observed immigration and extinction curves
for all islands were highly variable, with no apparent
pattern except for generally higher immigration
rates on nearer islands during the first
150 days. Rates were usually between 0.05 and
0.50 species/day. Employing a statistical method
285 DANIEL S. SIMBERLOFF
devised by Siiuberloff (1969) it was determined
that the very least the expected error in immigration
and extinction rates could be is 0.67 species/
day, and furthermore that the variance about the
"expected" immigration and extinction curves is
very high. The observed curves are consequently
of limited interest, since they yield only an extreme
lower limit for turnover rates and cannot be used
to intuit shapes of the "expected" curves. On the
other hand, the values of the turnover rates are
of considerable interest even when they can be approximated
only to the nearest order of magnitude.
They are of course surprisingly high, in the vicinity
of 1% of the equilibrial species number per
day or higher. Yet this is at least roughly consistent
with the MacArthur-Wilson (1967) model,
which predicts that the turnover (= extinction)
rate at equilibrium is 1.15 (mean S/tO.gO, where
mean 3 is the average equilibrial species number
and toso is the time (in days) required to reach
90% of the equilibrial number. According to this
formula, which is based on the simplest non-interactive
version of the model, the turnover rates in
our experimental islands should fall somewhere
between 0.1 and 1.0 species/day. The relation between
this version of the model and the more
precise stochastic form of the model will be treated
later at length by Simberloff (1%9). The MacArthur-Wilson
formulation is a special case of the
many cases covered by the stochastic version and
it has the advantage of permitting this first rough
(and approximately correct) prediction of turnover
rates.
In testing such predictions with measurements
in the field there is reason to expect that the invasions
not observed will occur by different means
than those which are recorded. The assumption
that most propagules arrived by air in the experimental
keys is therefore probably valid. The evidence
against a major seasonal component of dispersal
was given earlier. In sum, we have a large
body of information which implies that i, (invasion
rate of species a) is nearly constant through
time for all a. What we lack at present is quantitative
information on the sizes and distribution
of the i,.
Extinction rates, at least during most of the rise
of the colonization curve from 0 to a large fraction
of S are adequately represented by the unchanging,
species-characteristic e, , without an additional
S-dependent or density-dependent factor included.
The main arguments behind this assertion are:
i) Most of the species in the Florida Keys pool
are obligate transients on these small mangrove
islands. For a variety of reasons not
directly related to their own densities or to
AND EDWARD 0.WILSON Ecology, Vol. 50, No. 2
other species they are doomed to swift ex-
tinction.
ii) Population sizes during most of the rise of
S from 0 to near 3 are uniformly low.
The observed data from this ex~eriment~rovide
rough quantitative information on both the distribution
and sizes of the e, unlike with the i,.
Whatever the i, and e, , the expected curve of
immigration rate vs. time is represented during
the rise of S from 0 to a large fraction of the noninteraction
3 by:
while the expected curve of extinction rate vs.
time during the same period is :
Beyond this point accurately predicted curves
are impossible. It is nevertheless clear that during
the slight decline of S to an enduring 3, E(t)
must be, on the average, slightly greater than
I(t), while after the equilibrium is reached the
two must remain approximately equal. On any
real island, of course, the two curves would cross
and recross indefinitely. It also seems reasonable
that after interactions become important, the E (t)
and I ( t ) curves still do not change much, since
the contribution from their common major component,
the transients, does not change with time.
Whereas evidence on the specific agents of dispersal,
and hence on immigration, has been plentiful
during this study, observations on the causes
of extinction have been meager. Obviously the
probability of witnessing the death or disappearance
of the last member of a population is exceedingly
low. Some inferences can be drawn
from observed means of population decrease.
