1 1-BULLETIN
of the

FLORIDA STATE MUSEUM
Biological Sciences

Volume 22 1977 Number 4

ECOLOGY AND BEHAVIOR OF THE JAMAICAN WOODPECKER

ALEXANDER CRUZ

.

UNIVERSITY OF FLORIDA GAINESVILLE



Numbers of the BULLETIN OF THE FLORIDA STATE MUSEUM, BIOLOGICAL SCIENCES,
are published at irregular intervals. Volumes contain, about 300 pages and are not necessarily
completed in any one calendar year.

OLIVER L. AUSTIN, JR., Editor

RHODA J. RYBAK , Managing Editor

Consultants for this issue:

OLIVER L. AUSTIN, JR.

JEROME A. JACKSON

Communications concerning purchase or exchange of the publications and all manuscripts
should be addressed to: Managing Editor, Bulletin; Florida State Museum; University of Florida;
Gainesville, Florida 32611.

This public document was promulgated at an annual cost of $2,193.63 or
$2.193 per copy. It makes available to libraries, scholars, and all interested
persons the results of researches in the natural sciences, emphasizing the
circum-Caribbean region.

Publication date: December 21, 1977 Price: $2.25



ECOLOGY AND BEHAVIOR OF THE JAMAICAN WOODPECKER

ALEXANDER CRUZ'

SYNOPSIS:  Jamaican Woodpeckers (Melanerpes mdiolatus) occur wherever habitats are suitable,

from sea level to high elevations. Their presence depends on the occurrence of trees necessary for
feeding, nesting, and roosting. Highest densities were recorded in a wooded pasture and a
mesophytic forest, with an average of 22 and 20/km, respectively. The high figures for these
locales  are possibly related to their structural complexity-well developed vertical stratification,
high tree species diversity, and numerous epiphytes, all of which increase both the area available
for foraging and the amount of foods present. Censuses in Jamaica and Florida indicate that in
some habitats the Jamaican Woodpeckers maintain comparable or higher densities and biomass
than do several species of Florida woodpeckers in comparable habitats. Possible explanations
are: Jamaican Woodpeckers have more resources available in the absence of other species with
similar habitats, Jamaica has fewer predators than Florida, Jamaican habitats are more complex
than comparable Florida habitats, and primary productivity is probably greater in Jamaica than
in comparable Florida habitats.

No selection pressure for differential niche use between the sexes appears to have acted on
the Jamaican Woodpecker-the sexes are structurally monorphic, forage in similar fashions,
are syntopic, and take the same food. Food and foraging sites, which would be the two most
important selective forces, appeared to be more abundant in Jamaica than on some other islands.
When food is plentiful, both sexes profit by having a longer bill, for large predators (implied
here by longer bill lengths in birds) eat either an equal or a greater range of foods than smaller
predators. Bill size is also a go6d indicator of food size: birds with longer bills usually obtain
larger prey items than birds with smaller bills. As greater prey size implies greater biomass, which
in turn implies more calories and more energy, the sexes are thus equally efficient in obtaining
energy from the environment. In addition one might argue that by decreasing both the body
size and bill size of the female, the female will be less efficient in protecting the nesting hole
against competitors, of which several species are present on Jamaica.

The predominant foraging methods of the Jamaican Woodpecker are fruit-eating (28%),
probing (28%), and pecking (2096)  Cleaning, sallying, and probing into bromeliads accounted
for the remainder.

In the Jamaican Woodpecker's diet both animal and vegetable matter are well represented,
comprising 58.2% and 42.7%, respectively, of the total volume. The large and varied numbers
of foods taken strongly suggest that the Jamaican Woodpecker is diverse and opportunistic in its
feeding habits. The foraging behavior diversity of the Jamaican Woodpecker and of seven species
of Florida woodpeckers was measured by the Shannon-Weaver information theory formula.
The foraging diversity index of the Jamaican Woodpeeker was higher than any one species of
the Florida woodpeckers studied, and was almost equivalent to the pooled foraging diversity
indices of Florida woodpeckers, 1.59 to 1.72 respectively. No methods of feeding were noted
in the Jamaican Woodpeckers that the mainland woodpeckers did not use, but the foraging
methods were more evenly distributed in the different categories. There is no evidence that
limitations in the range or amounts of available food resources is the factor underlying these
changes in feeding methods, rather certain zones are incompletely exploited by other species
in Jamaica; and hence it is profitable for M. mdiolams to extend into them.

Jamaican Woodpeckers occupy overlapping home ranges and territorial defense is restricted
to the nest vicinity. The extended breeding season of at least 10 months is most likely a response
to the more uniform tropical climate that provides sufficient food throughout the year.

' The author is Assistant Professor of Biology, Department of Environ,nentill, Population, and Organi:mic Biology, University
of Colorado, Boulder CO 80309. Portions of this paper were sitbinitted earlier to the University of Florida in partial ful-
fillment of the Ph.D. degree (1973).

CRuz, Al.ExANDER. 1977. Ecology and Behavior of the Jamaican Woodpecker. Bull. Florida
State Mus., Biol. Sci. 22(4):149-204. -



150 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

TABLE OF CONTENTS

INTRODUCTION 150
ACKNOWLEDGEMENTS 152
METHODS OF STUDY 152
DESCRIPTION OF STUDY AREAS 155
RESULTS AND DISCUSSION - 159

DISTRIBUTION . 159
HABITAT ANALYSIS 160
ABUNDANCE 162
FEEDING ECOLOGY AND BEHAVIOR. 165

BREEDING BEHAVIOR AND BIOLOGY 184
VOCALIZATIONS. - 184
MECHANICAL COMMUNICATION 185
D]SPLAYS.  186
TERRITORY, TERRITORIALITY, HOME RANGE, AND

NEST HOLE COMPETITION . 186
THE ANNUAL BREEDING CYCLE 191

LITERATURE CITED.. , 202

INTRODUCTION

The study of insular biogeography and ecology has contributed signifi-
cantly to the development of evolutionary and ecological theories and con-
cepts. By studying species on individual islands or groups of islands, biologists
view a simpler mierocosm of that seemingly infinite complexity of continental
areas. By their very multiplicity and variation in shape, size, degree of
isolation, and ecology, islands provide the necessary replications in natural
experiments by which evolutionary and ecological hypotheses can be tested
(MacArthur and Wilson 1967).

One of the basic tenets of island biology is that species diversity of island
habitats is low in comparison to similar habitats on mainland regions (Darwin
1859, Wallace 1880). This concept has provided important opportunities f6r
ecological investigations on competition, niche exploitation, and behavior
patterns. Preston (1962), Hamilton and Rubinoff (1963), and MacArthur and
Wilson (1963) have demonstrated that the number of species on an island
depends upon diverse factors, the most important of which are: (a) the island's
size, (b) its distance from a source of additional species, and (c) the structure
of the habitat, which includes both topographic variation and the number of
vegetational strata. MacArthur and Wilson (1967) expanded the above factors
into an elaborate mathematical model of island biogeography, based on the
tenet that the number of species of an island represents an equilibrium be-
tween immigration and extinction rates, the level of which depends primarily
on the above factors. In the Greater Antillean region for instance, Cuba (the
largest and closest island to the continental mainland) supports five resident
species of woodpeckers, Hispaniola (the second largest) has two species, and
the smaller islands of Jamaica and Puerto Rico each have one species. In con-
trast, 79 species of woodpeckers are found in South America and 39 species are
found in North America (Meyer de Schauensee 1964).



1977 CRUZ: JAMAICAN WOODPECKER 151

Important studies of insular birds include those of Crowell (1961, 1962),
Selander (1966), and MacArthur, Diamond, and Karr (1972) on Bermuda,
Hispaniola, and Puercos Island respectively. Crowell's studies of three passer-
ine bird species resident on both Bermuda and in the eastern United States
demonstrated that absence of competition on the species-poor island of
Bermuda has allowed these species to attain greater densities than they do in
North America, although the total range of habitat use and feeding behavior
lies within the range of their abilities in North America. Selander found that
the Hispaniolan Woodpecker.(Melane,pes striatus) has attained the ability to
subdivide and perhaps expand the total feeding niche use of the population
by evolving sexual dimorphism in the feeding apparatus (bill) accompanied by
divergenee in the foraging behavior of the sexes. In the more recent study on
the Puercos Island avifauna, MacArthur et al. found that niche shifts between
island and mainland (Panama) birds included habitat expansions, wider ranges
of vertical foraging strata, and increase in numbers.

To test these important concepts, niche expansi6n and population den-
sity changes in the absence of related species and differential niche utilization
between the sexes, and to examine the breeding biology and behavior of an
insular species of bird, I studied the endemic Jamaican Woodpecker (Melan-
erpes radiolatus). Data for this study were obtained on six trips to Jamaica
during the winter, spring, and summer (14 June-17 August 1969, 20-29 De-
cember 1969, 14 April-24 May 1970, 14 june-28 July 1970, 4 June-24 june
1971, and 7 june-15 June 1972). The main ecological emphasis was on
various parameters of the Jamaican Woodpecker's niche. Hutchinson (1957)
defined the niche as a multidimensional space with each parameter corres-
ponding to a different requirement of the species. My study is a critical
examination of parameters related to population size, habitat preferences,
foraging patterns, niche expansion in the absence of other woodpeckers and
birds of similar foraging methods, sexual differences in niche use and intra-
and interspecific competition. My investigation of the Jamaican Wood-
pecker's breeding biology covered all aspects of the annual reproductive
cycle, from pre-pairing to fledging of the young. Behavioral information was
obtained on a variety of subjects, including foraging behavior, vocalizations,
displays, agonistic behavior, territoriality, and reproduction. This is of interest
because, although information is available on the breeding biology and be-
havior of the North American congeners (e.g. Melaneges carolinus and M.
aurifr(ms), little information is available on the insular species of this genus,
specifically M. radiolatus.

Investigations of Florida woodpeckers (Colaptes auratus, Dryocopus pilea-
tus, Melanerpes carolinus, M. erythrocephalus, Dendrocopos villosus, D.
pubescens, and D. borealis) were undertaken during the spring and summer
seasons of 1969-1972, mainly to compare the foraging behavior and popula-
tion densities of mainland woodpeckers with those of the Jamaican Wood-
pecker.



152 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

ACKNOWLEDGEMENTS

I am indebted to the members of my supervisory committee, Archie F. Carr, George W.
Cornwell, Dale H. Habeck, David W. Johnston, and Thomas H. Patton, for their encouragement
and helpful criticisms during the course of this study. I am especially grateful to David W.
Johnston, not only for his guidance and assistance as my supervisory committee chairman, but
for the many helpful discussions of the·projeet, and to Thomas H. Patton and Joshua C. Dickin-
son for making available the resources and facilities of,the Florida State Museum.and Worthy Park
Field Station, without which field research in many·parts of Jamaica would have been impossible.
A number of people participated ia field work-particularly Philip Clarke, Audrey Downer,
Neil Jones, David W. Johnston, Jean Klein, T. H. Patt6n, Roger Smith, Robert Sutton, Lisa
Salmon, and Michael Winegar. C. B. Lewis, Director of the Institute of Jamaica, was very
helpful and made available the Institute facilities. C. D. Adams (University of West Indies) and
George Proctor (Institute of Jamaiea) aided in the identification of plant  material, and john
Carrol, Dale H. Habeck, Jonathan Reiskind, Joel Rodriguez, and Fred.G. Thompson aided with
the identification of animal material. The Clarke family of Worthy Park not only provided
strategic lodging, but their friendly hospitality made the visits more pleasant.

Support during this investigation came from a National Institute of Health Grant awarded
to T. H. Patton, American Philosophical Society Grant to David W. Johnston, and a Frank M.
Chapman and Ford Foundation Fellowship awarded to me. To all the people of Jamaica,
known and unknown, who gave me indispensable aid, this report is especially dedicated.

METHODS OF STUDY

DISTRIBUTION, HABITAT PREFERENCES, AND ABUNDANCE

I tried to analyze each habitat occupied by the Jamaican Woodpecker to see what components
of the community might be consistently present in each of the places visited in order to de-
termine its habitat preferences and requirements. Some of the factors considered included:
ground cover, shrub layer, canopy height, tree species diversity, tree sizes, presence or ab-
sence of dead trees, and epiphytic growth.

The principal study areas were censused to obtain comparative data on woodpecker popula-
tion densities. In addition, censuses were made on mainland woodpeckers in Florida in various
communities to compare population densities and biomass between mainland and insular wood-
peckers. The Florida counts were also supplemented by the published literature, but no published
counts were found for the Jamaican Woodpeckers. Two census methods were used in jamaica:
(1) a linear strip count for all birds heard or seen along preselected routes, and (2) a census plot
count for all birds in a given area. In Florida only the latter method was usied. Censuses were
conducted on foot during the morning when woodpeckers are usually most active. In each of the
census areas, at least three counts were taken and then averaged.

FORAGING AND FEEDING METHODS

Detailed foraging and feeding observations of the Jamaican Woodpecker in the principal
study areas were obtained during the spring, summer, and winter, but chiefly during the breeding
season (spring and summer). Observations were carried out. at all times of the day, although it
was found that the woodpeckers' fed most frequently in the morning and late afternoon. The
specific technique involved walking along an undetermined path in the study area until a
w6odpecker was encountered. If the bird was foraging, information was recorded for foraging
height, behavi6r, and zones used. The tree was divided into three main feeding zones, trunk,
inner branches, and outer branches. Each of these main zones was in turn divided into three
subzones (Fig. 1). Sex was also recorded. Observations on the foraging behavior of Florida
woodpeekers were undertaken during the spring and summer of 1969-1972 in approximately
the same fashion.

The percentage of the total number of times the woodpeekers were recorded in each dis-
crete foraging zone was used to estimate the frequencies with which the woodpeckers used
each of these zones. These percentages were calculated by summing the observations recorded in
a particular foraging zone and dividing the total by the sum total of observations from all
zones.

Feeding behavior patterns of the woodpeckers were,categorized as follows:



1977 CRUZ: JAMAICAN WOODPECKER 153

396

8 5

27 4

. 1

FlcuRE 1.-Diagram of foraging zones. Numbers correspond to the following zones: (1) Lower
trunk, below lateral branches of crown; (2) Middle trunk, lower half of trunk within region of
crown; (3) Upper trunk, upper half of main stem within region of crown; (4) Proximal lower
1/3 of inner branches; (5) Proximal middle 1 /3 inner branches, (6) Proximal upper 1/3 of inner
branches; (7) Distal lower 1/3 of outer branches; (8) Distal middle 1/3 of outer branches;
(9) Distal upper 1/3 of outer branches.

PROBING,-In probing, a bird inserted the bill, tongue, or both into cavities, such as holes,

cracks, and crevices in the bark, weathered holes previously excavated by the woodpeckers, holes
in dead stumps, cracks between trunks, and in accumulations of plant debris.

BROMELiAD-PROBING.-A special category of probing was into bromeliads and other epiphytes.
As will be demonstrated quantitatively later, this special feeding category comprised a signifi-
cant proportion for these woodpeckers.

SEARCHING AND GLEANING,-In searching and gleaning, the birds moved slowly along limbs

and trunks, actively scanning the bark, epiphyte-covered branches, and clumps of leaves for
animal prey, which were picked up with the bill and tongue.

FRuIT-EATING.-In this category are included the use of berries, fruits, and seeds as a food.

The usual foraging procedure involved taking of fruits from a perching position and, less
commonly, by hanging on the elumps of fruits.

PECKING.-Pecking (exeavating) involved striking an object sharply with the bill. Usually it

is a repetitious activity, with several blows delivered in rapid succession. In addition to pecking,
the woodpeckers sometimes excavated in wood by prying off flakes of bark, especially in spots
where the bark was rotten.

SALLYING,-Both bird and prey are on the wing at the time of capture.
The percentage of the total number of observations the Jamaican Woodpecker was in each

foraging procedure was used to estimate the frequency with which the woodpeckers used



154 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

each of these methods. These percentages were calculated by summing all the observations
recorded.in a particular foraging method and then dividing the total by the sum of observations
for all zones.

Foraging method is usually indicative  of. the food items sought. Although in some cases I
was not able to see the food items taken, those identified were recorded, Observations on the
food and foraging ecology of other bird species in the study areas were also recorded to de-
termine the degree of feeding niche overlap and whether competition might exist for any
particular food source.

FOOD ANALYSIS

In addition to obtaining information on the foraging behavior and zones the Jamaican Wood-
pecker used, 29 adults were collected and the stomach contents examined. The primary purpose
was to determine foods taken, how the food items in the stomach correlated with foraging pat-
terns, and to determine if any differences existed between the sexes with respect to food type.
All woodpeckers were collected in the Worthy Park area in the summer of 1969, spring of 1970,
and summer of 1971. The stomach and intestinal tract were removed soon after death and pre-
served in 75% alcohol. Later the food samples present were initially separated into food classes
or groups (animal and vegetable) and analyzed both by volume and frequency of occurrence.
The reason for using more than one method of food analysis is that the different methods pro-
vide indications of different aspects of the feeding ecology of the species. A food class with a
high occurrence indicates that the food is consistently available (and attractive) to the bird, and
so is a reliable food source. The frequency of occurrence alone gives little indication of the
importance of each food in the birds' diet as it ignores the size of the-food item. This problem
is overcome by measurements of the volume of the various food classes in each sample to obtain
indications of the relative general importance of the different food classes in the birds' diet. In
addition the length of each prey item was measured to detefmine the importance of the various
classes of prey size and to see if the sexes differed in prey sizes eaten.