Population decline should be most apparent
when associated with interaction, especially predation,
yet the small sizes of most populations during
our experiment reduced interactions enormously.
A few cases of predation have already
been mentioned. Insectivorous birds, particularly
warblers and red-winged blackbirds, were fre-
289Early Spring 1969 COLONIZATION O F EMPTY ISLANDS
quently observed eating numerous insects of many
species. Wasps, both parasitic and nonparasitic,
were seen destroying several insects and spiders.
Some of these attacks may have led directly to
extinction, when the prey populations were small.
Examples include the parasitism on Autorneris io
caterpillars by Apanteles henzileucae (Braconidae)
and the destruction of the salticids Hentzia palmarum
and Stoidis aurata by Trypoxylon collinurn
(Sphecidae).
~xclusioncan also ~rovideindirect evidence of
extinction through interaction. A possible instance
is the apparent predation of crickets on E9
by a large population of the centipede Orphnaeus
brasilianus. E8, with no centipedes, had immense
populations of four cricket species. After defaunation
of E9 removed Orphnaeus, a large population
of the cricket Cyrtoxipha confusa was rapidly
established.
From observation of the pre-defamation distribution
of ant species' numbers and population
sizes on the various islands, it seems probable that
when one ant species is able to build up large
populations before other species invade, it can
exclude one or more other s~ecies. However.
direct evidence of aggressive behavior among the
mangrove ants is lacking; the observed exclusion
may have been the result of nest site pre-emption.
Most extinctions were probably not the result
of interactions, at least during the initial rise of
the colonization curves, but rather resulted from
the inability of most species in the Florida Keys
pool to colonize these tiny mangrove islands under
any conditions. Lack of proper food or nest site
and hostile physical conditions are probably common
causes. An example was the observation of
a dealate queen of Brachynzyrrnex sp. on E9.
Since this species nests in soil, which is lacking
on the experimental islands, it is not surprising
that no workers were subsequently observed, despite
the fact that the queen landed in a totally antfree
environment.
Even species which can survive on small mangrove
islands have high probabilities of extinction
not related to interaction. A number of animals
have been found drowned during this experiment,
including entire small ant colonies, numerous
lepidopterous larvae, beetles, and psocopterans.
It seems probable that during the hurricanes which
periodically buffet these islands such deaths would
be commonplace.
Similarly, numerous mangrove colonists inhabit
hollow twigs and even under normal conditions
many of these fall into the water. Certainly such
events are multiplied during storms. The ant Canzponotus
floridanus and the oedemerid beetle Oxycopis
sp. seem particularly vulnerable in this respect:
they typically inhabit low, weakly anchored,
hollow roots.
We wish to thank Joseph A. Beatty and Edward L.
Mockford for identification of specimens and advice on
experimental technique. Robert E. Silberglied performed
the survey of Keys arthropods and aided in monitoring.
S. Peck, H. Nelson, and R. B. Root participated in island
censuses. John Ogden cultured many arthropod colonists.
The following specialists gave generously of their time
to identify animals encountered during the experiment:
D. M. Anderson, E. W. Baker, J. R. Barron, J. C. Bequaert,
L. Berner, B. D. Burks, J. A. Chemsak, K. A.
Christiansen, W. J. Clench, C. M. Clifford, B. Conde,
R. E. Crabill, P. J. Darlington, G. W. Dekle, H. Dybas,
H. E. Evans, J. G. Franclemont, R. C. Froeschner, A. B.
Gurney, J. L. Herring, W. Herrnkind, R. W. Hodges,
R. L. Hoffman, D. G. Kissinger, J. P. Kramer, J. F.
Lawrence, H. W. Levi, J. E. Lloyd, E. G. MacLeod, P.
M. Marsh, R. Matthews, W. B. Muchmore, S. B. Mulaik,
L. Pinter, C. C. Porter, E. S. Ross, L. M. Roth, L. M.
Russell, R. L. Smiley, T. J. Spilman, L. J. Stannard,
R. L. Usinger, B. D. Valentine, G. B. Vogt, T. J. Walker,
L. M. Walkley, R. E. Warner, D. M. Weisman, F. G.
Werner, R. E. White, S. L. Wood, R. E. Woodruff,
P. W. Wygodzinsky. The work was supported by NSF
grant GB-5867.
Bristowe, W. S. 1958. The world of spiders. Collins,
London. 308 p.
Cotteau, E. 1885. A week at Krakatau. Proc. Geogr. 

Soc. Australia, N.S.W. and Victorian branches, 11.

Glick, P. A. 1939. The distribution of insects, spiders, 

and mites in the air. U.S.D.A. Tech. Bull. 673. 151 p.
MacArthur, R. H., and E. 0.Wilson. 1967. The theory
of island biogeography. Princeton University Press.
203 p.
Meyerriecks, A. J. 1960. Comparative breeding behavior
of four species of North American herons.
Publ. Nuttall Ornithological Club, No. 2. 158 p.
Mockford, E. L. 1967. Some Psocoptera from plumage
of birds. Proc. Entomol. Soc. Wash. 69: 307-309.
Simberloff, D. 1969. Experimental zoogeography of
islands. 111. A theory of insular colonization. Ecology
(in press).
Wilson, E. O., and D. Simberloff. 1969. Experimental
zoogeography of islands. Defamation and monitoring
techniques. Ecology 50: 267-278.
Wilson, E. O., and R. W. Taylor. 1967. An estimate
of the potential evolutionary increase in species density
in the Polynesian ant fauna. Evolution 21: 1-10.
DANIEL S. SIMBERLOFF AND EDWARD 0. U'ILSOX Ecology, Val. 50, NO.2
APPENDIX. The colonists of six experimentally defaunated islands. Headings identify the island and
give the number of days after defounation for each census ; "Pre" is the pre-defaunotion
census. Solid entries indicate thot a species was seen; shaded, thot it wos inferred to be
present from other evidence; open, thot it was not seen and inferred to be absent.
THE COLONISTS OF ISLAND E 1
THYSANL'RA l.episa\idac Lbplstm3 ap.
ORTHOPTERA Blattidar Lalibl~~tteliun. mp.
GryllidaC CyCloplllu111sp.
TafiUisc~lurid0
DEllhlAPTERA Labiduridau 1~IIidu1.urlpnl.ta
COLEOPTERA Anthicidae Sdpinlus fuivlprs
Curculiunidae Crypturhynchus nlinullsslmus
Psrudoucalles sp.
Gen. so. '
Phlaeothripidae
Thripidae
Peripsocldae Ectopsocus sp.
Perlpsocus paullani
Peripsocus slagnivsgus
Peripsocus sp. c
Psmidae Psocidus texanus
HEMIPTERA Aleyrcdidae Telraleurcdes sp.
Coccidae Cercplaste6 sp.
Miridae Paallus conspurcahis
NEUROPTERA Chrysopidae Chrvsopa collarls
LEPDOPTERA Eucleidae
Hesperiidae Phocides batabano
Lymantriidae Orgygia detrila
Olethreutidae Ecdytolopha sp.
Phycitidae Bema ydda
Ptineldae Nemapogon sp.
Pyraiidae Gen. sp.
Sahirniidae Automeris i c
DlPTERA Hippobscidae Lynchla albipennis
Olfersia sormda
HYMENOPTERA Bethyiidae Scieroderma macrogaster
Chalcidae Brachymeria psyche
Eumenidae Pachodynerus nasidens
Formicidae Camponotus tortuganus
Co1obopsis sp.
Crematogaster ashmeadi
Pseudomyrmu elongatus
Calilephiaites ferrugineus
Casinaria texana
ORDER UNK. Fam. Unk. Gen. sp.
ARANEAE Araneidae Eustala sp. I
Nephila clavipes
Saiticidae Hentzia p a l m a r m
Fam. Unk. Gen, sp.
Argasidae Argas radialus
ASddae Arctoseius sp.
Dermanyssidae Ornunonysaus bursa
Peilonysaus sp.
Galumnldae G a l m n a 8p.
Oribahllidae Scheloribatea sp.
Phytoseiidae Amblyseius sp.
ISOPODA & y 8 ~ 0 h l ~Bp.
DIPLOPODA Lophoproctinus bal?schi
EMBlOPTERA Teratembiidae
ORTHOPTERA Blattidae
YSOPTERA
COLEOPTERA 	 Anoblidae
Anthicidae
Buprestldae
Cerambycidae
Lampyridae 

Oedemeridae 

Scolytidae 

Fam. Unk 

THE COLONISTS OF ISLAND E 2 	 QJ 6 o\ \ \ % 9 @ @ 4 b $
(L c c 4'& s ' 4 0 ~ @ $ \ % \ ! + \ ! ? ? ( i ~
Gen. sp. 

Agiaopteryx sp. 

Latiblanella n. sp. 

LBtiblatlella rehnl 

Cycloptilum ap. 

Qrtoxlpha sp. 

Orocharls gryllodes 

Tafallsca iurlda 

Turpilia rostrata 

Kalotermea Jmteli 

Neotermes castanew 

Tricorynus sp. 

Saplntue falvlpes 

Actenodes auronotala 

Chrysobthrls sexlasclatus 

Chrysobthris tranqllebxic 

Ataxia sp. 

Styloleptus blushis 

Gen, sp. 

Gen. ap. 

Cryptorhynchu~mlnutlestm 

Pseudoacalle8 sp. 

Micronaspie f lorldana 

Copldlta auturalle 

Oxacis sp. 

Orjcople sp. 

Paecilips rhirophorae 

Trlschidlas atoma 

Gen. sp. 

-1y Spring 1969 	 A' OF EMPTY ISLANDS
APPENDIX -PART 2 

THE COLONISTS CIF ISLAND E 2 (corrt.) 