In the volumetric analysis method, food volume was ascertained with reasonable accuracy
by noting the displacement of water in a graduated cylinder accurate to 0,1 ml. The volume
of the total items in each of the food and prey size classes was summed for each class, and
expressed as percentage of the total volume of the food and prey size in all classes. The fre-
quency of occurrence of each food class is presented as the percentage.of samples in which the
class.was represented.

MORPHOLOGICAL ANALYSIS

Morphological data were obtained by standard mensural methods to see if any sexual dimor-
phism in body structures of possible ecological significance existed. Bill length was measured
from the anterior margin of the nostril to the tip; the tarsometatarsus wasi measured from its
posterior proximal end to the distal edge of the most distal unbroken scale crossing the bases
of the two forward toes; and the outer front toe (number 3) was measured from its proximal end
to the distalmost scute (not including claw). Body weights were obtained on all the specimens
collected. Foot volume was measured in a graduated cylinder accurate to 0.1 ml. Linear measure-
ments are in millimeters, weights are in grams, and food volumes in ml. All weights and measure-
ments were usually taken on the day the bird was collected.

BREEDING BEHAVIOR AND BiOLOGY

Field studies of breeding Jamaican Woodpeckers were conducted primarily at the Worthy
Park (Lluidas Vale) field station during the summer of 1969, spring and summer of 1970, and
summer of 1971. The part of the annual cycle covered by the above field work extended from ~
early April to late August. Some additional field work was conducted during December of 1969.

The primary method of study was by direct observation and recording data on breeding
biology, vocalization, territoriality, and inter- and intraspecific behavior. To record the activities
inside the nesting hole a rotatable mirror attached to the end of a rod and with a built-in light
source was used. Reproductive condition was noted for all collected specimens.



1977 CRUZ: JAMAICAN WOODPECKER 155

To investigate factors involved in species recognition, territorial behavior, and territory size,
I performed a series of experiments in which dummy male Jamaican Woodpeckers (study
skins) were placed at varying points from the nesting hole, and the reaction thereto of the
breeding pair was recorded. Presumably the points where the occupant pair fail to show aggres-
sion towards the visual representation of another woodpecker marks the boundary of the ter-
ritory. These experiments were. supplemented by noting the reactions of contiguous family
groups of woodpeckers toward one another, and the reactions of nesting woodpeckers to arti-
ficial drumming produced by tapping a clip-board with a pencil. Home range size in the Jamaican
Woodpecker was determined by measuring the distance traveled from the nesting  hole by 12
breeding pairs in Worthy Park during the summer of 1969. The distances traversed were
plotted on a field map, and the home range boundaries were determined by drawing a line
through the extreme points where the woodpeekers were recorded (modification of the method
used by Odum and Kuenzler 1955).

DESCRIPTION OF STUDY AREAS

PHYSIOGRAPHY

jamaica lies at 18° N. Lat; in the western Caribbean, approximately 150
km south of Cuba and 200 km west of Hispaniola. The nearest mainland is
Honduras, approximately 610 km southwestward. With an area of 11,740
sq km, Jamaica is the third largest island in the West Indies, exceeded in size
only by Cuba and Hispaniola. Most of Jamaica is mountainous, with more than
one-half of the island over 305 m (1000 ft) in elevation. The greater part of
this area comprises the Central Upland plateau at 600-915 m, and the Blue
and John Crow Mountain ranges in the east. The Blue Mountains, the highest
in Jamaica (maximum height 2155 m), extend westward from the John Crow
range in the eastern part of the island to Mount Telegraph (1275 m), ap-
proximately one-third the length of the island. In the Central Upland plateau
are the Dry Harbor Mountains (St. Ann Parish) and the Mocho Mountains
(Clarendon Parish), separated by the Main Ridge group. Farther westward is
the Cockpit country, a succession of cone-like hills with alternating en-
closed conical depressions or "cockpits," typical karst country with under-
ground drainage, subterranean rivers, sinkholes, and caves. Mount Diablo
(1000 m), in St. Ann and St. Catherine parishes, is a precipitous, calcareous
plateau, similar to the Cockpit country. Dolphin Head (542 m) is an isolated
limestone peak on the western end of the island. Along the southern coast of
Ja~aica from Morant Bay (eastern Jamaica) westward to Portland Ridge, a
distance of over 80 km, lies an intermittent line of limestone hills, mostly under
300 m in elevation.

CLIMATE

Rainfall is the most important single factor affecting the vegetation. The
island lies in the path of the moisture-laden easterly trade winds that blow
throughout the year. The uplifting and cooling of the winds causes conden-
sation that strikes first the high limestone John Crow Mountains and then the
northern flank of the Blue Mountains. Hence Portland in the northeast is one



156 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

of the wettest parishes with an average annual rainfall of 381 cm. In con-
trast, the dry southern parish of St. Andrew has only 89 cm of rain. Another
feature of the rainfall is its seasonal periodicity. Rainfall records for the island
indicate that the heaviest rainfall occurs in May and June and again from
August to November. The major dry period is from January to March (Asprey
and Robbins 1953).

VEGETATION AND STUDY AREAS

The vegetation of Jamaica, except in the higher montane region, is tropi-
cal. In general the average annual temperature decreases with increase in
ground elevation, whereas rainfall tends to increase from south to north and
also with elevation. The annual mean surface temperature is 26°C at Kings-
ton on the southern coast and 13°C at the Blue Mountain Peak (2255 m).
These conditions (rainfall, elevation, and temperature) operate to bring about
a very diverse vegetational pattern. The description of the following major
communities and principal study areas follows the terminology of Asprey and
Robbins (1953).

STRAND WOODLAND ASSOCIATION.-This association is characteristic of
coastal Jamaica and occurs on both coral and sandy substrata. Characteristic
trees are sea grape (Coccoloba uvifera) and seaside mahoe (Thespesia popul-
nea). This is a low (3-6 m) scrubby, open community which may include
palms and a mixture of other trees and shrubs. Field investigations on wood-
peckers in this community were conducted at Morant Point (St. Thomas
Parish), Negril (Westmoreland Parish), near Falmouth (Trelawny Parish),
and various points in St. Elizabeth Parish.

COASTAL PLAINS PLANT COMMUNITIES.-Large low-lying coastal plains
extend along the dry southern coast. They once supported several types of
seasonal evergreen or deciduous forests. Man's activities have now produced
successional communities that may lead to a secondary savanna type forest
consisting primarily of mesquite (Prosopis iutiflom), acacia (Acacia lutea), or
logwood (Haematoxglum campechianum). Much of the area is under cultiva-
tion, primarily for sugar cane, bananas, and coconuts. Where irrigation is
impractical, the land is used for grazing and has a savanna-like appearance
that consists mainly of guinea grass (Panicum maximum) and guango trees
(Samanea saman). Field investigations in the coastal plain were undertaken
in various localities in the parishes of St. Thomas, St. Catherine, Clarendon,
and St. Elizabeth.

MANGROVE WOODLAND.-In protected coastal areas where silt is deposited,
mangrove woodlands develop in which four New World mangrove species are
present. The red mangrove (Rhizophora mangle) usually forms pure stands on
the seaward side, whereas white mangrove (Laguncularia racemosa), black
mangrove (Avicennia germinans), and buttonwood (Conocarpus erecta) are
characteristic of mud swamps and occur farther inland. One of the principal



1977 CRUZ: JAMAICAN WOODPECKER 157

study areas, near Falmouth (Trelawny Parish), was in this community.. Charae-
teristic trees of this mangrove community included white, black, and button-
wood mangrove. Red mangrove was not present. The forest was low, the trees
averaging approximately 4.5 m in height.

DRY LIMESTONE FOREST.-Asprey and Robbins (1953) recognized two dis-

tinct vegetational types on limestone rocks: dry limestone forest and wet
limestone forest. Dry limestone forest occurs where the annual precipitation
is less than 101 cm and is best represented along the southern coast of Jamaica,
but also occurs at Negril on the extreme western end of the island and in
the parish of Trelawny on the northern coast, Asprey and Robbins noted: "Dry
limestone is a sparse, vegetation cover of low forest and tall scrub growing
on bare limestone rock. No soil is present except for that deposited in small
crevices or washed down to level areas. Leaf litter is almost nil and the floor
is either a jumble of broken stones or a more or less continuous mass of jagged
honey comb rocks." There is no distinct stratification and heights vary from
low scrub to a thin forest with trees rarely exceeding 9 m in height with
occasional emergence of red birch (Bursera simamba) and cotton tree (Ceiba
pentrandra) up to and over 18 m. Many of the tree species are semi-
deciduous during the dry season. Field investigations in this community were
conducted at Portland Ridge (Clarendon Parish), Hellshire Hills (St. Catherine
Parish), and near Discovery Bay (St. Ann Parish). Portland Ridge (152 m in
elevation and one of the principal study areas) forms the western end of the
intermittent line of limestone hills that occur along the southern coast from
Morant Bay westward, a distance of approximately 80 km. This type of
forest has been subject to much human interference in jamaica, but the
Portland Ridge region is relatively undisturbed. Some of its characteristic trees
are red birch, torchwood (Am!/ris basamifera), thatch palm (Thrinax parui-
flora), burnwood (Metopium browni), and dildo (Cephalocereus spp.) among
many others.

WET LIMESTONE FOREsT,-This community is developed on limestone
rock where the rainfall is over 190 cm and may be as high as 380 cm. Most
of this forest grows inland at elevations from 300-760 m. It is more mesophytic
and luxuriant than the dry type, with more tree species, epiphytes, lianas,
aroids, and bromeliads. Ground vegetation and leaf litter are more evident,
although soil might be absent on the hillsides but deep in the valley. The
canopy is dense and twice as tall as the dry limestone forest, with emergent
trees up to 30 m or more. Investigations in this community were conducted
in the Lluidas Vale (Worthy Park) and Mount Diablo area (St. Catherine
Parish), Cockpit country (Trelawny Parish), and Dolphin Head Mountain
(Hanover Parish). Investigations in Worthy Park (one of the principal study
areas) were conducted both in forests and upland pastures at elevations
ranging from 370 m in the valley to 950 m in the surrounding hills and
mountains. Some of the characteristic trees are broadleaf ( Tenninalia /«ti-



158 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

folia), Jamaican cedar (Cedrela odorata), sweetwoods (Nectandra spp.), bullet-
woods (Daphnopsis spp.), prickly yellow (Fagam martinicensis), and figs (Ficus
spp.). In the upland pastures where some of the original vegetation has been
removed, characteristic trees also include guango, pimento (Pimento offici-
nalis), trumpet tree (Cecropia peltata), logwood, and citrus trees (Citrus spp.).
Many of the trees in this region support epiphytes, bromeliads, and lianas
growing in profusion.

LowER MONTANE RAIN FOREsT.-Floristically some relationships exist be-
tween the lower montane rain forest and the wet limestone forest, but the
former occurs at higher elevations (up to 1070 m). Annual rainfall exceeds 250
cm and may reach nearly 700 cm. Although this community has been subject
to much human disturbance, lower montane rain forest is still to be found
in the Blue and John Crow mountains., Investigations in this community were
conducted near Hardwar Gap (St. Andrew) and Corn Puss Gap (St. Thomas
Parish). Corn Puss Gap, one of the principal study areas, lies at an elevation of
610 to 686 m on the southwestern slopes of the John Crow Mountains. This
forest consists of many of the tree species found in the wet limestone forest
and has a similar appearance. Characteristic trees of this region also include
santa maria (Catophyllum lacquinii), rodwood (Eugenia sp.), mountain guava
(Psidium montanum), coby wood (Matal/ba apetala), and many others. The
tree fern (C!/athea) is also characteristic of this area.

MONTANE MisT FoREST.-One of the few tracts of original vegetation in
Jamaica is the montane mist forest covering the upper reaches of the Blue
Mountains, which form the central mountain system in the island's eastern
end. It is a region of high atmospheric humidity with a high annual rainfall
(over 200 em) and mist covers it almost continuously. The mist forest is low-
eanopied, seldom exceeding 12 m. Much variation is found in the structural
and floristic composition of the mist forest, and variations are almost entirely
correlated with degree of exposure. The chief variation is between that of the
deep sheltered ravines and that of the exposed ridges. In both structure
and floristics the montane mist forest shows some temperate forest features
(Asprey and Robbins 1953). Hardwar Gap, one of the principal study areas,
lies at an elevation of 1220 to 1373 m in the Port Royal Mountains, a range
subsidiary to the Blue Mountains. Dominant trees in this community are yacca
(Podocarpus urbani), bloodwood (Cyrilla racemiflora), jumba (Alchornea lati-
folia), and wild fig (Clusia spp.). Undergrowth shrubs and lower order plants,
such as mosses and ferns, including the tree fern, are also abundant.

ELFIN WOODLAND.-In Jamaica elfin woodland grows on the exposed
summits and northern ridges of the Blue Mountains at 1525 m and higher.
It is an open woodland of gnarled and twisted trees, often short, windblown,
and laden with mosses, lichens, ferns, and epiphytes. Elfin woodland is re-
garded as an open stunted fasciation of mist forest, brought about by more
exposed conditions. Investigations in this community were conducted on
Abraham's Peak, St. Thomas Parish.



1977 CRUZ: JAMAICAN WOODPECKER 159

FLORIDA STUDY AREAS

Observations on woodpeckers in Florida were made in various natural and
m-m-modified communities in Alachua County. A description of the major
Florida study areas follows.

MORNINGSIDE PARK.-This city park in eastern Gainesville is composed of

two main vegetational types, longleaf pine-turkey oak sandhills and longleaf
pine flatwoods. In the former the predominant trees are longleaf pine (Pinus
palustris), turkey oak (Quercus laeuis), and bluejack oak (Quercus cinerea).
The undergrowth is sparse, but in places patches of saw-palmetto (Serenoa
repens) occur. In the longleaf pine flatwoods the predominant tree is the long-
leaf pine, although slash pine (Pinus elliotii) was also present. There is a dense
understory in which the most conspicuous plants are saw-palmetto, gall-
berry (Iicx glabra), and fetterbush (Demothamus lucidus). Canopy height in
both areas is less than 15 m.

DICKINSON STUDY AREA.-A xerophytic hammock in southwestern

Gainesville, characterized by the presence of large live oak trees (Quercus
virginiana) exceeding 10 m in height. Also present were laurel oak (Quercus
laurifolia), sweetgum (Liquidambar st!/raci~ua). pignut hickory (Car!/a
glabra), and many other trees.

SAN FELASCO STUDY AREA,-This extensive tract of climax mesophytic

hammock 11 km northeast of Gainesville has, unfortunately, been cutover in
some places, but it is still dominated by large trees and is the most extensive
stand of mesophytic forest in Alachua County. Characteristic trees are
southern magnolia (Magnolia grandiftora), laurel oak, pignut hickory, sweet-
gum, and many others.

MEI)ICINAL GARDEN STUDY AREA.-The medicinal garden on the Uni-

versity of Florida campus, Gainesville, was originally a mesophytic hammock,
but the removal of some of the trees and the undergrowth gives it a park-like
appearance. Characteristic trees include loblolly pine (Pinus Meda), pignut
hickory, blue beech (Carpinus caroliniana), water oak (Quercus nigm), south-
ern magnolia, sweet gum, and others.

LOBLOLLY PINE-PASTURE STUDY AREA.-This tract west of Gainesville

near interstate highway 75 consisted of pasture land with loblolly pines,
ranging in height from seedlings to 15 m.

RESULTS AND DISCUSSION

DISTRIBUTION

The Jamaican Woodpecker occurs throughout the island wherever suit-
able habitat is present, from sea level to high altitudes (Fig. 2). It occupies
forests, mangrove woodlands, and various man-modified communities, such as
wooded pastures, park-like areas, and tree crop areas. Destruction of the
Jamaican forests, which began when Columbus discovered the island in 1497,



160 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

0 5 4 852
Km

.<ca•®~E} .~**:.f'.'.'.~0~.9~7 
4

- iffsii!11!~*=1-.
...... *M.*AF.2

...... p.,8.*Ir --<'200'503%15*R!~
FIGURE 2.-Distribution of the Jamaican Woodpecker (Melanerpes mdiolams). Black circles
= distribution of Jamaican Woodpecker based on field observations, museum specimens,
published literature, and personal communications. Open area = 0-152 m, dotted = 153410 m,
horizontal = 610-1071 m, vertical above 1071 in.

has doubtless reduced the total range occupied by both the Jamaican Wood-
pecker and other predominantly forest-dwelling birds. Asprey and Robbins
(1953) noted that the land area consists of 18% forest, 47% agricultural use,
and 35% second growth scrub, thorn scrub, and mangrove woodland. Of the
agriculture area, 8% can be classified as tree crops (cocoa, pimento, citrus,
coconut, and bananas).