THYSANOPTERA 	 Phlaeothripidae
Thripidae
Fam. Unk.
C Q R R O D E ~ M 	Caeciliidae
Lepidopsocidae
Liposcelidne
Peripaacidae
Psocidae
HEMIPTERA 	 Acandoniidae
Aievrcdldae
Cicadellidae
Cmiidae
Fiatidae
Miridae
Pseudacoccidae
Tropiduchidae
Fam. Unk.
NEUROPTERA Chrysopidae.
LEPIDO~ERA Geometridae
Hesperiidae
Noctuidae
Olethreutidae
Phycitidae
Ptineidae
Pyraiidae
Saturniidae
DlPTERA 	 Cecidomyidae
Culicidae
Syrphidae
Fam. bnk.
m E N O m E R A Bethylidae
Braconidae
Chdcedectidae
Chlcidae
Encyrtidae
Eumenidae
Eupelmldae
Sphecidae
Arane~dae
Liothrips n. sp, 

Pseudothrips inequalis 

Gen. sp. 

Caecillus ap. n-b 

Echmepteryx hageni b 

Gen. sp. 

Perlpsocus sp. 

Psooidus texanus 

Acanalonia latifrons 

Aleurothrlxus sp. 

Paraleyraiea ap. 1 

Paraleyrodes sp. 2 

Tetraleurodes sp. 

Scaphytopius sp. 

Gen. eo. 

Flatoidinus punctatus 

PsaUus conspurcatus 

Pseudococcus sp. 

Neurotmela breviceps 

Gen. sp. 

Chrysopa collaris 

Chrysopa externa 

Chrysopa rufilabris 

Oxydia sp: 

Phocides batabaco 

&lelipotis sp. 

Ecdytolopha sp. 

Bema ydda 

Nemapogon sp. 

Gen. sp. 

Automeris io 

Gen. sp. 

Aedes taeniorhynchus 

Gen. sp. 

Gen. sp. 

Cephalonornia waterstuni 

Heterospilus sp.
Iphiauiax eplcus
Euchrysia sp.
Gen. sp.
Coencyrtus submetallicus
Pachodynerus nasldens
&letapipeimaschwarzi
Neanaslatus sp.
Gen. sp.
Componolus fiortdanus
Camponotus tortuganus
Crematogasler ashmeadi
Monomarium flar~coia
Paracryptocerus varians
Pseudomyrmex elongatus
Pseudomyrmex "nav~dula"
Tapinoma I ~ t t o r d e
Xenomyrmex fjor~danus
Caiilepiualtes ferrugineus
CiiSinaria texana
Trypoxyion coliinurn
Argiope argentah
Eustala sp. 1
E~6tai.asp. 2
Casteracantha elli2soides
Gen, sp.
Nepiuila clavlpes
Anyphaena sp.
A)sha sp.
Gen. sp. I
Gnnphosa sp.
hleionola sp.
UBilus sp.
llenlzla palmarum
hletacj rba undah
Gen. sp. I1
Ariadna arthuri
I.eucaugr venusta
TeLragwlha sp. 

Cen, sp. 

Rhizoglyphus callae 

Gen. sp. 

Choylet~dwelist 

Gen. sp. 

rn1ull)ssus sp. 

(:alulnna gp. 

Scheloribates sp. 

Anit1)srlus sp. 

Telranyehus sp. 

Rhyscutus sp. 

Luphopruct~nuslbarlschi 

DANIEL S. SIMBER.LOFFAND EDWARD O. WILSON Ecology, Vol. 50, No. 2
APPENDIX - PART 3
THE COLONISTS OF ISLAND E 3
COLLEMBOLA ~ c d u r l d a e
ELIBIOPTERA Teralemb~idae
ORTHOPTERA Blattldae
Cryllidae
ISOPTERA Kalctermitidae
COLEOPTERA Ancbiidae
Anthicidae
Bupreatidae
Cerambycidae
Curculionidae
Lathridiidae
Oedemeridae
Scclytidae
THYSANOPTERA Phlaeothripidae
Thripidae
Fam. Unk.
CORRODENTIA Liposcelldae
Peripsocidae
Psocidae
HEMIPTERA Aleyrcdidae
Allthocoridae
Cicadellidae
Lygaeidae
Membracidae
Mlridae
Nabidae
NEUROPTWA Chrys2pidae
LEPDOPTERA Eucleidae
Geomel~idae
Hesperlidae
Olelnreutidae
Phycilidae
Ptineldae
Saturniidae
HYMENOPTERA Bethylidae
Braconidae
Chalcidae
Ency rlidae
Eulophidae
Eumenidae
Eupelmidae
lchneumonidae
Scelionidae
Spnecidae
Vespidae
ARANEAE Arvlcidae
Clubionidae
Salticidae
Segestriidne
Totragnathidae
Thcridiidae
Acarid'au
A acid&=
DcrmvlyssW'
Eupodidae
Calumnidac-
Orihntulidoe
Phyiosoiidau
Cen. ep.
Dlradlus carlbbeana
Latiblanella n. 8p.
Cyclaptilwn speclabile
Tafdiaca lurida
Kalotermes jouteli
Tricorynus sp.
Sapidua fulvlpea
Actenodes auron,tata
Chryaobothris tranquebarica
Styloleptua bQatua
Cryptorhynchua minutiaalmus
Pseudoacalles sp.
Melanophlhalma floridana
Oxacis sp.
mycopis sp.
Trischidiaa atama
Haplcthrips flavipes
Liothrips n. sp.
Neurothrips magnafemoralis
Pseudolhrips inequalis
Gen. sp.
Liposcelis b?slrych~plulus
E~lopsocopsiscryptmeriae
Ectopsocus SP. c
Peripsocus pauliani
Psocidus LeXBlus
Telmleurcdes sp.
Dulouriellus afer
Orius sp.
Scaphylupius sp.
Gen. sp. 2
Gen. sp.
Psallus conspurcatus
Carthasis decnralus
Cnrysopa collaris
Chrysopa externa
Cnrysopa rufllabris
Alarodia si~ssoniae
Oxydia sp.
Pnocldes batabono
Ecdytolopha sp.
Bema ydda
Nemapogon sp.
Aulomeris i~
Nesepyris flol.idanus
Scleroderma macrogasler
Gen. sp.
Heterospilus sp.
Iphiaulax epicus
Macrocentrus sp.
Brachi m e n a psych6
Gen. sp.
Gen. sp.
Enledontini sp.
Melittobia chalybii
Pachodynerus nasidens
Gen. sp. I
Gen. sp. 2
Colobopsis sp.
Crematogaater ashmeadi
Monornorim lloricola
Paracryptocerus varians
Paratrechina bou~.Innica
Pseudomyrmex el,,ngatus
Tapimma lit torale
Xenomyrmex [loridanus
Calliephialtes ferrugincus
Casrnarin lexana
P ~ o ~ Z P ~ C O ~ U SSP. 

Telonomus rp. 

Trypoxylon collinum 

Gen, sp. 

Eustala sp. I 

Gasteracanlna eiiileoidos 

Anypilaena sp. 

Aysha sp. 

S e r k ~ o l wsp. 

Hontzia palmavun, 

A r i d n u arlhuri 

leucauge Mnusla
Telr&gllillha 3111i11S11R

Can. ap. 

Tyrq,haguS Pulresccnliar 

ABCXSP. 

c*n. wp.

EuimJes sp. nr. fusifer 

Gklumnv 91,. 

Schcloriba'es sp. 

Aml>iyscius SP. 

I'oiyxenidae iapnoprocti~luabartschi
I l h y a c ~ t ~ s t i i l i i i i i i i i i i lsp.
Early Spring 1969	 COLONIZA'I:ION O F EMPTY ISLANDS
APPENDIX -PART 4 

THE COLONISTS OF ISLAND ST 2 

EMBIOPTERA Teratembiidae
ORTHOPTERA Blattldae
Gryllldae
ISOPTERA 	 Kalotermitldae
COLEOPTERA 	 Anobiidae
Anthicidae
Buprestldae
Cerambycidae
Curculionidae
Lathridiidae
Oedemeridae
THYSANOPTERA Phlaeothripidat:
Thripidae
CORRODENTIA Llpscelidae
Peripsocldae
Psccidae
HEMIPTERA Aleyrodidac
Lygaeidae
NEUROPTERA 	 Chrysopidae
LEPIDOPTERA 	 Eucloidac
Geometridae
Hesperiidae
Lymantrlidae
Olethreutidae
Phycitidae
Ptineidae
qiraustidae
DIPTERA Fam. Unk. 1
Fam. U*. 2
HYMENOPTERA Bethylidae
Braconidae
Eurnenidae
Eupelmidae
Formicidae
Ichneumonidac
Sphecidac
Apmrita
ARANEAE Araneidae
Salticidae
Segestriidae
Tetragnathidae
Theridiidae 

ACARLNA Ascidae 

Bdellidae
Eupodidae
Oribatulidae
Phytoseiidae
Veigaiidae
Fam. Unk.
CHELONETHIDA Cheliferidac 