HABITAT ANALYSIS

Analysis of the structural components of the various communities visited
(Table 1) shows that the presence of the Jamaican Woodpecker in a given
habitat is directly related to the presence of trees necessary for feeding,
nesting, and roosting. Communities visited that were composed primarily of
shrubbery or scrub thickets, such  as the dry limestone scrub and the el_fin
woodland, had no woodpeckers. Tree species diversity did not seem to be a
factor in determining the presence or absence of woodpeckers, but it did seem
to be correlated with the numbers of woodpeckers present. This aspect will
be discussed under the census results. Woodpeckers were present in habitats
ranging from low tree species diversity and foliage height, such as mangrove
woodland and citrus groves, to areas of high tree species diversity and foliage
height, such as wet limestone forest and lower montane rain forest. The
presence of woodpeckers was not correlated with the presence of ground or
shrub layers. Places visited where woodpeckers were present ranged from
habitats without or with very little ground cover, such-as mangrove wood-
land and dry limestone forest, to places with heavy ground cover, such as
savanna (park-like) areas. Woodpeckers were also found in habitats ranging



1977 
C

R
U

Z: JA
M

A
IC

A
N
 W

O
O

D
P

E
C

K
E

R
 

161

TABLE 1.-HABITAT ANALYSIS OF COMMUNITIES VISITED.

Ground' Shrub Tree Tree' Tree species Dead trees·' Fruiting
layer layer layer size diversity or branches Epiphytes trees

Strand woods 2 2 3 5 1 U R U
Mangrove 1 1 3 S 1 U R R
Thorn scrub 1 3 1 - - R
Lowland savanna 3 2 2 M,L 1 C R U
Dry limestone scrub 1 3 1 - - R R
Dry limestone forest 1 2 3 S,M 2 U U U
Wet limestone forest 1 1 3 M,L 3 C C C
Modified wet limestone 3 2 3 M,L 3 C C C

forest
Upland tree crops 2 2 3 S,M 1 U U C
Montane rain forest 1 1 3 M,L 3 C C C
Montane mist forest 1 2 3 S,M 3 U C C
Elfin woodland 2 3 1 S 1 U C U

'Strongl developed (3), moderately developed (2), poorly developed or absent (1)
'Small (f),.4.6 meters, medium (M) - 4.6 to 9.2 m, large (L) » 9.2 meters
·'Rare or absent (R), uncommon (U), common (CD



162 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

from zero or low shrub density (e.g. park-like areas and wet limestone forest)
to those of high shrub density (mist forest and dry limestone forest). All these
habitats contain gradients in types of food available, nest sites, microclimates,
tree species diversity, and stratification, which help determine the size of the
population present.

ABUNDANCE

POPULATION DENSITIES IN DIFFERENT HABITATS.-Table 2 lists the den-

sities of Jamaican Woodpeckers in various natural and man-modified habitats
in number of birds per km. The highest densities were recorded at Grass
Piece, wooded upland pasture bordering on wet limestone forest (avg 22/km),
and at Coco-Ree, wet limestone forest (avg 20/km). No woodpeckers were
recorded on each of three Visits made to the Port Henderson area (dry lime-
stone scrub), Hellshire area (scrubby thickets), and Stoddard's Peak (elfin
woodland). The number of individuals was lowest in the mangrove woodland
(4/km) and dry limestone forest (6.8/km). The results of the other censuses
varied from 9/km in the montane mist forest to 15/km in the induced
guango savanna.

A species is probably found in greatest numbers in those habitats that
best meet its ecological requirements. The density figures indicate that
middle elevation wooded pastures and wet limestone forest are optimal habi-
tats for the jamaican Woodpecker. The absence of woodpecker occupation
in the dry limestone scrub, scrubby thickets, and elfin woodlands indicate that
these habitats do not meet the necessary minimal requirements, namely
a well developed arboreal component. The low numbers in the mangrove
woodland and dry limestone forest indicate that these areas are suboptimal
for Jamaican Woodpeckers.

The high density figures for both the wooded pasture (which has the
original wet limestone vegetation plus trees characteristic of more open areas)
and wet limestone forest are possibly related to their structural complexity.
Both areas have a well developed vertical stratification, high tree species
diversity (including many fruiting trees), and numerous bromeliads, all of
which increase the total area available for foraging. The results of the food
and foraging behavior analysis (pp. 165-184) indicate that fruits form a signifi-
cant portion of the foods eaten and that bromeliads and other epiphytes
are important foraging sites. The high densities of Jamaican Woodpeckers in
these places may also be related to the presence of large trees that furnish
a variety of suitable sites for roosting, nesting, and foraging. In contrast the
simpler forest structure of the mangrove woodland and the dry limestone
forest (i.e. lower canopy, low tree species diversity-including fruiting trees,
and little epiphytic development) apparently offers fewer sites for feeding,
nesting, and roosting purposes. Although both areas have a simple forest
structure, the less complex structure of the mangrove woodland probably



TABLE 2.-CHARACTERISTICS OF THE CENSUS AREAS AND THE RESULTS OF THE COUNTS.'

Distance Census
1977 

C
R

U
Z: JA

M
A

IC
A

N
 W

O
O

D
P

E
C

K
E

R
 

163
Censused Results Number of Mean Number

Site Community Type (Kilometer) Mean (Range) Censuses per Kilometer
Grass Piece Wooded pasture bordered

by wet limestone forest 0.6 14(12-15) 6 22
Coco-ree Wet limestone forest 0.75 15(14-16) 4 20
Caymanas Induced Guango savanna 0.3 4.5(46) 3 15
Corn Puss Gap Lower montane rain forest 2.6 29(25-33) 3 14
Hardwar Gap Montane mist forest 0.7 6.3(5-8) 3 9
Portland Ridge Dry limestone forest 2.2 15(12-18) 3 6.8
New Falmouth Mangrove woodland 0.5 2(24) 3 4
Stoddard's Peak Elfin woodland 0.6 0 2 0
area
Port Henderson Dry limestone scrub 2.5 0 3 0
Hellshire Scrubby thickets 1.75 0 3 0

'All censuses were taken during the spring and summer (0800-1200)

TABLE 3.-COMPARISON OF WOODPECKER DENSITIES IN JAMAICA AND NORTH AMERICA (IN NUMBERS AND GRAMS PER 40 HA),

Biomass
Species Habitat N/40 ha (g/40 ha) Sourel-

Jamaican Woodpecker modified mesophytic forest 36.4 3,942 This study
Florida woodpeckers' modified mesophytie forest 31.2 2,962 Dennis (1951)

(five species)
Florida woodpeckers' modified mesophytic forest 34.0 ' 2,330 Woolfenden and Rohwer

(two species) (1959)
Jamaican Woodpecker mesophytic forest 30.3 3,281 This study
Florida woodpeckers' mesophytic forest 31.5 2,598 Dennis (1951) Kale and

(four to six species) (average for four areas) (27 to 42) (2,359-3,075) Webber (1968), and
this study

'Cotaptes auratus, D,!ocopus pileatus, Melanerpes carolinus, M. erythrocephalus, and Dendrocopos pubescens.
1Cofaptes auratus and Melane,pes carolinus
'Colaptes auratus, Dryocopus pileatus, Melane,pes carolinus, M. erythrocophalus, Dendrocopos pubescens, and D. cillostis



164 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

accounts for the lower number of woodpeckers recorded there. In addition,
the Jamaican Woodpecker uses mainly dead trees or dead branches of live
trees for excavating nesting and roosting holes. An examination of over 100
woodpecker holes in various communities showed that they are usually
placed higher than 4.6 m and excavated in trunks or branches at least 15 cm
in diameter. Although trees meeting these requirements are present in both
the dry limestone forest and the mangrove woodland, they are not common.
The other communities censused (Table 2) showed a gradient of these factors.

COMPARISON OF WOODPECKER DENSITIES IN JAMAICA AND FLORIDA.-

When a species occurs together with many similar species that exploit some
of the same resources, one might anticipate that population numbers and
niche breadth might be reduced, in contrast to the situation that would
exist in the absence of the competitors (see Crowell 1962). Selander (1966)
estimated that the Hispaniolan Woodpecker's population density was four
to five times greater than that maintained in continental areas by any species
of woodpecker with which he had field experience. Unfortunately no census
counts were made in that study. The Hispaniolan Woodpecker shares its
habitat with the Antillean Piculet (Nesoctites micromegas), a woodpeeker that
is very different in size, morphology, and habits.

The Jamaican Woodpecker is in a somewhat similar ecological situation.
It is the only resident woodpecker on Jamaica, so that one might expect
its population size to be greater than those of some of the mainland wood-
peckers.

To test this assumption and to see whether the increase in numbers (if
any) is sufficient to compensate for the absence of other woodpeckers, i.e.
whether the total population density of Jamaican Woodpeckers is as high as
the combined population densities of mainland woodpeckers, census counts
were undertaken in Jamaica and Florida. The F16rida counts were also sup-
plemented by counts appearing in the literature, but no published census
counts were available for the Jamaican Woodpecker.

These censuses (Table 3) indicate that in some habitats the Jamaican
Woodpecker maintains comparable or higher densities than the combined
population densities of several species of Florida woodpeckers. Thus Jamaica
has 36.4 and 30.3 woodpeckers per 40 ha in modified mesophytic and meso-
phytic habitats respectively, and Florida has 32.5 (average for two plots) and
31.5 (average for four plots) per 40 ha in modified mesophytic and meso-
phytic habitats. Although number of individuals per unit area is the customary
census unit, Salt (1957) and Crowell (1962) pointed out the value of biomass
figures as an index of abundance of different species. Crowell (1962) noted:
"when considering the amount of living matter supported by a unit area, it
is weight which should be used." Standing crop biomass for woodpeckers in
grams per 40 ha for Jamaica was calculated to be 3,942 for the modified
mesophytic habitat, 3,281 for the mesophytic habitat, and for similar areas



1977 CRUZ: JAMAICAN WOODPECKER 165

in Florida it was 2,962 and 2,598. These data show that comparable (similar
structure) Jamaican communities support a greater biomass of woodpeckers
per unit area than do mainland habitats.

Some of the possible explanations for the high densities and biomass
figures for Jamaica are the following. (1) More resources are available to
Jamaican woodpeckers through the absence of other bird species (and animal
groups) with similar habits. (2) The average weight of the Jamaican Wood-
picker is 25% greater than the combined average weight (81 grams) 6f Florida
woodpeckers (range 24.3 to 237.5 grams). In addition the larger species of
Florida woodpeckers, such as the Pileated Woodpecker (Dr~ocopus pileatus),
were not as common as the smaller species. The most common species was
the Red-bellied Woodpecker (Melanerpes carolinus) with an average weight
of 66.4 grams. (3) Apparently more energy is available to the avian segment
of the community in Jamaica than in the North American mainland, because
productivity is probably greater in Jamaican habitats than in comparable
Florida habitats. Although no data are available for Jamaica, Odum and
Pigeon (1970) found that the net productivity in a mature mesophytic forest
in Puerto Rico was 13,000 kcal/m' /year, and Woodwell and Whittaker (1968)
found that the net productivity of an oak-pine forest in North America was
5,000 kcal/nV/year. (4) Jamaican habitats are structurally more complex than
comparable Florida habitats, evidently because more fruiting trees, epiphytes,
lianas, and foraging niches are available in Jamaica (Asprey and Robbins
1953, Adams 1972). (5) Jamaica has fewer predators on birds than the main-
land, where snake predation is one of the most common causes of woodpecker
mortality (Kilham 1957, Nolan 1959, Stickel 1962, Jackson 1970, Dennis
1971). A hole-nesting bird such as a woodpecker, although having a safer
nest in many respects and an easier nest to defend than birds nesting in the
open, is still vulnerable to snake predation especially if surprised. Jamaica
has five species of snakes, only one of which possibly preys on woodpeckers.
This snake, the Jamaican boa (Epicrates subflauus), is very rare and con-
sequently the numbers of woodpeckers it eats, if any, are probably negligible.

FEEDING ECOLOGY AND BEHAVIOR

This section describes feeding behavior and food habits in both a qualita-
tive and quantitative manner, stressing the following aspects: sexual di-
morphism, foraging heights, foraging zones (upper trunk, outer branch, etc.),
foraging behavior (pecking, probing, etc.), food habits (based' on stomach
analyses and field observations), interspecific food competition, and structural
adaptations relevant to foraging method. There follows a discussion of the
morphological data obtained by standard methods to assess any sexual di-
morphism in body structure with ecological significance.

LACK OF SEXUAL DIMORPHISM IN THE JAMAICAN WOODPECKER.-NO

statistically significant sexual differences were found between the sexes in



166 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

bill length, weight, and tarsometatarsal length at the 0.001 probability level
(t-test). There was a difference in outer-front toe length at the 0.001 proba-
bility level. The data presented in Table 4 indicate that the Jamaican Wood-
pecker is monomorphic, with the males averaging slightly larger in culmen
length, tarsometatarsal length, and outer-front toe length. The greatest dif-
ferences between the sexes were in body weights, the male woodpeckers
averaging 12 grams heavier than the female woodpeckers (114.8 to 102.0
respectively), and outer-front toe length, the male woodpecker averaging 1.55
mm longer than the female woodpeckers (24.3 to 22.75 respectively).

TABLE 4.-MEASUREMENTS OF Melanerpes radiolatus.'

Males Females
N Mean *S.D. Range N Meant S.D. Range

Bill length' 11 28.04* .61 26.8- 29.1 14 27.01* 1.05 25.8- 29.2
Weight 11 114.72*8.6 97.3-130.5 14 102.02 *8.9 91.6-118.5
Tarsometatarsal 11 25 34 f .868 23.7- 26.7 13 24.80 f .65 23.6- 25.3
Foot volume 8 .8 .75- .85 9 .8 .75- .85
Outer front toe(3) 11 24.3 f .85 23.0- 25.5 12 22.75:E .97 20.7- 23.9

'No Ntatistically significant differences were found between the Nexes either in lilli length, weight. or tar%.inetatarNal length
(.001 prol,al,lity level. t-test).

Measure,nents in mm. weights in-g. and foot volnme in mi.

These data contrast sharply with measurements on another insular species
of woodpecker, Melanerpes striatus of Hispaniola (Selander 1966). In all
linear dimensions and in weight, the male Hispaniolan woodpecker is larger
than the female, usually with little or no overlap between the sexes. The
greatest sexual differences were in bill length in which there was a 21.3%
difference. Males use the longer bill to do more pecking and probing, whereas
the shorter-billed females do more surface gleaning. This contrasts sharply
with the Jamaican Woodpecker, where the bill of the female averages only
3.67% shorter than that of the male. In body weights females of Mehnerpes
striatus average 18.1% less than the male, but females of M. radiolatus average
only 11.1% less than the male.

Selander's study demonstrated a relationship between the degree of sexual
differences in foraging behavior and the degree of sexual dimorphism in the
feeding apparatus. Davis (1965) and others suggested that "bill size would
probably be the most important character involved" in determining the range
of food items taken and the foraging techniques employed. One may reason-
ably ask whether males and females of structurally monomorphic species
may exhibit differences in foraging and food items chosen, as such differences
could be of a strictly behavioral nature (such as foraging heights or zones),
rather than one directly correlated with differential morpholbgical characters
(Ligon 1968, Morse 1968, Williamson 1971).



1977 CRUZ: JAMAICAN WOODPECKER 167

FORAGiNG HEIGHTS.-Almost complete overlap in foraging heights for
male and female woodpeckers is seen in the pooled data for all the study
areas as well as in the data for individual study areas (Fig. 3). On the other

100 - - r- Mlmis =sm RAS =44#6

75-

50- r-

-

25-

...... ... -.... - -*. L

469 d d9 9 d 8 9 d 69 9 d 0 9

12345
FIGURE 3.-Percent distribution of foraging heights in the Jamaican Woodpecker. (1) Pooled re-
suits for all study areas; (2) Modified wet limestone forest; (3) Dry liinestone forest; (4) Montane
mist forest; (5) Mangrove woodland. Stippled area = 0-4.6 m, vertical area = 4.6-9.2 m and white
area = 9.2 m and over. Numbers at top of bar = number of observations.

hand, the foraging heights were correlated with the heights of the trees in
the various foraging zones. In the Worthy Park study area, for example, the
trees were usually taller than the trees in the other study areas, and this
greater height is reflected in the foraging heights of the woodpeckers, 26% in
the upper zone (greater than 9.2 m), 45% in the middle zone (4.6 to 9.2 m),
29% in the lower zone (less than 4.6 m). In the mangrove woodland the
trees averaged considerably smaller, and consequently most of the foraging
observations were recorded at lower heights: 68% in the lower zone, 30% in
the middle zone, and 2% in the upper zone. In the other study areas
tree heights on the average fell between the Worthy Park study areas and the
mangrove woodland study area, averaging taller in the Hardwar Gap study
area than in the Portland Ridge study area. In the Hardwar Gap area, 35%



168 BULLETIN FLORIDA STATE MUSEUM Vol, 22, No. 4

of the foraging observations were recorded in the lower zone, 49% in the
middle zone, and 16% in the upper zone; in the Portland Ridge study area,
46% were recorded in the lower zone, 46% in the middle zone, and 8% in
the upper zone. The pooled results for all the study areas indicate that the
lower zone was used 35%, middle zone 45%, and the upper zone 21%. These
results indicate (1) an absence of stratal segregation between the sexes, and
(2) that both sexes forage most commonly in the middle zone.