PAUROPODA 

DIP1,OPODA I)ol>xonidac 

Diradius caribbeana
Latlblattella n. sp.
Cycloptilurn ap.
Cyrtoxipha sp.
Tafallsca lurida
Kalotermea jouteli
Trlcorynus sp.
Gen. sp.
Gen. ap.
Styloleptua blustus
Cryptorhynchus sp.
Pseudoacallea sp.
Melanophthalrna floridana
mycopis sp.
Haplothrips flavipes
Liothrlps n. sp.
Paeudothrips inequalis
Lipascelis bwtrychophilue
Ectopsocus sp.
Peripsmus pauliani
Peripsmus stagmvagus
Psocldus texanus
Tetraieurodes sp.
Blissus insularis
Gen sp. 2
Chrysopa collaris
Chrysopa externa
Chrysopa rufilabris
Alarodia slossoniae
Gen. ap.
Phocides batabano
Orgygia detrlta
Ecdytolopha sp.
Bema ydda
Nemapogon sp.
Gen. sp.
Gen. sp.
Gen. so.
Scleroderma macrogaster
Iphiaulax epicus
Macrocentrus sp.
Gen. sp.
Brachymerla psyche
Gen. sp.
Pachodynerus nasidens
Gen. sp.
Camponotus floridanus
Colobopsis sp.
Crematogaster asluneadi
Crematogaster atkinsoni
Paracrfltmerus varians
Pseudomyrrnex elongatus
Tapinoma littorale
Xenomyrmex floridanus
Calliephialtes ferrugineus
Casinaria texana
Trypxylon collinum
Gen. sp.
Eusiala sp. I
Eusiala sp. 2
Gasteracantha ellipsoides
Nephila clavipes
Anypha+ na sp.
Aysha velox
Hentzia palmarurn 

Arladna arthuri 

Leucauge venusta 

Tetragnatha sp. 

Thcridion sp. 

Arctoseius sp. 

Asca sp. 

Bdella sp. 

Eupodos sp. nr. fusifcr 

Scheloribates ap. 

Amblyseiue sp. 

Veigaia sp. 

O"", sp. 

Tyrannochelifc!. n. sp. 

Gen. sp. 

laphoproctinus bartschi
DANIEL S. SIM IFF AND EDWARD 0. WILSON Ecology, Val. 50, No. 2
APPEND1X-PART 5 

THE (>OLONISTS OF ISLAND E 7 % e8Q@&d.@$@400~g
4 ? 8 4 4 $ , ! $ ~ + q , ~ ~ + 
+ + t.,
EMUIOPTERA Cen sp.
ORTHOPTERA Latlblattella n, sp.
Cycluplllulll sp.
T a f d I s ~ nlurldn
DERMAPTERA LabMu~.orlpilrla
COLEOPTERA Trlcorynua ap.
Gm. sp.
Artenudus auronolnlv
ChrysoI*thl.l~ tranquebarlco
Gen, sp.
Cerumbycidae Leptostvlus sp.
Styloleptus blii~lus
Cucujldoe Gen. sp.
Curcullunidue 1'seduacallt.s sp.
Oedetnerldoe Cell ap.
Soydm~~snid~~eGen, sp.
Phlaeothrlpldae Haplothrlps flavlpes
Thrlpidae
Archlpccldae
Llposcelldae
Myopsocldae
Psccidae
HEMIFTERA Nabidae
NEUROPTERA
LEPIDOFTERA
HYMENOPTERA
Chalcedmtidae
Chalcldae
Eumenidae
F~rmicidae
ARANEAE
ACARMA
Neurothrlps magnafemor'alla
Pseudothrlps lnequalle
ArcNpsocus pnnama
Embldopsccus latleeps
Upcacells bostrychophllus
Llchenomima ap, b
Gen. sp.
Ectopsocopsls crypi~merlae
EC~MBOCUBmaldronl
Perlpaocus pauliani
Perlpaocus stagnivagus
Perlpsccus sp. b
Perlpsccus sp. c
Psucldus kxanus
Psocldus sp, nr. blslgnabls
Peocidus sp. I
Psocldus sp. 2
Carlhrnis demratus
MetUroplphom~belfragei
Pseudomccus sp.
Chry~opacollaris
Chryeop rufilabris
Phocides batabano
Orgygia detrlta
Ecdytolopha sp.
Bema ydda
Nemapogon sp.
Cephalonomia vaterstoni
Scleroderma macrogaster
Apanteles hemileucae
lphiaulax eplcus
hrchrysia sp.
Cen sp.
Pachodvnerus nasldens
Hypoponera opacior
Paracryptccerus varlanh
Pardrechlna bourbonica
Fneudomyrmex elongatua
Peeudomyrmex "flavidula"
Xenamyrmex noridanus
Calllephialtes ferruglneus
Caslnarla texana
ProbBryconus sp.
Trypoxylon colllnum
Gen sp.
Arglope argentata
Eustala sp. I
Euatala sp. 2
Gaalerwantha ellipsoides
Mangora sp.
Nephlla clavlpes
Aysha sp.
Dlctyna sp.
Seraolus sp.
Dolomedes sp.
Henula palmarum
Maevla vllata
Paraphldippus n a v ~
SloIdl~Bursts
Arladna arthurl
Leuoauge venusta
Tetragnatha antillam
Argyrodes nsphUae
Therldlon adsmuan1
Therldlon ?tropvlftatum
Mlswnenop sp.
Uloborw sp.
Gen sp.
Tyrophsgus putresoenUne
Proeolpelaps hypudnsl
Pro~tolaelaplpygmaew
Ornlthonyseua bursa
O Q l m M .p.
Amblymelw sp.
Oen. sp.
Trlophtydars sp.
Tyrannochellfer ep.
Rhyrotus sp.
OrphnPeul brMUlnnul
rly Spring 1969 	 COLONIZATIOX' OF EMPTY I S L A N D S
APPENDIX-	 PART 6
THE COLONISTS OF ISLAND E 9
ORTHOPTERA Gryllldae 	 Cycloptilum sp. 