FORAGING ZONES.-Table 5 and Figure 4 summarize the percent of the
total number of times the Jamaican Woodpecker was noted in the different
foraging zones-trunk, inner, and outer portions. The Jamaican Woodpecker
used different zones preferentially and both the pooled results for all the
study areas and the results for the individual study areas show a nearly
complete overlap in the foraging zones used by male and female woodpeckers.

The pooled data for both sexes for the different study areas indicate that
almost one-half of the time (46%) was spent foraging in the inner portion,
one-third (32%) in the outer portion, one-seventh (14%) on the trunk, and the
remainder (8%) on the vines and in the aerial zones. These values varied from
area to area, ranging from 38% for the inner zone in the Portland Ridge
area to 57% in the mangrove woodland, 21% for the outer zone in the man-
grove woodland area to 34% in the Worthy Park area, and from 10% for
the trunk in the Hardwar Gap area to 22% in the mangrove woodland.

Within each major foraging zone (inner branches, outer branches, and
trunks), different subzones were preferentially used (Fig. 4). The pooled data
for the different study areas (Table 5) show that 20% of the foraging observa-
tions were in the middle inner branches, 13% in the middle outer branches,

59 4 9

8 -3

)< 2-11

2-20 . - -20t'. f, +r- 20
((L\\ ~/  -12|-3 ~-4

, 1-13-4 ~. 4 14 12 --0 1-L 15ii- k Ls 10 -13~~194 < -t15

Vine- 6 ~ Vine - 5 ~ Vine - 7 -31- 2Aerial- 2 ~-3 Aerial- 1 ~_ Aerial - 3

FIGURE 4.-Foraging zones of the Jamaican Woodpecker. The numbers represent the percent
of the total number of observations. The subzones of greatest usage. within each major zone are
shown by stipling.



TABLE 5.-USE (IN PERCENT) OF VARIOUS FORAGING ZONES lN DIFFERENT HABITATS BY THE JAMAICAN WOODPECKER. 1977 
C

R
U

Z: JA
M

A
IC

A
N
 W

O
O

D
P

E
C

K
E

R
 

169
Trunk Inner OuterNumber of

Sti,dy Areas 01}servations Sex LT' MT UT LIB MIB UIB LOB MOB UOB Vines Aerial
Wooded pasture 725 3 2 10 3 :13 20 10 12 15 9 2 2

(Worthy Park)
Dry limestone forest 52 8 4 8 4 16 14 10 12 12 6 4

(Portland Ridge)
Montane mist forest 259 3 2 5 2 17 19 12 12 12 7 12 -

(Hardwar Gap)
Mangrove woodland 32 8 6 10 - 28 34 - 16 6 -

(New Falmouth)
Pooled 1068 6 2 9 3 14 20 11 13 14 8 51
Wooded pastitre 552 9 3 7 3 13 18 14 10 13 8 6 5

(Worthy Park)
Dry limestone forest 68 9 3 13 6 10 19 7 6 15 6 15 -

(Portland Ridge)
Montane mist forest 141 9 2 5 4 18 24 13 11 9 4 10

(Hardwar Gap)
Mangrove woodland 76 9 7 10 8 28 22 4 13 8

(New Falmouth)
Pooled 537 9 3 8 4 15 20 12 10 12 6 73
Wooded pasture 1277 3 9 3 9 3 13 19 12 11 14 9 43

(Worthy Park)
Dry limestone forest 117 8 9 3 11 5 13 17 8 8 14 6 15

(Portland Ridge)
Montane mist forest 400 3 9 2 5 3 17 21 12 12 11 6 11 -

(Hardwar Gap)
Mangrove woodland 108 0 9 6 10 6 28 26 3 14 7

(New Falmouth)
Pooled 1905 & 9 3 8 3 15 20 11 11 13 8 6 2

' Letter symt)01% salne as in Figure 3.



170 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

and 8% in the middle trunk. The use of these subzones yaried from area to
area, ranging from 17% for the middle inner branches in Portland Ridge to
26% in the mangrove woodland, 7% for the middle outer branches in the man-
grove woodland to 14% in both the Worthy park and Portland Ridge study
areas, 5% for the middle trunk in Hardwar Gap to 11% in Portland Ridge
(Table 5). Although some differences between the 5exes exist with regard to
the subzones used, a wide range of overlap is evident. The greatest differences
between the sexes were in the use of the lower inner branches in the Portland
Ridge region, 16% for the male and 10% for the female, and the middle inner
branches in the mangrove woodland, 34% for the male and 22% for the female,
a 6% difference in the former, and a 12% difference for the latter. In most of
the subzones the range of overlap was from complete overlap to 3% dif-
ferences. What is apparent is a correlation between the foraging zones used
and the foraging behavior.

FORAGING METHODS.-Foraging by the Jamaican Woodpecker in the
principal study areas is shown in Table 6. The pooled results for both sexes
from all the study areas indicate that the predominant foraging methods are
fruit-eating (28%), probing (28%), and pecking (20%). Probing into bromeliads
(13%), gleaning (9%), and sallying (2%) accounted for the rest of the foraging
procedures.

A comparison of the foraging patterns used by male and female Jamaican
Woodpeckers (Table 6) shows that the area of overlap between the sexes was
large and that neither sex used one method solely. The greatest difference
noted between the sexes was in pecking-the combined results of the dif-
ferent study areas indicate that males do 5% more pecking than females, but
this is not the complete picture. In some individual study areas (e.g. dry
limestone forest) females peck more frequently than males (4% differences).

The results from the individual study areas indicate that certain feeding
repertoires predominate in some places, whereas in others that same feeding
repertoire plays a lesser role or is absent (Table 6). The high percentage of
fruit in the diet of the jamaican Woodpecker in some of the study tracts,
for example, appears to be correlated with the diversity of fruiting trees that
provide a variety and abundance of fruit at all seasons. Fruit-eating was
highest in Worthy Park (33%) where an abundance of different fruiting trees
occurs, and lowest in the mangrove woodland (0%) which was a low tree
species diversity and few fruit-bearing trees.

Foraging pattern type was correlated with foraging zone used. For ex-
ample, in Worthy Park where the incidence of fruit-eating was high, the use
of the outer branches for foraging was also high (34%). This is obviously re-
lated to the fact that most of the fruit-bearing branches of the tree occurred
in the outer zone. In the mangrove woodland, which has a low fruiting tree
diversity, the predominant mode of foraging was pecking (58%), and this is
reflected by the foraging zones used, the trunk and the inner branches (21%



TABLE 6.-USE (IN PEHC:EXT) OF VARIOUS FORAGING METHODS lN' DIFFERENT HABITATS BY THE J~MAiCAN WOODPECKER.

1977 
C

R
U

Z: JA
M

A
IC

A
N
 

W
O

O
D

P
E

C
K

E
R
 

171
Study Areas N' Sex Pecking Probing Bromeliads Cleaning Fruit-eating Sallying
Wooded pasture 731 6 22 22 14 9 31 2

(Worthy Park)
Dry time%tone forest 50 3 30 28 - 10 32 -

(Portland Ridge)
Montane mist forest 259 6 16 47 13 7 17

(Hardwar Gap)
Mangrove woodland 32 3 66 25 - 9 -

(New Falmouth)
Pooled 1072 3 22 29 12 9 27 1
Wooded pasture 552 9 10 21 17 10 37 5

(Worthy Park)
Dry limestone forest 68 9 34 29 - 9 28

(Portland Ridge)
Montane mist forest 140 9 14 36 16 8 26

(Hardwar Gap)
Mangrove woodland 76 9 55 37 - 9

(New Falmouth)
Pooled 836 9 17 26 14 9 31 3
Wooded pasture 1283 8 9 16 22 15 10 33 4

(Worthy Park)
Dry limestone forest 118 8 9 32 37 - 10 21

(Portland Ridge)
Montane mist forest 399 8 9 15 44 14 7 14

(Hardwar Gap)
Mangrove woodland 108 8 9 58 32 - 9

(New Falmouth)
Pooled 1908 3 9 20 28 13 9 28 2

' Represe* the nult,1,et  of <,IN.vatio,2. -



172 BULLETIN FL0RIDA STATE MUSEUM Vol. 22, No. 4

and 57% respectively). In all the study tracts pecking was confined mainly to
the trunk and the inner branches. The other study areas also showed a cor-
relation between the foraging behavior and the foraging zones used.

PROBING.-Probing and fruit-eating were the most common foraging pro-
cedures used by the Jamaican Woodpecker, each accounting for 28% of the
total. These values exclude probing into bromeliads, which is included in a
separate section and accounts for 13% of the total foraging observations.
Probing was used in all the study areas, ranging from 22% in the wet lime-
stone forest (Worthy Park) to 44% in the montane mist forest (Hardwar Gap).
A low frequency of probing for Worthy Park is possibly correlated with the
availability of alternate food sources, such as fruiting trees and bromeliads. On
the other hand, frequency of probing in the montane mist forest is possibly
related to the presence of a large number of weathered holes and rotten
branches (the result of high atmospheric humidity and rainfall) that provide
many suitable sites for probing.

In probing both the bill and the tongue are used to explore a variety of
natural cavities and accumulations of plant material. Cavities in which the
woodpeckers probed included fissures and cracks in the bark, knot holes,
weathered holes, holes previously excavated by woodpeckers, holes in ends
of rotten or dying branches, and stumps. Accumulation of plant material in-
cluded debris between trunks and lateral branches, spaces in and under-
neath mosses, lichens, and epiphytic growth on trees.

The long and protrusible tongue of the Jamaican Woodpecker, which
extends almost 5 cm beyond the tip of the bill, is highly adaptive for probing.
Kilham's studies (1963) on the use of the tongue by various woodpeckers
showed that the tongue of the congeneric Melanerpes carolinus, while not as
long as the tongues of Colaptes or Dryocopus, appears to function more
adroitly in maneuvering objects at a distance, within crevices. Neither
Sph!/rapicus varius nor Melanerpes ergthrocephalus has the long-distance
control and adroitness of M. carolinus.

Kilham (1963) stated: "C. carolinus spends almost no time excavating holes
when in search of grubs and other prey, but appears to accomplish the same
objective by using bill and tongue to explore natural crevices. When I brought
rotting logs from the woods to the aviary, the Pileated, Downy, and Hairy
Woodpeckers, as well as the Yellow-shafted Flicker, and to a lesser extent, the
sapsuckers, were immediately interested and each began pecking or hewing,
in its individual style, to get ants, termites, or whatever might turn up. The
strategy of C. carolinus was effective in this competitive situation... this spe-
cies explored the logs with its tongue, occasionally twisting its head upside
down to do so. In this manner it sometimes pulled out beetle larvae 2.5 to 5
centimeters long before any of the other woodpeckers had dislodged prey of
any size. It may be that the tongue-probing of Centurus, in contrast to pecking
and hewing methods, enables it to come upon its victims quickly and without
disturbance."



1977 CRUZ: JAMAICAN WOODPECKER 173

My field observations on both the Jamaican Woodpecker and the Red-
bellied Woodpecker in Florida bear out Kilham's Observations, but the
Ja~aican Woodpecker does more pecking than the Red-bellied Woodpecker,
in which pecking as a feeding method accounted for only 4% of its total
foraging procedures. This low figure is possibly related to the presence of

"

other woodpeckers in Florida that are more effective "peckers, whereas the
higher pecking figure for the Jamaiean Woodpecker is probably due to the
absence in Jamaica of other woodpeckers that might compete with and be
more efficient in using the pecking niche.

EpipHYTES.-The use of this foraging niche was correlated with the quan-

tities of epiphytes (particularly bromeliads) present. The mangrove woodland
and the dry limestone forest support few bromeliads. In contrast the wet
limestone forest and the montane mist forest have luxuriant growths of bro-
meliads and Other epiphytes, and probing into bromeliads constituted 15% and
14.3%, respectively, of the total foraging methods for each of these habitats.

Foraging involved probing the inner and the outer leaves of the bro-
meliads, using the bill to push or throw aside debris or portions of the bro-
meliads until a suitable prey item was found. Foods seen taken from the
bromeliads included snails (Gastropo(la), spiders (Araneae), insects (Orthop-
tera, Coleoptera, larval lepidopterans, and other unidentifiable insects), frogs
(H!/la spp.?), and an occasional fruit that had fallen into the bromeliad. The
fauna of five bromeliads (two large, base diameter 20 cm; two medium,
base diameter 13 cm, and one small, base diameter 5 cm) representative of
the types used by the jamaican Woodpecker consisted of snails (Gastropoda),
earthworms (Oligochaeta), spiders (Araneae), insects (Tettigoniidae, Acridi-
dae, Blattidae, Formicidae, unidentified hemipterans, larval and adult coleop-
terans, and lepidopterans), millipedes (Diplopoda), sowbugs (Isopoda), and
frogs (Hilda).

PECKING.-The Jamaican Wobdpecker has the family's structural speciali-
zation of obtaining food by pecking, but pecking does not play so prominent
a role in its foraging as it does in some of the continental species (Table 11),
other than its mainland congeners, M. carolinus and the M. erythrocephalus.
Observations in Florida indicated that pecking accounted for 4% and 2%
of the total foraging behavior of the Red-bellied and Red-headed wood-
peckers, respectively. In contrast, pecking in the Jamaican Woodpecker
accounted for 20% of the pooled results for the different study tracts, ranging
from 15% in the montane mist forest to 58% in the mangrove woodland (Table
6). This difference between the insular species and its mainland congeners
is probably related to the absence in Jamaica of other species of woodpeckers
that would compete for the pecking niche. In Florida, for example, pecking
plays a prominent role in the Pileated, Hairy, and Downy Woodpeckers,
accounting for 62%, 91%, and 89% of the total foraging behavior (Table 11).

Pecking constituted a greater proportion of the foraging behavior in those



174 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

habitats where fruiting and epiphyte-laden trees were scarce, such as the
mangrove woodland and dry limestone forest. In the wet-limestone and mon-
tane mist forests pecking played a minimal role.

Pecking for food was confined mainly to dead or dying portions of the
tree, ranging from large tree trunks to the outer branches, and to the hollow
branches of Cecropia. The latter, occupied by ants (Formicidae), was a favor-
ite pecking site. This type of predation on ants living in Cecropia inter-
nodes is not peculiar to the Jamaican Woodpecker for it has also been
reported by Skutch (1945) and Janzen (1969) for Central American wood-
peckers. In addition this type of predation is not peculiar to ants living in
Cecropia, for I have seen the Jamaican Woodpecker feed on the twig- and
branch-inhabiting ants in other tree species, although it was not as common.
The pecking procedure did not involve Splitting the twigs, but punching
a series of holes through the twig wall at approximately 2-8 em intervals.
The spacing and the size of the hole the woodpecker made varied according
to the size and type of twig involved. Food items obtained by pecking, other
than ants, included various woodboring larvae (e.g. Buprestidae).

In pecking the Jamaican Woodpecker assumes the woodpecker stance de-
scribed by Spring (1965) for some North American woodpeckers. The whole
body is held away from the tree trunk and the blow delivery momentum
appears to come from regions posterior to the neck, although it appears
that the neck also plays a role. While on a vertical surface (trunk), either
pecking or climbing up, the toes are usually held in the position shown by
Bock and Miller (1959) for the Hairy and Pileated Woodpeckers. While
the fourth toe is directed laterally at right angles to the trunk, the second
and third toes are held anteriorly and the hallux points down the trunk, and
not in a zygodactyl position.

FRuIT-EATING.-Although the Jamaican Woodpecker is structurally
specialized to obtain food by pecking, fruit-eating constitutes a greater
proportion of the pooled foraging behavior than pecking, the former account-
ing for 28% and the latter for 20%. In addition, the results of the stomach
analysis indicate an even greater use of fruits (41%). The use of fruits as food
was directly correlated with the presence, abundance, and diversity of fruiting
trees. In the mangrove woodland few f~uiting trees were present and none
was bearing fruits during my visits, and therefore no fruit-eating was recorded.
The wet limestone forest had a diversity and abundance of fruiting trees
(Table 7), and fruit-eating accounted for 33% of the total. In both the mon-
tane mist forest and dry limestone forest fruiting trees were also present,
but not in the abundance or diversity of the wet limestone forest, which is
reflected in the lower numbers of fruit-eating observations reported there
(Table 7).

Melanerpes radiolatus showed considerable agility in moving over the
outer branches of trees where most of the fruits grew. Birds often balanced



1977 CRUZ: JAMAICAN WOODPECKER 175

TABLE 7.-RECORDS oF FRUIT-EATING By Me/cinerpes radiolatus.