Cyrioalpha confuse. 

Orocharis sp. 

DERMAPTERA Labldurldae 	 Labidura riparia
COLEOPTERA Anobitdae 	 Cryplorama minuturn 

Tricoriilus sp. 

S q i n i u s fuivlpes 

Vacusue viclnus 

Actenodes auronoata 

Chryeobothr~st r a n y e b a r l c a 

Canthar~dae Chauliognathus marginatus
Cerambycidae styloleptus biustus
Curculionldae Cryptorhynchus m ~ n u t i s s m u s
P s e u d o ~ c a l l e ssp. 

Lathridiidae Haloparamecus sp 

O e d e m e r ~ d a e Oxacia s p . 

F a m . Unk. Gen, ap. 

THYSANOPTERA 	 P h l a e a l h r ~ p ~ d d e Haplothrips f l a v ~ p e s
Troglomorpha
HEMIPTERA 	 Anthocorldae
C i u l d a e
Alirldae
Pentatom~dae
Fam. Uilli
NEUROPTERA 	 Chrysopidae
LEPIDOPTERA 	 Eucle~dae
Olethreutldae
Phycltldae
Ps,chdae
P t i n e ~ d a e
Pyralidae
Sarurruldae
F a m . Uilli.
DlPTERA Hippobosc~dae
Fam. Unk
EIYMENOPTERA Braconldac
Eulopiudae
Eumerudae
E u p l m i d a e
F o r m i c ~ d a e
Pteromalldae
Sphecidae
Vespldae
ARANEAE 	 Araneidae
Clubionidae
Dictyllldae
Gnaphosidae
Linyphlidae
Lycosidae
Saltlcidae
S ~ f l u d l d a e
Neurathrips magnafemoral%s
Pseudothrips Inequal~s
Caecillus s p np
Lachesdla n s p
Echmepteqn hageru b
Belaphotrmtes okalensls
Embldopsocus l a t ~ c e p s
UPDGCBIISSP. not bostr)chophilc
Ectopsmus sp. tv
P e r i p s w u s stagmiagus
Psocldus texanus
P s o c ~ d u ssp. I
Gen s p
Dufouriellus afer
O l ~ a r u ss p
P s a u u s conspurcatus
oebaius pugnar
Gen, sp.
Chrysopa collails
Chrysopa enterna
Chrysopa r u f l l a b r ~ s
Alarodla slossoruae
E c d i i o l o p k sp.
Bema ydda
Olket~cusabbottu
Nemapogon sp.
Tholeria reversal18
A u t o m e r ~ ~LO 

Gen. Q.
Oifersla Bordida
Gen, sp.
Apanteles hemtleucae
Apanteles margmlventns
Calldiormms blfasclatus
Ecphylus n. sp. nr. c h r a m r s i
lph~aulanepicus
Gen. 8p. 1
Gen, sp. 2
Gen. sp. 3
Gen, sp. 4
Euderus sp.
P a c h o d ~ n e r u snaaldens
Gen. sp.
Brachymyrmex s p
Camponotus U a r ~ d a n u s
Camponotus sp.
Crematogaster ashmead,
hionomorrum floncoia
P a r a c r > p t m e r u s v a r i a m
P ~ e v d o m ) r m e xelongatus
Pseudomy rmex "llaiidula"
Taplnoma llttorale
Xenomyrmex floridanus
Gen sp.
Calliephlallra f e r r w i n e u s
C a s h a r m tenana
Uroleple r u f l p s
Trypoxilun colllnum
Polisles s p
Arglope ar~enlala
Eriophora s p .
Eu~Lalasp.
Casteracantha elllpsi,ldas
Mctepeira l a h I hlheu
Nephila clauij,c~s
Aysha s p .
Dictyna s p
Serglolus s p .
Meioneta s p
M r a t a sp.
Henkia palmarum
scytudee sp.
VIEL s. s RLOFY Ecology, Val. 50,
APPENI31X - PART 7
THE COLONISTS OF ISLAND E9 (cont.)
Leucauge venusta
T e t r a g ~ t h aantUlana
Tetragnatha sp 2
Therldlidae Cen. ap.
Ulobor~dae Uloborw sp.
Argasidae Argae radlatw
Ascldae La8108dY6 Bp.
Mellchares sp.
PdeUa sp.
Carpoglyphus laclls
Sphaerolophus sp.
CallllllM BP.
Amblyaeiw ep
ISOPODA Rhy8~0U8Bp. 