Trees and Shrubs Limestone Forest Montane Mist

Family Species Wet Dry Forest

Ulmaceae Trema inicranthum' 5
Moraceae Trophis race,nosa 10

Cecropia peltata 138
Ficus trigonata 70
Ficus pertusa 21
Ficus citrifolia - 2

Lauraceae Ocotea sp 20
Nectandra antit[ana 15 -
Nectandra coriace<1 11 2

Rutaceae Citrus atirantilim 4-
Fagara tnartinicensis 8 -

Simaroubaceae Picrcisina excelsia 57 -
B, irseraceae Bursern si,11(imbci - 6

Euphorbiaceae Alchornea latifolia 26

Anacardiaceae Mangifem indica 8

Thymelaeceae .Wetopium browni 12

Thymelaeceae Daphnopsis mnencana 48

Flacourtiaceae Casaeria hirst, ta 10
Castieria niticlu - I
Lactia thmnnia 4

Araliaceae Dendropanax arborems 19

Sapotaceae Bwne/ki,saticifolia - 2
Pouteri~ mititiflom 2
Chrysophylit,in cainito 5

Symplocaceae Symplocos octopet{11(1 10

Boraginaceae Cordia collococca 4
Verbenaceae Petitia domingensis' 4
Solanaceae Dunalict arborescens 19

Total number of observations 429 25 80

' Botanical names taken from Adain~ (1972).

between two limbs, with legs spread apart, holding one branch with each foot,
and reaching out for the fruits. Sometimes individuals flew to a dump,
clung to the fruits, and plucked them. Occasionally I saw the birds hang
upside down with the body parallel to the ground to reach an otherwise
inaccessible fruit.

Small fruits (<25 mm) were plueked and swallowed and if the fruit was
very small (5 mm), it was not uncommon to see woodpeckers with more than
one fruit in the bill at a time (up to seven in one instlince). If the fruit was
large (> 25 mm) (e.g. ripe mango), the bird first pecked a hole in it, and then
used the bill and tongue to probe and feed on the fleshy pulp. In feeding
on Cecropia catkins (up to 6 cm), the birds broke off terminal pieces and
swallowed them whole.



176 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

I recorded 29 different types of fruit eaten by the Jamaican Woodpecker.
Fruit from trees of the family Moraceae accounted for one-half of the fruits
they ate in the wet limestone forest. In the dry limestone forest and mon-
tane mist forest, where fewer cases of fruit-eating were recorded, Alchomea,
Dunalia, and Dendropanax fruits accounted for more than three-fourths of
the fruits eaten in the latter, and Metopium accounted for nearly one-half of
those eaten in the former (Table 7).

GLEANING.-Cleaning accounted for 9% of the total foraging behavior re-
corded. This consisted of searching along the limb and trunk surfaces for
invertebrate or vertebrate prey. On trunks and larger branches, gleaning
Jamaican Woodpeckers move in a fashion characteristic of most woodpeckers,
using their stiff rectrices as a prop and the laterally directed toes acting in
juxtaposition as a pincer. This stance was used while gleaning on horizontal
and vertical surfaces and even when moving along the underside of large
branches. On vertical branches, gleaning birds usually moved outward from
the trunk and then usually flew back to where another branch joined the
trunk and repeated the process. The identifiable animals taken by gleaning
included tree snails (Gastropoda), insects (Orthoptera, Coleoptera, and larval
Lepidoptera), and Anolis lizards.

SALLYING.-Jamaican Woodpeckers rarely engaged in sallying flights
(Table 6), which usually took place in late afternoon from the tops of trees.
The flight pattern took the form of a circle or a loop and the bird returned
to the same perch or part of the tree.

STOMACH ANALYSES.-Tables 8 and 9 list all food items found in the
stomachs of Jamaican Woodpeckers, as well as frequency and volume per-
centages of each prey taxon in the diet. Figure 8 shows the size distribution
of intact animal prey.

In the Jamaican Woodpecker diet both animal and vegetable matter are
well represented, comprising 58.2% and 42.7%, respectively, of the total
volume. A striking general result of the present study is the demonstration
of the major role played by vegetable material (fruits) in the diet of a member
of a family considered primarily insectivorous.

The animal food embraced 3 classes, 7 orders, and 11 families. Insects
were most important in the woodpecker diet, comprising 48.6% by total
volume. The most important taxa were Orthoptera, Lepidoptera, and Coleop-
tera, accounting for 18.2%, 14.2%, and 11.5% of the total volume, respectively.

The low percentage of wood-boring larvae and limb or twig dwelling in-
sects, accounting for (at the most) 10% of the total volume, is in agreement
with the results of the foraging behavior where pecking accounted for 16%
of the total in Worthy Park. Included here are the wood-boring larvae of
Coleoptera (Buprestidae) and twig-dwelling ants (Formicidae). Most of the
invertebrate prey taxa (48%) live on wood surfaces, in crevices, in accumula-
tion of plant material, and in epiphytes. They included tree snails (Poma-



1977 CRUZ: JAMAICAN WOODPECKER I77

TABLE 8.-ANIMAL FOODS IN THE STOMACHS OF Melanerpes radiolattis.

Percent Occurrence Percent Volume
Prey Taxon 3(15) 39 9(14) S (15) 89 9 (14)

Mollusca
Gastropoda

Pomatiasidae - 6.9 14.2 - 2.0 4.7
Helicinidae - 3.4 7.1 - 0.6 1.5
Undetermined 6.6 6.9 7.1 2.8 2.4 1.9

Arthropo(la
Arachnida

Heteropidae 6.6 6.9 7.1 1.4 3.6 6.6
Undetermined 6.6 6.9 7.1 0.3 0.2 0.2

Insecta
Undetermined 13.3 6.9 - 2.5 1.5 -
Mallophaga

Menoponidae 6.6 3.4 - 0.3 0.1 -
Orthoptera

Locustidae 26.7 22.1 21.4 5.3 7.7 11.0
Tettigonidae 13.3 13.8 14.3 7.0 6.7 6.3
Gryllidae 6.6 6.9 7.1 2.3 3.7 5.7

Coleoptera
Buprestidae 13.3 6.3 - 6.3 3.6 -
Cerambycidae 6.6 3.4 - 2.5 1.5 -
Undetermined 26.6 31.0 35.7 2.6 3.6 4.9

Lepidoptera
Noctuidae 20.0 13.8 7.1 11.3 9.1 5.9
Undetermined 13.3 10.3 7.1 6.0 5.2 4.0

Hymenoptera
Forinicidae 33.3 34.4 35.7 4.9 5.1 5.3
Undetermined - 7.1 13.8 - 0.2 0.4

Subtotal: Animal - - - 55.5 58.2 58.4
Subtotal: Plant - 42.7 41.044.2

tiasidae and Helicinidae), spiders (Heteropidae), grasshoppers (Locustidae
and Tettigoniidae), crickets (Gryllidae), beetles (Cerambycidae and others
of undetermined families), most of the lepidopteran larvae (Noctuidae), and
possibly some of the ants. These prey items are obtained primarily by non-
pecking means, such as probing and gleaning. The Noctuidae larvae, for
example, belong to a group that is active during the night, but during the
day hides underneath the bark, in crevices, and plant accumulation (Dale H.
Habeck, pers. comm.). The proportion of the invertebrate prey obtained by
nonpecking means (48%) is in close agreement with the results obtained in
the foraging methods section, where foraging patterns other than pecking
and fruit-eating accounted for 51% of the total volume.

Plant materials consisted of fruits and seeds, representing three identified
families and ten genera. The family Moraceae was the most important in
the diet, their fruits and seeds accounting for 25.4% of the total volume. The
most important taxa were Cecropia and Ficus spp., accounting for 17.5%



178 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

TABLE 9.-VEGETABLE FOODS IN THE STOMACHS OF Melanerpes radiolatus.

Percent Occurrence Percent Volume
Plant Taxon S (15) &9 9 (14) 315) 29 9(14)

Moraceae
Cecropia 51.7 44.8 42.9 23.5 17.5 9.3
Ficus 26.6 20.6 14 .3 7 .7 7 .3 6 .6
Trophis 6.6 6.8 7.1 0.5 0.6 0.8

Lauraceae
Bailschmie(tia - 3.4 7.1 - 2.4 5.7
Nectandra 6.6 6 .8 7 . 1 0 .8 1 .3 1 .9
Ocotea 6.6 6 .8 7 . 1 4 .9 5 .6 6.7

Simaraubaeceae
Picrasma - 3.4 7.1 1.1 0.6 -

Flacourtiaceae
Casaeria - 3 .4 7 . 1 - 1 .2 2 .8

Thymelaeceae
Daphnopsis 13.3 13.8 14.3 5.3 5.3 5.1

Araliaceae
Dendropanax 6.6 6.8 7.1 - 0.2 0.6

Undetermined 6.6 6.8 7.1 0.4 0.9 1.5
Subtotal: Plant - - - 44.2 42.7 41.0
Subtotal: Animal - - - 55.5 58.2 58.4

and 7.3% of the total volume. Both of these trees were visited by large numbers
of birds to feed on their fruits. In Worthy Park I counted 11 species of birds
feeding in a single Cecropia peltata tree and 19 species in a single Ficus tri-
gonata tree (Cruz 1974). Other plants important in the diet (in percent of
total volume) were fruits and seeds of the families Lauraceae 7.0%, Thymela-
ceae 5.3%, and Simaraubaceae 3.1%.

Figure 5 shows the distribution of prey in different size classes by
percent frequency. The prey ranged from small ants (4 min) to large cater-
pillars (45 mm). Most of the prey items taken by both sexes were in the 1-10
mm range, but as only intact prey items were measured and the prey cap-
turing technique fractured proportionately more prey items, the sampling is
biased against larger prey items (greater than 10 mm). Figure 5 shows a
large prey size overlap between the sexes, so they are not selecting different
prey sizes.

The large and diverse numbers of animal and plant species taken strongly
suggest that the Jamaican Woodpecker is exceedingly diverse and opportun-
istic in its feeding habits, taking nearly all the animal and fruit material
(within a certain size range) that it encounters while foraging. Fruit size is
probably not of great importance, as the woodpecker uses its tongue and bill
to feed on the fleshy pulp of larger fruits. If one assumes that woodpeckers
forage in a manner that maximizes energy intake per unit of time and energy
expended (MacArthur and Pianka 1966, Emlen 1966), it is unlikely that their



1977 CRUZ: JAMAICAN WOODPECKER 179

100 _

tn
m

>
5 -I77-...

C •••

2 so.

O

C
C> 25 -
0

(1)a- EM El-i mTL1:::rl .r'.4-*-1
1110 11-20 21-30 31-40 41-50

FIGuRE 5.-Percentage of prey individuals in five categories of prey Nize for the Jamaican
Woodpecker. Data are based on intact prey items found iii the stomachs of 114 males and 141
females. Open bar = females; dotted bar = males. Food size in min.

rate of energy intake could be increased, or even maintained by selectively
discriminating against certain prey items unless they are highly unsuitable
for other reasons, i.e. toxic, distasteful, or hard to catch.

All three of the techniques of measurement and analysis (percent occur-
renee, percent volume, and prey size distribution) showed an overlap in the
kinds of foods male and female Jamaican Woodpeckers eat (Tables 8,9 and
Figure 6). Usually not only were the same families present but the same prey
species of both animals and plants.

INTERSPECIFIC FOOD COMPETITION.-Orians and Willson (1964) suggested
that in cavity-nesting birds the two resources most likely to be limiting and
thus subject to interspecific competition are nest sites and food, On Jamaica
interspecific competition for nesting holes was apparent and is discussed
below, but in contrast I found little direct evidence that interspecific compe-
tition for food existed, though the feeding habits of the Jamaican Woodpecker
overlapped in varying degrees with those of some of the species occurring in
the same habitat. Table 10 shows the various bird species that were seen
feeding either on similar foods or with similar foraging techniques. In all eases,
however, the species that overlapped the Jamaican Woodpecker were mor-
phologically or behaviorally adapted to obtain foods not available to the
Jamaican Woodpecker, and conversely, the woodpecker's adaptations for
pecking and probing enable it to forage and obtain foods not available to the
other species in this table.

LACK OF DIFFERENTIAL NICHE UTILIZATION BETWEEN THE SEXES.-NO
selection pressure for differential niche utilization between the sexes appears



180 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

TABLE 10,-NICHE AND MORPHOLOGICAL CHARACTERISTICS OF BIRDS THAT OVERLAPPED IN THEIR

FOODS AND FORAGING TECHNIQUES WITH THE JAMAICAN WOODPECKER.'

Body' Bill ' Feeding' Foraging, Foods
Species size length heights methods taken Mean

Columba let,cocephala 3 3 2 4 4 3.2
Aratinga nana 3 3 2 4 4 3.2
Saurothera cettda 2 2 1 4 4 2.6
Hgetomis plucialis 2 2 1 4 4 2.6
Platypsaris niger 3 2 1 3 3 2.4
Tgrannus caudifasciatus 3 2 2 4 3 2.8
Comus iamaicensis 3 4 1 3 2 2.6
Tt,rdus pmaicensis 2 2 3 3 2 2.4
Turdtis at,mntius - 2 2 3 3 2 2.6
M!/adestes genibarbis 4 3 1 3 3 2.8
Vireo altiloquus 4 3 1 3 , 3 2.8
Coereba flaveola 4 4 2 4 \ 4 3.6
Euneornis campestris 4 3 2 4 3 3.2
Pgrrhuphonia iamaka 4 4 1 4 4 3.4
Spindalis zena 3 3 1 4 4 3.0
Icterus let,copteryx 3 2 1 2 2 2.0
Nesopsar nigerrimus 3 2 1 2 3 2.2
Loxigilla ciolacea 3 3 3 3 3 3.0
Lsxipasser anoxanthus 4 4 3 4 4 3.8

'Differences: 1 =Poorly developed, 2= moderately developed,
3 = moderately well developed, 4 = strongly developed.

:Bod. size based on weights taken in the field and ineasurements from Ridgway ( It}02-16).
'Billlength froin Ridgway (1902-16).
'Feeding heights and foraging methods based on field ohservatiom.
'Foods based on field observations and stomach analyses.

to have acted on the Jamaican Woodpecker-both are structurally mono-
morphie, forage in similar fashions, are syntopic, and take the same food. Food
and foraging sites, which would be the two most important selective forces,
appeared to be abundant in Jamaica as opposed to some other island situa-
tions (Grant 1968). The main reasons for the abundance of food and foraging
sites in Jamaica are topographic and climatic diversity, which in turn produce
a number of distinct habitats with an abundance of fruiting trees, bromeliads,
and other microforaging sites. In addition interspecific competition for food
or foraging sites is slight in the absence of other woodpeckers, birds of similar
foraging behavior, and other true frugivores.

As in the case of Asyndesmus lewis (Bock 1971), the Jamaican Wood-
pecker appears to be an opportunistic feeder concentrating on temporarily
abundant food sources. A pair of Jamaican Woodpeckers that made over 20
trips in one day to a fruiting Ficus trigonata abandoned it after 2 days when
its fruits were depleted and responded to the availability of a new food
source, Daphnopsis fruits, and made 25 trips in one day to the latter tree.
During the same periods the two woodpeckers also engaged in generalized
probing, probing into bromeliads, pecking, and gleaning. Bock (1971) noted
that: "Niche dimorphism is unlikely in these instances on theoretical grounds,



1977 CRUZ: JAMAICAN WOODPECKER 181

since selection should favor the ability of both sexes to exploit the full range
of food types. That is, if most important prey species or food sources become
abundant sequentially or are generally evenly distributed, each sex should
be unspecialized enough to feed upon all types." Bock further noted: "It is
important to realize that the selection pressure of intraspecific competition
can act only to increase an already established sexual dimorphism. In a
theoretical monomorphic population, a large female would have the same ad-
vantages as a large male in exploiting foods unavailable to the bulk of the
population. While individual variability might increase there would be no
reason to expect sexual dimorphism to occur."

In a situation where food is not scarce both sexes will profit by having a
longer bill, for Schoener (1967) noted that very large predators (implied here
by longer snout lengths in Anolis lizards and bill lengths in birds) eat either
an equal or a greater range of foods than smaller predators. In addition bill
size is a good indicator of food size: birds with longer bills in general obtain
larger prey items than birds with smaller bills (Grant 1968). As greater prey
size implies greater biomass, which in turn implies more calories and more
energy, both sexes would thus be equally efficient in obtaining energy from
the environment.

In addition one might argue that by decreasing both the body size and bill
size of the female, the female will be less efficient in protecting the nesting
hole against competitors, of which a number of species are present on jamaiea.
Another disadvantage, which may be of greater importance on the mainland
than in insular regions, is the likelihood of predation. Islands usually have
fewer predators so that larger size as a defense against predation may not be as
important, but larger size may still confer a selective advantage. Possible
predators of the Jamaican Woodpecker would be the Red-tailed Hawk (Buteo
jamaicensis), Kestrel (Falco sparuerius), Barn Owl (T!/to alba), and the Ja-
maican Owl (Pseudoscops grammicus). Hence the possible disadvantages of
competing less with the male is counterbalanced by other selective advantages
for the female.

COMPARISON OF WOODPECKER FORAGING DIVERSITY ON J~MAICA AND ON

MAINLAND FLORIDA.-As some related sympatric species of land birds reduce

competition through compression of their foraging niches (Gibb 1954, Haf-
torn 1956, MacArthur 1958, Dixon 1961, Brewer 1963, and Newton 1967), one
might expect that a species occurring in an area with no related species that
might compete for the same resources would have a broader feeding niche.
To test this hypothesis, the foraging behavior of the Jamaican Woodpecker
was compared to the foraging behavior of resident Florida woodpeckers.
Foraging behavior diversity was measured by the information theoryformula
(Shannon-Weaver formula), -Ipilnpi where pi is the percentagd of the total
foraging behavior located in the ith interval (MacArthur and Mac Arthur 1961,
Crowell 1962, MacArthur, Recher, and Cody 1966).



182 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

The results of the comparison of the foraging behavior of the Jamaican
Woodpecker with Florida woodpeckers (Table 11) indicate that the foraging
diversity index of the Jamaican Wbodpecker was almost equivalent to the
pooled foraging diversity indices of seven Florida woodpeckers, 1.59 to 1.72
respectively. In addition the foraging diversity index of the Jamaican Wood-
pecker was higher than any one species of mainland woodpecker. The highest
mainland foraging diversity indices were those of the related Red-bellied and
Red-headed Woodpeckers, 1.27 to 1.35 respectively. The lowest mainland
foraging diversity indices were those of the Dendrocopos woodpeckers
(Downy, Hairy, and Red-cockaded), which had values ranging from 0.36 to
0.61.

No feeding methods were noted in the Jamaican Woodpecker that the
mainland woodpeckers did not use, but the foraging methods were more
evenly distributed in the different categories in the Jamaican Woodpecker.
Thus while both the mainland Red-headed and Red-bellied woodpeckers have
the same number of feeding categories as the Jamaican Woodpecker, the
distribution was not as equitable. Approximately one-half the foraging of
both species were in one foraging category, probing in the Red-bellied
and fruit-and-mast-eating in the Red-headed Woodpeekers. An even more
striking example is furnished by the Red-cockaded Woodpecker, which has
one less feeding category than the Jamaican Woodpecker, but one (pecking)
was used 84% of the time. These results show that the Jamaican Woodpecker
has a more diverse foraging niche than any one of the mainland woodpeckers.

No evidence suggests that limitations in the range of available food
resources are factors underlying these results. Rather certain feeding zones are
incompletely exploited in Jamaica because of the depauperate nature of its
avifauna, and hence it is profitable for Melanerpes mdiolatus to extend into
them. Thus while on the mainland, pecking as a foraging method is rarely
used by the Red-bellied and Red-headed woodpeckers, presumably because
other mainland woodpeckers can peck.more effectively and they (Red-bellied
and Red-headed) are better adapted to obtain food by other means; the re-
lated Jamaican Woodpecker uses pecking to a greater extent (20% of total).

For example, Jamaica has a sufficient diversity of habitats to support
several species of woodpeckers and birds of similar foraging types, as in
continental habitats and islands near the mainland. For instance, the smaller
island of Trinidad (2,980 sq km) has six species of woodpeckers and five of
woodcreepers. Consequently Jamaica has unoccupied niches that permit
niche expansion by species able to exploit the vacant habitats and niches as
well as their own preferred ones (Crowell 1962, Selander 1966). This is
achieved not by species abandoning their former way of life, but by their
becoming inbreasingly diversified in food and foraging behavior as has the
Jamaican Woodpecker.

An interesting and somewhat parallel situation was reported by Zusi (1969)



1977 
C

R
U

Z: JA
M

A
IC

A
N
 W

O
O

D
P

E
C

K
E

R
 

183

TABLE 11,-FORAGING BEHAVIOR OF THE J~MAICAN AND RESIDENT FLORIDA WOODPECKERS.

Species No.' Pecking Probing Cleaning Vegetable Ground Sallying Epiphytes Scaling FDI'

Red-bellied 632 4 53 21 16 1 - 5 - 1.27
Red-headed 248 2 5 26 47 6 14 - - 1.35
Yellow-shafted 386 5 3 2 8 82 - - - 0.69
Pileated 149 62 18 3 16 - 1 - 1.04
Downy 127 89 8 3 - - 0.41
Hairy 34 91 6 3 - - - 0.36
Red-cockaded 315 84 1 2 4 - - 9 0.61
Total 1891 29 21 12 15 18 2 2 1 1.72
Jamaican 1908 22 29 9 27 - 1 12 - 1.59

'Number of observations
-Foraging diversity index
'Percent of total observation.



184 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

on Dominica where no woodpeckers or woodereepers occur. There the
Trembler (Mimidae, Cinclocerthia ruficauda) feeds on a variety of animal
and plant foods, but d6es most of its feeding among epiphytes and clumps of
dead leaves, and takes food from crevices on trunks or tangled vines. Zusi
stated: "In forests of the mainland, members of the Dendrocolaptidae and
Furnariidae forage among epiphytes and leaf trash and on tree trunks. Neither
family is present on the Lesser Antilles, and no mimid has a comparable for-
aging niche on the mainland. The Trembler's feeding adaptations probably
evolved largely on the islands in the absence of species specialized for arboreal
rummaging." In a similar fashion, the Jamaican Woodpecker's use of bro-
meliads and other epiphytes may be related to the absence of Dendrucolap-
tidae and Furnariidae.

Thus through a diversification in its foraging patterns, the Jamaican Wood-
pecker is able to use foods and foraging zones of species that are absent,
and to exploit its insular environment to a degree that in continental habi-
tats is achieved only by the combined efforts of several species of woodpeckers
and birds of similar adaptive types.

BREEDING BEHAVIOR AND BIOLOGY

The jamaican Woodpecker has an extensive communications and display
repertoire, the former by both vocal and mechanical means. Both sexes take
part in the vocalization, drumming, and displays discussed below. In contrast
to the situation described by Kilham (1961) for the Red-bellied Woodpecker in
which the males are the more active participants, the female Jamaican Wood-
pecker appears to be almost as active a performer as the male. The categories
used below are based on Kilham (1961), Lawrence (1967), and my interpre-
tations of their meanings based upon the circumstances attending their com-
munications and displays on repeated occasions.

VOCALIZATIONS

LOCATION CALL.-A loud kaaa that may be repeated two or three times in
succession. This call is heard throughout the year, but is heard more frequently
during the breeding season. The location call is used to reveal a bird's presence
and location to another woodpecker or to its young. Upon hearing it, one or
more Jamaican Woodpeckers usually answer with the same call, or signal a
response by drumming and, depending upon the circumstances, one of the
participants may fly toward the other.

TERRITORIAL CALL.-A very loud kaaaah that usually includes shaking or
vibrating of the entire body. Its use is to advertise the territory, and it is also
given when disturbed by a human intruder. Upon hearing it one or more
woodpeckers answer with the same call or signal a response by drumming.

Low INTENSITY ALARM CALL.-The low intensity alarm call kao is usually
expressed singly but is sometimes repeated in a continuous and scolding



1977 CRUZ: JAMAICAN WOODPECKER 185

fashion. It suggests, as Lawrence (1967) noted, alertness to danger, to the
unexpected, unusual, or strange. Birds uttered it, for example, when I sur-
prised an individual feeding, or came too close to a nesting tree or to an adult
bird with a juvenile.

HIGH INTENSITY ALARM CALL.-The high-intensity alarm note, a loud

wee-cha wee-cha, is usually uttered in intraspecific encounters between
woodpeckers not belonging to the same family group.

BREEDING CALL.-Whereas the above calls were heard during December,

the breeding call, krirr, krirr, and the following mutual recognition call
were heard only during the breeding season. The breeding call was heard
throughout the reproductive period, but was uttered more frequently during
the early reproductive period before egg-laying. It is somewhat reminiscent
of the kwirr note of the North American Red-bellied Woodpecker, although
harsher. It appears to have the same functions as in the Red-bellied Wood-
pecker, namely to bring the sexes together and to establish and maintain
pair bonds during the breeding season (Kilham 1958, 1961, Stickel 1965).

MUTUAL RECOGNITION OR CONTACT NOTES.-A low intimate whirr-whirr

sound was given by members of a pair and apparently functions in mutual
recognition and to reinforce the pair-bond. This sound was heard during court-
ship and when one member of the pair alighted near its mate to relieve it
from excavation or ineubation duties.

MECHANICAL COMMUNICATION

Mechanical means of communication are important in this species, as they
are in many other woodpeekers. They serve a variety of functions, i.e. ex-
pressing excitement, territorial dominance, or displacement.

DRUMMING.-Drumming is the burst of very closely spaced taps produced

when the woodpecker's bill hammers on any resonating structure, such as a
dead or hollow branch. Drumming in the Jamaican Woodpecker appears to
have the same function as in other woodpeckers (Kilham 1959, 1961, Law-
renee 1967, Bock 1970, 1971), that is, the assertion of territorial dominance
and the attraction of a mate. In contrast to the Red-bellied Woodpeeker, in
which drumming is far less common and is used frequently or solely by the
male (Kilham 1961), both sexes of the Jamaican Woodpecker drum.

RITUAL AND MUTUAL TAPPING.-Ritual and mutual tapping were noted in

the Jamaican Woodpecker and appear to serve the same function as in the
Red-bellied Woodpecker, the strengthening of the pair bond (Kilham 1958).
The arrival of the mate while the partner is in or near the nest hole appears
to be necessary for the performance of ritual or mutual tapping, and it may
be enacted either by the male 6r female in response to the appearance of its
mate. In ritual tapping the mate begins tapping in one spot upon the arrival
of the partner at the nest hole. Tapping may either take place outside or in-
side the nesting hole. Outside the nest hole, tapping usually takes place just



186 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

below the lower rim of the hole. An elaboration of ritual tapping is mutual
tapping in which both members of a pair tap together.

DISPLACEMENT TAPPING.-Displacement tapping is the tapping produced
under stress or confusion. This was witnessed, for example, when a dunimy
Jamaican Woodpecker skin was put near the nest hole while the woodpeckers
were away. Upon arrival the male Was confused, did not enter the nesting
hole, and circled the dummy woodpecker. Instead of attacking the dummy
bird (as it did later), it flew to the base of a large bromeliad and tapped for
50 seconds.
DISPLAYS

CREsT-RAINING.-Crest-raising was done alone or accompanying other dis-
plays. It is lisually indicative of a disturbed or excited condition.

BILL-WAVING.-The bill-waving display is usually done iii conjunction
with either low or high intensity alarm calls. The woodpecker, with rectrices
spread and crest raised, swings its head from side to side iii a 180 degree angle.
Lawrence (1967) described similar types of displays in the Hairy and Downy
woodpeckers, Yellow-shafted Flicker, and Yellow-bellied Sapsucker, but it has
not been reported in other Centums woodpeckers

RESTING MOTIONLESS.-Members of a pair or a family group may cease
all activities and remain motionless within close proximity of each other for
periods of up to 20 minutes. This form of behavior has also been observed
in the Red-bellied and the Hairy Woodpeckers by Kilham (1966); and in
palearctic woodpeckers by Pynnunen (1939). Kilham noted: "for such active,
energetic birds, these... inotionless periods are a striking form of behavior."
TERRITORY, TERRITORIALITY, HOME RANGE, AND NEST HOLE COMPETITION

DEFINITIONS.-A territory in this study is defined as a space within which
an animal is aggressive toward and usually dominant over certain categories
of intruders (Emlen 1957), whereas the home range is the total area that a
bird habitually occupies and normally confines its movements to (Burt 1943).
In some birds home range may be the same as the territory if the bird
defends the whole area, and in other birds the territory may be restricted to
the nesting area (for a fuller treatment of territories see Nice 1941, Hinde
1956, Brown 1964). I found that the Jamaican Woodpecker occupies over-
lapping home ranges and that territorial defense is restricted to the vicinity
of the nest.

HoME RANGE.-Home ranges of 12 breeding pairs of jamaican Wood-
peckers obtained during the 1969 breeding season are given in Figure 6. The
size of the home range in the Worthy Park area (modified wet limestone
forest) varied from ,74 to 2.6 ha (x = 1.4 ha). The size of the home range
appears to depend upon the suitability of the area for foraging. For instance
pair No. 1, which occupied a more open area with fewer trees, had the largest
home range. In contrast, pair No. 3, which occupied a heavily wooded area



1977 CRUZ: JAMAICAN WOODPECKER 187

73 146rn

8 7• 6
5.

11. .410
9

3.•12

2.

1o
FILL·RE 6.-Home ranges of 12 breeding pairs of Jamaic·an Woodpeckers during the 1969
breeding season in Worthy Park. Circles indicate nesting tree. Those.ranges without circles are
for family groups with postfledgling young.

with many fruiting trees and foraging sites, had the smallest home range.
Frequently (27 observations) pairs occupying neighboring home ranges fed
within full view of one another and no aggressive behavior was detected. In
one instance (30 June 1969), the male and female of pair No. 5 and the male
of pair No. 4 fed without conflict in the same fruiting tree.

TERRITORIAL UNIT.-In the Jamaican Woodpecker, the territorial unit con-
sisted of the nest tree and its immediate surroundings. Using the nest tree as
the approximate center, the territory appears to have a maximum radius of
approximately 40 meters. The principalelements of the territorial behavior
included (Lawrence 1967): (1) calls and drumming, expressing challenge or
protest; (2) display, elaborate conspicuous movements expressing protest and
having the effect of threat; (3) pursuits, taking the form of stalking, chasing,
or supplanting attack (in which the attacker dislodges the trespasser by flying
at him and taking his place); and (4) fighting, involving bodily contact.

TERRITORIAL DEFENSE.-Adjacent pairs often fed within full view of one



188 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

another in the area of home range overlap, but if a strange or neighboring
woodpecker intruded the vicinity of the nest area (black circles in Fig. 6), the
resident woodpecker actively defended it. The size of the defended area was
difficult to quantify because intrusion of the territorial boundary of nesting
woodpeckers by another woodpecker was rarely recorded. In three seasons of
nesting observations, territorial conflicts were recorded only nine times-six
of the conflicts took place when a strange woodpecker landed on a tree
adjacent to the nest tree. In each case the intruder was repelled by either
chasing or supplanting attacks. The minimal distance that an intru(ling bird
was chased was 15 m and the maximal distance was 40 m. In one instance
the resident male chased an intruding male for approximately 30 m before
flying back to the nest tree. The intruding male flew to an exposed branch
on a tree 50 m from the nest and remained for approximately 5 minutes;
although the intruding bird was still in the home range and in full view of the
resident male, no further territorial behavior was recorded.

An example of territorial behavior is taken from my field notes. On 7
july 1969, a strange male landed on the tree where the Swansea pair was
brooding the young. The resident male arrived momentarily. Instead of enter-
ing the nest hole, he flew to a dead branch on the same tree and started
to drum. The intruding male remained close to the branch and did not re-
spond to the drumming. The resident male with crest raised flew toward the
intruding male (supplanting attack), which flew off to another branch on the
same tree. Next a series of pitrsuits began, the resident male chasing the in-
truding male from one limb to another or following him around the tree
trunk. This lasted for approximately 3 minutes, culminating with the resident
male chasing the intruder off the tree and following him for approximately
35 m. During this time, the female did not join in the chase, but flew to an

exposed dead branch on the nest tree and began uttering high-intensity alarm
notes.

EXPERIMENTS WITH DUMMIES.-Before describing the experiments with
dummies and mechanical stimuli, I must emphasize that while these experi-
ments are useful and helpful in gaining insight into woodpecker behavior,
specifically territorial behavior in this case, the results, if not corroborated
with the action of live woodpeckers, might be misleading in some cases.
For instance Lawrence (1967) noted: "The woodpecker's encounters with
dummies... which elicited ferocious attacks and pecking carried to the
extreme, certainly give a distorted idea of the birds' normal way of life. In
the wild the woodpeckers display to live opponents that move and never
remain still in one place. Swiftly and impressively the woodpecker reacts to
the sight of an aggressively displaying bird of its own kind and every one of
its movements has, in turn, a crucial impact upon the latter's further be-
havior." Nevertheless, I feel that (bearing these limitations in mind) ex-



1977 CRUZ: JAMAICAN WOODPECKER 189

periments with dummies and mechanical stimuli are of value in gaining
insight into behavioral phenomena that one has little opportunity to observe
in nature.

I experimented with dummy woodpeckers on three different nesting pairs
in the summers of 1969 and 1970. The following results obtained from a pair
nesting in a prickly yellow tree are characteristic. On 2 July 1970 at 0900
I placed a dummy male Jamaican Woodpecker 1 m from the prickly yellow
tree nest while the parents were away. At 0908 the male landed on the
nest tree and saw the dummy. Instead of going into the nest hole as he nor-
mally did, he uttered high intensity alarm notes and moved his head from side
to side (bill waving) and simultaneously started approaching the dummy.
When he was 2 m from the dummy he flew to another tree and began tapping
at the base of a bromeliad (displacement tapping). While he was tapping,
the female landed on the nest tree, visibly agitated with crest raised, and
uttered high intensity alarm notes. The male flew back to the nest tree with
crest raised and began attacking the dummy with blows directed at the
head region. The female did not join in the attack, but remained on the tree
uttering high intensity alarm calls. The resident male's excitement was so
great that he continued the attack while I was removing the pole with the
dummy from its position. The pole with the attached dummy was next placed
at distances of 5, 10, 15, and 20 m from the nest hole. At 25 m from the nest
tree, the male discontinued the attack, but remained on the nest tree uttering
high intensity alarm notes for 6 minutes, after which he entered the nest
hole and resumed brooding the young. The female flew to the nest tree 18
minutes later and relieved the male. The male, completely ignoring the
dummy, flew off the nest tree and began to forage.