CHILOPODA Orphnaeua brasillanus 

EXPERIMENTAL ZOOGEOGRAPHY O F ISLANDS : A MODEL 

FOR INSULAR COLONIZATION 

DANIELS. SIMBERLOFF~
Biological Laboratories, Harvard University, Cambridge, Massachusetts 02138
(Accepted for publication December 22, 1968)
Abstract. A distinction is made between immigration rate (in spp./time) for an island,
and invasion rate (in propagules/time) for a species and an island. An analogous distinction
is drawn between an island extinction rate and a species extinction rate (or intrinsic probability
of extinction in a given time interval). It is claimed that the most objective definition
for "propagule" is any animal or group capable of population increase under any conceivable
circumstances. Immigration and island extinction curves are unique only if plotted against
time, not against number of species.
A model for non-interactive colonization is discussed, and its equilibrium number of species,
3 derived. It is shown that data from the defaunated Florida Keys can be interpreted as
arising from non-interactive colonization to an S near the non-interactive 3 (and to a point
above a more enduring interactive S ) , followed by a slow decline in S (as population sizes
and interaction increase) to an enduring S near that obtaining before defaunation.
The effect on this scheme of increasing distance from source area is shown, and a simulation
of the non-interactive part of the scheme was performed which corroborates that part of
the model.
Explicit equations for the "expected" colonization, immigration, and island extinction curves
are given for non-interactive colonization, and predictions are given about the general changes
in these curves as interaction becomes significant.
SYMBOLSUSEDI N THE TEXT i ,(a) probability that species a invades in time
C'(t) colonization rate, in species/time, at time . period of length z.
f z , invasion rate of species a, in propagules/
e,(a) probability that species a, if present, is time
extinguished in time period of length z. I ( t ) immigration rate for island, in species/
P , species extinction rate of species a, de- time, at time t
lim I ( contribution to immigration rate of spefined
analogously to i, (q.v.) ; cies a ; defined in text+
e , (a) lim P number of species in species pool
z+o. S ( t ) number of species on island at time t
3 equilibrium nurnher of species for island
If species a were replaced immediately S, ( t ) species indicator variable for species a
upon extinction, e ,would be the frequency = 1 if species a present on island durof
extinctions/time. ing time period t
E ( t ) extinction rate for island, in species/time, =0 otherwise
at time t
E, (t) contribution to island extinction rate of
species a ; defined in text A discussion of theoretical zoogeography must
Present address : Department of Biological Science, be based on clear definitions of the following often
Florida State University, Tallahassee, Florida 32306 ambiguous terms :
You have printed the following article:
Experimental Zoogeography of Islands: The Colonization of Empty Islands
Daniel S. Simberloff; Edward O. Wilson
Ecology, Vol. 50, No. 2. (Mar., 1969), pp. 278-296.
Stable URL:
http://links.jstor.org/sici?sici=0012-9658%28196903%2950%3A2%3C278%3AEZOITC%3E2.0.CO%3B2-U
This article references the following linked citations. If you are trying to access articles from an
off-campus location, you may be required to first logon via your library web site to access JSTOR. Please
visit your library's website or contact a librarian to learn about options for remote access to JSTOR.
Literature Cited
Experimental Zoogeography of Islands: Defaunation and Monitoring Techniques
Edward O. Wilson; Daniel S. Simberloff
Ecology, Vol. 50, No. 2. (Mar., 1969), pp. 267-278.
Stable URL:
http://links.jstor.org/sici?sici=0012-9658%28196903%2950%3A2%3C267%3AEZOIDA%3E2.0.CO%3B2-J
An Estimate of the Potential Evolutionary Increase in Species Density in the Polynesian Ant
Fauna
Edward O. Wilson; Robert W. Taylor
Evolution, Vol. 21, No. 1. (Mar., 1967), pp. 1-10.
Stable URL:
http://links.jstor.org/sici?sici=0014-3820%28196703%2921%3A1%3C1%3AAEOTPE%3E2.0.CO%3B2-N
http://www.jstor.org
LINKED CITATIONS
- Page 1 of 1 -