TERRITORIAL DRUMMING,-The use of dummies was not the only way to

elicit territorial defense. On many occasions tapping on a clip board with a
pencil near the nesting hole excited the parents. For example on 14 July
1969, the female of the pair nesting in the guango tree was brooding young
and the male was presumably foraging. At a distance of 5 m from the nest
branch, I began tapping. Immediately the female stuck her head out, uttered
a territorial call, flew to a dead branch on the nest tree, and began drumming.
Momentarily the male joined the female on the tree, and began drumming
and uttering high-intensity alarm notes. I repeated the tapping 15 m from
the nest, the male flew toward me, landed on a vine 5 m from me, and with
raised crest and bill-waving motions began uttering the high-intensity alarm
call. I moved to 25 m from the nest and repeated the same procedure, again
obtaining the same response from the male, but at a lower intensity. At 30 m
from the nest, the resident male answered my tapping, but he remained on
the nest tree. The following morning I repeated the same procedure, but
this time started 45 m from the nest. No response was obtained at 45 or 40



190 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

m from the nest, but at 35 m the male responded by drumming. At 27 m
from the nest the male not only responded to my tapping, but landed about
6 m from me and began uttering his high-intensity alarm note.

The results of the experiments with dummies and mechanical stimuli (tap-
ping) were not unexpected in view of the previous observations on territorial
behavior between resident and intruding woodpeckers, and served to rein-
force and clarify those observations.

INTERSPECMIC TERMTORIAIATY.-Although the observations above demon-
strate that the Jamaican Woodpecker exhibits:intraspecific territoriality, I saw
no interspecific aggressive behavior. On some trees containing active wood-
pecker nests other species of birds nested, and some even used unoccupied
woodpecker holes. In one instance three hole-nesting birds, a Saffron Finch
(Sicalis flaueola), a Starling (Stumus oulgaris), and a Jamaican Woodpecker
nested in different holes in the same branch. The Saffron Finch often chased
the woodpecker as it entered or left the nest hole, but the woodpecker never
reacted aggressively to either species.

On 14 occasions in a 3-day period I saw from one to three of the in-
troduced common Parrotlets (Forpus passerinus) enter an active woodpecker
hole containing young. Although in two instances the parent woodpeckers
were on the nest tree, they watched the Parrotlets with indifference and made
no efforts to drive them away. Another time I watched the male woodpecker
land outside the nest hole with food while the Parrotlets were still inside the
hole. Again the woodpecker acted indifferently and waited until the Parrot-
lets left. Apparently the Parrotlets did not harm the young because two
juveniles fledged successfully from this nest, but because of its inaccessibility
I was unable to determine the original number of young. Skutch (1948) also
observed a similar lack of aggressive behavior between the Golden-naped
Woodpecker ( Tripsurus chrysauchen) and other Central American birds.

Murray (1971) hypothesized that interspecific territoriality is misdirected
intraspecific territoriality, and that it usually occurs in closely related species
which share similar features that stimulate intraspecific territorial aggression.
Thus Selander and Giller (1959) found that the Red-bellied and Golden-
fronted woodpeckers (Melane,pes aurifrona) maintained exclusive territories
where they occurred together near Austin, Texas. These two species re-
semble each other in plumage, vocalization, foraging, and nesting. The lack of
interspecific territoriality in the Jamaican Woodpecker may be related to the
absence of other species of similar morphology, vocalization, or habits that
might otherwise stimulate territorial aggression.

DiscussioN.-In the Jamaican Woodpecker territorial behavior is confined
to the space around the nest and does not include the home range as it
does in some mainland woodpeckers (Kilham 1958, 1961, 1968, 1969, Selander
and Giller 1959, Ligon 1970). The reason for this seems to be that food
is not a limiting factor in Jamaica and large territories are uneconomical,



1977 CRUZ: JAMAlCAN WOODPECKER 191

in terms of time-energy budgets. In contrast, on the mainland where six or
seven species of woodpeckers are sympatric, the species do not appear to have
as diverse foraging niches as the Jamaican Woodpecker, and holding a large
territory that includes the species' foraging zone may be advantageozis. Brown
(1964) noted that aggressive (territorial) behavior is generally employed by
individuals in the acquisition of goals that tend to maximize individual sur-
vival and reproduction. Natural selection should favor aggressive behavior
within a population when these goals are consistently and easily accessible
through aggression, but should not favor it when they are not obtainal,le. For
example when food is abundant or transient (fruiting trees) as appears to he
the case in most Jamaican Woodpecker habitats, no territorial system is
needed to defend foraging zones, and the territory if present is restricted to
the space around the nest.

THE ANNUAL BREEDING CYCLE

Studies on the breeding biology of Jamaican Woodpeckers, primarily in
Worthy Park, enabled me to complete a fairly accurate picture of their annual
reproductive cycle. Figure 7 shows the combined breeding data of 19 pairs
of woodpeckers. Most data (17 pairs) were obtained in a wet limestone
forest, and the dates for certain events in the annual cycle may differ in
other habitats or parts of the island. For instance, in New Falmouth (man-
grove woodland) I saw a female woodpecker feeding a fledged bird on 11
May 1970; and Perkins (1970) saw a pair of woodpeckers in Trelawny ex-
cavating a nest hole in December 1933, from which young fledged iii April.
Snow and Snow (1964), in their work on the breeding seasons and annual

POST-NESTLI NG K..................„................................i,

NESTLING 6:imu:..:..:..::..::..:....:. I::.:1:1:.--'.-1

INCUBATION r.............................................d

EGG-LAYING i 01

EXCAVATION 1.............:......I..........................v

DEC | JAN | FEB |MAR |APR |MAY |JUNE |JULY |AUG |SEPT |OCT |
FiC:l'RE 7.-The annual breeding cycle of the Jainaican Woodpec'ker based on 17 pairs in \Vorthy
Park (St. Catherine), one pair at Baron Hill, one pair at New Falmouth (both in Trelawny).



192 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

cycles of Trinidad landbirds, found that the breeding season varied locally,
being longest in regions of high rainfall and probably shortest in the very dry
parts of Trinidad.

The breeding season of the Jamaican Woodpecker (excavating to post-
nesting) extends 'at least 10 months (December to September), which is
probably longer than that of any mainland temperate woodpecker. Lawrence
(1967) found that in the four species she studied (Hairy and Downy wood-
peckers, Yellow-shafted Flicker, and Yellow-bellied Sapsucker) the breeding
cycle (excavating to post-nesting) extended from 4 to 5 months. The ex-
tended breeding season in the Jamaican Woodpecker is most likely related
to the more uniform tropical climate, which provides sufficient food for the
nestlings and adults throughout most of the year. In temperate zones with
their seasonal variations in weather, reproduction must be timed to a period
of minimum stress on the adults and maximum probability for the survival of
both parents and the young, usually in the spring and summer.

PAIR-BOND RELATIONSHIPS.-Although no birds were banded in these
studies, the recording of male and female Jamaican Woodpeckers feeding
together when no reproductive activities were under way in Worthy Park
suggests that they maintain year-round pair bonds. During December 1969,
I made 21 observations of 7 different male and female woodpeckers (pre-
sumably pairs) foraging together in the home ranges occupied by pairs 1,3,
5, 7, 10, 11, 12 (Fig. 6). During this period male and female woodpeckers
occupying the same home range kept in contact by using location calls. These
results differ from those of Kilham (1958, 1961) and Stickel (1965) for the
congeneric Red-bellied Woodpecker in which the sexes do not maintain a
year-round bond and occupy different home ranges during the nonbreeding
season. My data are similar to Skutch's (1969) observations on the Golden-
fronted Woodpecker (M. aurifrons) in which the sexes remain together
throughout the year.

NEsT SITES.-Data on 68 nest sites in Jamaica (Table 12) show that the
jamaican Woodpecker prefers to nest in dead branches on live trees, such
sites accounting for 53% of the total. Other sites, listed in order of preference,
were dead trunks, dead branches in dead trees, and live branches, No nests

TABLE 12.-NESTING SITES OF THE J~MAlCAN WOODPECKER.

Live Tree Dead Tree
Height of Nesting

Hole Trunk Live Branch Dead Branch Trunk Branch

under 4.6 m - 3 4 2
4.6-9.2 m - 1 26 11 5

over 9.2 m - 1 7 2 6
Total - 2(396) 36(53%) 17(25%) 13(19%)



1977 CRUZ: JAMAICAN WOODPECKER 193

were in live tfunks. Aside from the existence of suitable dead trees or dead
branches for nesting, the main reason for the preference of such sites is that
the Jamaican Woodpeeker belongs to a group of woodpeckers that are not
structurally specialized for pecking (see Burt 1930, Bock and Miller 1959,
Spring 1965). It thus avoids the use of the heavier live woods more difficult
to dig into. Nest heights varied from 3 m to over 20 m, with an approximate
average of 9 m.

Dimensions were measured at 16 nest holes. The shallowest hole was 15 cm
deep (measured from lower rim of entrance hole) and the deepest hole was 60
cm, the average depth was 31 cm. (David Johnston told me of a hole 180 cm
deep in a dead coconut palm in St. Andrew Parish.) The differences in
depths could be related to the type of wood used, because the shallower
holes were in live wood. In addition the preempting of nesting holes by
Starlings probably forced the construction of hastily made second holes in
some cases. Dennis (1969) noted that the unusually shallow nesting holes made
by the Flicker on Nantucket Island were probably related to the fact that
Starlings took over the original holes thus forcing the Flickers to construct
a second hole quickly. The nest hole diameter of the 16 nests ranged from 47
to 77 mm, with a mean of 65 mm. Rarely was the same nest hole used for
more than one season in Worthy Park. The most important factor involved
in this appears to be the presence of aggressive nest-hole competitors such as
Starlings and Saffron Finches. One nest hole in a heavy woodland, where no
Starlings or Saffron Finches were present, was used by woodpeckers for three
seasons.

NEST-SITE SELECTION.-Although most woodpeckers that I noted had
already started their nest holes, in one pair that had not started excavating
the male took the lead in nest-site selection. On 18 April 1970 I watched a
male Jamaican Woodpecker moving up a dead prickly yellow tree, tapping
it lightly with his bill. Upon reaching the top of the dead trunk, he began
uttering his breeding call, and the female answered. A few minutes later the
female landed on the tree about 3 m below the male. The male flew toward
the female and landed approximately 0.5 m below her. He moved up toward
the female and, when they were at approximately the same level, they en-
gaged in mutual tapping. After mutual tapping three times, the female flew
off followed by the male. The next day the male began excavating a nesting
hole in the prickly yellow tree. These observations are similar to Kilham's
(1961) report for the Red-bellied Woodpecker.

IMPORTANCE OF THE JAMAICAN WOODPECKER TO OTHER HOLE-NESTING

BEDS.-The Jamaican Woodpecker, by creating nesting and shelter sites for
many species of Jamaican birds, plays an important role in the community.
I have noted the following species using woodpecker holes for breeding:
Sparrow Hawk, Yellow-billed Parrot (Amazona collaris), Common Parrotlet,
Rufous-tailed Flycatcher (Mgiarchus validus), Purple Martin (Progne sums),



194 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

Starling, and Saffron Finch. In addition the following birds, based on their
nesting habits (Bond 1971), quite likely also use woodpecker holes: Black-
billed Parrot (Amazona agilis), Barn Owl (Tgto alba), Jamaican Owl, Stolid
Flycatcher (M!/iarchus tuberculifer), and Dusky-capped Flycatcher (M, barbi-
rostris)  With the exception of the Starling and possibly the Saffron Finch,
those species that I observed appeared to be using abandoned woodpecker
holes and did not seem to compete with woodpeckers for active nest holes.

STARLING-WOODPECKER NEST HOLE COMPETITIoN.-Starlings were first in-
troduced into Jamaica in 1903 near Annoto Bay (Saint Mary's Pafish) and since
then have become well established in many parts of the island, including
Worthy Park. As is true for Europe (Howard 1920, L6hrl 1956) and North
America (Kilham 1958, 1960, 1968, Lawrence 1967, Dennis 1969), Starlings
are effective nest-hole competitors of some woodpeckers, and the same was
also true in Jamaica. The following two accounts from the summer of 1969
demonstrate this competition and show the lack of inteispecific aggression
on the part of the Jamaican Woodpecker. On 17 june 1969 I  found a pair
of woodpeckers in the final stages of nest hole construction in a dead branch
of a guango tree. By 22 June 1969 the pair had apparently completed the
hole. That same day two Starlings arrived and perehed on the branch con-
taining the hole. The woodpeckers appeared to be disturbed, but showed
no aggressive behavior toward the Starlings. The Starlings used an aggressive
display involving various shrill calls and wing-flapping. The same day, when
both woodpeckers were away from the tree, the Starlings went into the nest
hole. When the male woodpecker arrived he made no attempt to take over
the nesting hole, but remained on a nearby branch. As heavy rain began at
this time, I was unable to continue watching. The next day the Starlings
were taking nesting material into the woodpecker hole, and the woodpeckers
were excavating a new cavity in the same branch. On 26 June 1969 I noted
Starlings with young in a woodpecker hole in a dead branch of a guango tree.
On another dead branch on the same tree a pair of woodpeckers were ex-
cavating a nest hole. The Starlings fledged their young before the w,ood-
peckers finished their hole. The woodpeckers immediately stopped work on
their hole, modified the Starlings' hole, and nested in it. Possibly the wood-
pecker originally built this hole, and the Starlings took it over. The fact that
in both cases the woodpeckers were able to fledge young seems that (for
the present) the Starlings are not an obstacle to the nesting success of the
Jamaican Woodpecker. Subsequent increases in the Starling populations may
pose a threat in the future, not only to the Jamaican Woodpecker but to
other hole-nesting birds.

COPULATION.-COpulation in the jamaican Woodpecker was observed 12
times in 4 different pairs. It occurred most frequently during the early part
of the breeding cycle (excavation and egg-laying), but also once during in-
cubation. In one pair it was seen approximately 28 days before egg-laying.



1977 CRUZ: JAMAICAN WOODPECKER 195

Copulation occurs in a manner similar to that described for other woodpeckers
(Kilham 1958, Stickel 1965, Lawrence 1967, Bock 1970). The male, with
wings fluttering, mounts the female, turns his body at a right angle to hers, and
after copulation falls off to the left. Copulation always took place on a hori-
zontal branch and lasted from 5 to 15 seconds. Mutual recognition notes were
always exchanged as the male approached the female, but no sounds were
uttered during copulation.

ExcAvATION PERIOD.-Both male and female Jamaican Woodpeckers ex-
cavate the nest hole. Of 1721 minutes spent watching the construction of
four nests the male performed 62% of the excavating and the female 38%. The
average working period for both sexes lasted 21.5 min. One excavation I
watched from the start of nest-hole construction to completion took 21 days.

Lawrence (1967) divides the process of excavation into three stages: (1)
boring the corridor that forms the entrance part of the cavity, (2) the curved
link between the corridor and the cavity, (3) the cavity itself (for information
on how each stage was determined see Lawrence 1967). For the nest men-
tioned above stage 1 took 9 days, stage 2 took 4 days, and stage 3,8 days. The
length of time spent in each stage is probably highly variable (as Lawrence
noted), depending upon the condition of the wood (whether live or rotten) and
the amount of disturbance. During nest-hole construction members of a pair
maintained contact with one another through location and breeding calls and
engaged in ritualistic and mutual tapping, both thought to be associated with
the maintenance and strengthening of a pair bond.

CLUTCH SIZE.-Although most nest holes were inaccessible because of their
height, I was able to determine the contents of two nests low in dead trees.
The clutch size was three and four eggs, respectively. In addition the examina-
tion of 4 females revealed the presence of from 3 to 5 follicles on the surface
of the ovary.

LENGTH oF INCUBATION.-Examining the inside of one nest with a rotatable

mirror from the time of egg-laying to hatching revealed that one white egg
was laid daily until a clutch of three was reached. The female laid each egg
during the early morning hours. Hatching started 13 days after the first egg
was laid and continued for 2 more days. These results are similar to the report
of Skutch (1969) for the Central American Melanerpes aurifrons, a species
with an incubation period of 12 days. Stickel (1965) reported an incubation
period for Melanerpes carolinus of 11.5 days * approximately 6 hours.

ArrENTIVENESS.-Both sexes take part in incubation during the day, but
only the male incubates at night. During 1955 minutes of observations on
three nests at least one adult was inside the nesting hole 88% of the time, the
average incubation session lasting approximately 31.7 minutes. The division
of daytime attentiveness was 47% for the male and 41% for the female. Percent
attentiveness in North American woodpeckers ranged from 68% in Asyndes-



196 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

mus lewis (Bock 1970) to "not less than 95% of the time during the day" in
the four species studied by Lawrence (1967).

THE RELIEF RITUAL.-Nest relief in the Jamaican Woodpecker varied from
no contact (one bird leaving before the other arrives) to elaborate ritualistic
behavior involving mutual tapping and mutual recognition notes. In many in-
stances when the incubating individual wanted to be relieved, it would look
out the nest hole and utter its location call. The mate usually arrived at the
nest shortly thereafter. The relieving woodpecker usually landed below the
nest hole and moved toward the nest. Often it did not go directly into the nest,
but inserted its body two or three times in intention movements into the
cavity before entering. Lawrence (1967) noted that the number of intention
movements performed by North American woodpeckers depends on the
nervous state of the birds as well as on the situation outside of the nest. No or
few intention movements denote quiet surroundings and a low level of ner-
vous tension in the bird.

HATCHING DAY AND BROODING DURATION.-Lawrence (1967) reported that
when the eggs begin to hatch "the brooding rhythm changes from the slow
regular pattern of incubation to a chopped-up pattern of shorter, irregular

"brooding sessions. Table 13 compares the rhythm of incubation between the
last day of incubation and hatching day, which was also the first day that
food was brought to the nest. The date of hatching of the prickly tree pair
was also verified by the use of a mirror. Whereas the duration of brooding
per session decreased from incubation to hatching, no decrease in attentive-
ness was detected at either nest.

TABLE 13.-THE DIFFERENCE IN INCUBATION RHYTHM BETWEEN THE L+~sT DAY OF INCUBATION
AND HATCHING DAY, DATA OBTAINED DURING MORNING HOURS (0800-1200)

Last Day of Incubation Hatching Day
Number Number

of Average Attentiveness of Average Attentiveness
Pair Sessions Minutes Percent Sessions Minutes Percent

Guango tree
Pair 1969 7 31.2 91 21 10.5 92
Prickly Pair
1971 6 35.3 88 23 9.4 90

These changes in rhythm are so distinct that, together with the initial food
brought, they serve as reliable criteria for determining the duration of the
nestling period (hatching day to fledgling). Because the nest of the guango
tree pair, which was studied intensively during the summer of 1969, was in an
inaccessible place, the nestling period and the fledgling time were determined
by these criteria. The first nestling left the nest 29 days after hatching and
the second nestling after 30 days. Stickel (1965) rep6rted that the nestling



1977 CRUZ: JAMAICAN WOODPECKER 197

period for M. carolinus in Illinois varied from 22 to 27 days, and Kilham (1961)
noted the nestling pefiod for the same species in Maryland to be 26 days.
Skutch (1969) reported that the young of M. aurifrons remain in the nest for
about 30 days, and that the young of M. rubricapillus flew 31 and 33 days after
the parents were first seen to take in food.

BROODING RHYTHM.-Both parents shared brooding the young during the
day almost equally, 44% for the male and 56% for the female, based on 2,488
minutes of observation at three nests. During thenight only the male brooded
the young. Equal sharing of nesting duties during the day enables each parent
to spend the maximum time possible away from the nest. Bock (1970) and
Stickel (1965) reported similar brooding behavior in the Lewis and Red-bellied
woodpeckers. Lawrence (1967) found that in the Yellow-bellied Sapsucker, the
sexes share brooding almost equally, whereas in the Flicker and the Hairy
Woodpecker the female, but in the Downy Woodpecker the male, assumes
most of the duty.

In the Jamaican Woodpecker diurnal brooding attentiveness was sustained
at a high level until the 10th day, ranging from 93 to 83%. By the 18th day,
brooding was shortened to 41%, and brooding ceased entirely by the 26th day
(Fig. 8), 3 days before the first young left_the nest.

100

in
Lf) ,
CD
C
4 75
>

C
Cl)

< 50

C
CD
U

11)
0- 25

O 1
10 20 30

Day
FIC,rRE 8.-Variation of diurnal brooding attentiveness in the Jainaican Woodpecker Chased on
1853 minutes of observation in the gliango tree pair)  Percent attentiveness refers to the percent
of the total period of observation that the parents were inside the nesting hole (presumably
brooding).



198 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

FEEDING RATES.-Nesting pairs averaged 10.3 feedings of the young per
hour (N - 335) with a range of 2 to 22. This feeding range is within the ranges
reported in the literature for North American woodpeckers. Stickel (1965)
found that the diurnal variation in the feeding rate for the Red-bellied Wood-
pecker was from 4.5 to 15.7 feedings per hour (no average given). Lawrence
(1967) found that the average feeding rate of the Flicker was 2.2 feedings
per houf, whereas that of the other three species varied from 8.8 in the Yellow-
beilied Sapsucker to 14.8 in the Downy Woodpecker. For the Lewis Wood-
pecker, Bock (1970) reported feeding rates varying from 2 to 62 with an
average of 15.1 feedings per hour.

The mean feeding rate per hour was 5.7 for the male and 4.6 for the female
in the Jamaican Woodpecker. This same pattern was also present in Law-
renee's (1967)·study of North American woodpeckers and Pynn6nen's (1939)
study of European woodpeckers. Lawrence noted that this is due to the fe-
male's preoccupation with brooding and defense, activities for which she
usually undertakes with greater responsibility. Although this may be true
for the woodpeckers that she studied, there was no evidence that the female
played a greater role than the male Jamaican Woodpecker in these activities.

In general the Jamaican Woodpeckers spent a greater proportion of the
morning and afternoon hours foraging, and this variation in diurnal activity
pattern has a direct bearing on the rate of feeding the young (Fig. 9), with
peaks during the morning (0800-0900) and afternoon (1600-1700). Stickel
(1965) observed a similar pattern in the feeding rates of M. Carolinus.

The rate of nestling feeding was lowest during the first and last week and
highest during the second and third wek of the nestling period (Fig. 10). Law-
renee (1967) found that in the course of the nestling period the Flicker's feed-
ing rate increased gradually until it reached a level that was sustained to the
end. The feeding rates of the other three species of woodpeckers studied by
Lawrence (Hairy, Downy, and Sapsucker) showed a peak about the middle of
nest life after which the feeding rates declined, reaching a low at the time of
fledgling. The decline in the feeding rate toward the end of the nesting cycle
is probably related to behavioral changes in the adults and may play a role
in encouraging the nestlings to leave the nest.

RELIEF RITUAL.-The relief ritual described in the incubation section re-

mained unchanged into the nestling period, but as the rate of feeding the nest-
lings increased, the meeting at the nest by the parents occurred less fre-
quently, and as a consequence:relief rituals were less frequently performed.

FOOD AND FEEDING BEHAVIOR,-Feeding of the young commenced almost

immediately after the first egg hatched in the prickly yellow tree nest. Food
items at first were small invertebrates and fruits, and the proportion of ~
larger food items increased as the nestlings grew older. Analysis of 283 feeding
visits in which the food items brought could be identified showed that animal
foods accounted for 59% and vegetable foods 41%. Diets of the nestlings



1977 CRUZ: JAMAICAN WOODPECKER 199

23

20

1 13
0

(1)

10

C

(D
LL- 5

lilli11111111
6 7 8 9 1011121 2343 6 7

Hour
FIGURE 9.-All day observations of feedings by the Jamaican Woodpecker on the 18th day of the
nestling period Chased on 783 minutes observation of guango tree pair, 1969).

were almost as diverse as those of the parents, ranging from small fruits to
insects to lizards, identified vegetable foods included fruits of Cecropia,
Ficus, and Daphnopsis. On one day the guango tree pair (1969) brought the
nestlings fleshy fruits of the Daphnopsis tree 35 times. Identified animal foods
included snails, lepidopterans, orthopterans, ants, coleopterans, and small
Anolis.

Lawrence (1967) divided the nestling period into three stages according to
the method of feedings: inside feeding, corridor feeding, and outside feeding.
She noted that each stage is determined by the phase in the development
of the nestlings. These three stages were readily identified in the Jamaican
Woodpecker where inside feeding, in which the parent goes all the way into
the hole, lasted until the 17th day. As the nestlings grow older, they began
climbing up to the entrance as soon as they heard the parents arrive. The
parent bringing food met the nestlings in the corridor and fed them there.
Outside feeding, in which the parents do not enter the hole, began on the
20th day.

NEsT SANITATION.-During 1853 minutes of nestling observations, the
parents removed fecal material,from the nest 39 times, the male accounting
for 65% of them. These observations differ from Kilham (1961) who never saw



200 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

20...

~ 15
-C

6-

11)

0 10
Clo
C

CD
CD

LL- 5--

Day 1 1 1
0 10 20 30

Nestling Period
FIGURE 10.-The feeding rates of a pair of Jamaican Woodpeckers during the nestling period.
Based on 15 hours of observations during five mornings (0800-1100).

M. carolinus remove fecal material, and from Stickel (1965) who saw only
the male M. carolinus remove fecal material from the nest. Lawrence (1967)
found that males of the Sapsucker, Flicker, and the Hairy and Downy wood-
peckers played a greater role than the females in the removal of feces from
the nest, the male accounting for 70% of the total observations. The feces of
the nestling Jamaican Woodpeckers are enclosed in a whitish membrane that
makes the droppings easy to carry. Similar types of membranes have been
reported by Kilham (1962) for the Flicker and by Sielmann (1958) for the
European Black Woodpecker (Dryocopus martius). In contrast the Sap-
sucker and Hairy Woodpecker have either a thin or no membrane around
the feces (Lawrence 1967). The Jamaican Woodpecker never used a special
disposal station, but each time flew in a different direction from the nest.
This is similar to what Lawrence found for the Flicker but differs from the
Sapsucker, which consistently deposited the excreta at a particular spot.
Lawrence also reported that the woodpeckers she studied often ate the fecal
material rather than removing it, but I noted no such behavior in the Jamaican
birds.

PosT-NESTLING PERIOD.-During the first 2 days after emergence from the
nesting hole, the young made no attempt to reenter the hole at dusk, which
the adult male now used for roosting. Skutch (1969) noted that in M. rubri-
capillus the parents repulsed the fledglings that tried to enter the roosting hole



1977 CRUZ: JAMAICAN WOODPECKER 201

with them. One juvenile Jamaican Woodpecker roosted in an abandoned
woodpecker hole after the 6th day of fledgling life. I feel that juvenile Jamai-
can Woodpeckers probably use abandoned holes until they are sufficiently
strong to exeavate their own holes. The juvenile woodpeckers are approxi-
mately the same size as the adults, but their excavating tool (the bill) is 22%
shorter (X billlength = immature 21.5 mm, adult 27.5 mm). In Worthy Park
there appeared t6 be enough abandoned woodpecker holes to afford the juve-
nile woodpeckers plenty of shelter. In other parts of Jamaica, such as the dry
limestone forest and mangrove woodland, fewer holes were present, and prob-
ably a longer period must elapse·before the young birds find proper shelter.

The two fledglings of the guango tree pair (1969) remained within 40 m
of the nest tree during the first 2 days after leaving the nest. During this time
they did not attempt to follow the parents, but kept in contact by means
of location calls. Whenever the parents obtained food, thsy uttered a loca-
tion call, the young answered, and the parents flew toward them. As the
fledglings detected the parents approaching with food, their begging calls
intensified. During this time the young made feeble attempts at pecking and
probing, but apparently found no food. On the 3rd day the young began fol-
lowing the parents. Although the family more or less traveled together,
one of the young followed the male, and the other the female. The juveniles
did not travel as fast as the adults and often lagged behind. Once after 4
minutes of pecking into a dead logwood branch, the female removed a bu-
prestid larva about 2.5 cm long. She began uttering location calls, and one
of the young arrived within less than a minute. The adults fed the young
both animal and vegetable matter.

During the first 2 weeks after emergence, the young of the guango tree
pair relied upon the adults almost esclusively for their food. Although the
young tried to obtain food during this period, they were apparently un-
successful. On the 15th day after leaving the nest I saw the first successful
food capture by a young bird when it removed a grasshopper-like insect from
a bromeliad. After this the young fed more and more independently, but the
parents still continued to feed them.

On the 21st day, I noted the first repulsion of a juvenile by the parents.
While the adult female was feeding on Ficus fruits, the juvenile male ap-
proached her, and she made a Stabbing lunge at him, causing him to fly off.
After this day repulsions of the juveniles by the parents occurred more fre-
quently, but they still continued feeding the young until the 24th day. After
the 24th day, although the family group still remained together, no more
feedings were detected. This particular family group (Guango Tree pair)
remained together until I left Jamaica on 17 August 1969 (28th post-nestling
day). Another family group first seen 19 June 1970 was still together when I
left on 28 July 1970. Thus the young remain with the parents for at least a
month. Kilham (1961) noted that in M. carolinus the young remained with



202 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

the parents for nearly 21k months. He did not specify whether or not the
young were completely dependent upon the parents. Lawrence (1967) found
that for the Hairy, Downy, Flicker, and Sapsucker the period of dependence
of the fledglings ranged from 1 to 3 weeks.

LITERATURE CITED

Adams, C. D. 1972. The flowering plants of Jamaica. Glasgow, Scotland, The University Press.
848 p.

Asprey, G. F.,and R. G. R6bbins. i953 The vegetation of Jamaica. Ecol. M6nogr. 23:3592412.
Bock, C. E. 1970. Theecology and behavior of the Lewis Woodpecker (Asyndesmus lewis). Univ.

Calif. Publ. Zool. 92:1-91.
. 1971. Relations between Lewis' and Red-headed Woodpeckers in sotitheasten1

Colorado. Wilson Bull. 83:237-248.
Rock, W., and W. DeW. Miller. 1959. The scansorial foot of woodpeckers, with comments on the

evolution of perching and climbing feet in birds. Amer. Mus. Nov. 1931:145.
Bond, J· 1971. The birds of the West Indies. Boston, Mass., Houghton Mifflin Co. 256 p.
Brewer, R. 1963. Ecological and reproductive relationships of Blackcapped and Carolina Chicka-

dees. Auk 80:9-47.
Brown, J· L. 1964. The evolution of diversity in avian territorial sy.stems. Wilson Bull. 76:160-169.
Burt, W. H. 1930. Adaptive modifications in the woodpeckers. Univ. Calif. Pul,1. Zool. 30:455-

524.
1943. Territoriality and range concepts as applied to mammals. J  Mammal. 24:346-

352.
Crowell, K. L. 1961. The effects of reduced competition in birds. Proc. Nat. Acad. Sci. 47:

240-243.
1962. Reduced interspecific competition among the birds of Bermuda  Ecology

43:75-88.
Cruz, A. 1974. Feeding assemblages of Jamaican birds. Condor 76:102-119.
Darwin, C. 1859. On the origin of species by means of natural selection. London, John Murray.

2 Vols.
Davis, J. 1965. Natural history, variation, and distribution of Strickland's Woodpecker. Auk

82:537-590.
Dennis, j. V. ·1969. The Yellow-shafted Flicker (Colaptes auratus) on Nantucket Island, Mas-

sachuketts. Bird-Banding 40:290-308.
1971.Species using Red-cockaded Woodpeeker holes in northeastern Sotith Carolina.

Bird-Banding 42:79-87.
Dixon, K. L. 1961. Habitat distribution and niche relati6nships in North American species of

Parus. pp  179-216, IN W. F. Blair, ed., Vertebrate Speciation. Austin, Univ. of Texas
Press.

Emlen, J. M. 1966. The role of time and energy in food preference. Amer. Natur. 100:611-617.
Emlen, J. T. 1957. Defended area? A critique of the territory concept and of conventional

thinking, Ibis 99:352.
Gibb, J. 1954. Feeding ecology of-tits, with notes on tree creeper and goldcrest. Ibis 96:513-543.
Grant, P. R. 1968. Bill size,.body size, and the ecological adaptions of bird species to competitive

situations on islands. Syst. Zool. 17:319-333.
Haftorn, S. 1956. Contribution to the food biology of tit% especially about storing of surplus

food. Pt. Iv. Kgl. Norske Vidensk. Selsk. Shrifter. No. 4.
Hamilton, T. H., and I. Rubinoff. 1963. Isolation, endemism, and multiplication of species in the

Darwin finches. Evolution 17:388-403.
Hinde, R. A. 1956. The biological significance of the territories of birds. Ibis 98:340-369.
Howard, H. E. 1920. Territory in bird life. L.6ndon, William Collins and Sons, Ltd.
Hutchinson, G. E. 1957. Concluding remarks. Cold Springs Harb6r Symp. Quant. Biol. 22:415-

427.



1977 CRUZ: JAMAICAN WOODPECKER 203

Jackson, J. A. 1970. Fredation of a Black Rat Snake on Yellow-shafted Flicker nestlings. Wilson
Bull. 82:329-330.

janzen, D. H. 1969. Birds and the ant-aeacia interaction in Central America, with notes on
birds and other myrinecophytes. Condor 71:240-256.

Kilham, L. 1957. Pilot Black Snake and nesting Pileated Woodpeckers. Wilson Billi. 71:191.
1958. Pair formation, mutual tapping, and ne%t hole selection of Red-bellied Wood-

peckers. Auk 75:318-329.
1959. Behavior and methods of communication of pileated woodpeekers. Condor

61:377-387.
1960. Courtship and territorial behavior of Hairy Woodpeckers. Auk 77:259-270.

. 1961. Reproductive behavior of Red-bellied Woodpeckers. Wilson Bull. 73:237-254.
1963. Food storing of Red-bellied Woodpeckers. Wilson Bull. 75:227-234.
1966. Reproductive behavior of Hairy Woodpeckers. I. Pair formation and court-

ship. Wilson Bull. 78:251-265.
1968. Reproductive behavior of Hairy Woodpeckers. II. Nesting and habitat. Wil-

son Bull. 80:286-305.
1969. Reproductive behavior of Hairy Woodpeckers. III. Agonistic behavior in re-

lation to courtship territory. Wilson Bull. 81:169-183.
Lawrence, L. de K. 1967. A comparative life-history study of four species of woodpeckers.

Ornithol. Mono. No. 5, Am. Ornithologists' Union, Lawrence, Kansas. 156 p.
Ligon, J. D. 1968. Sexual differences in foraging behavior in two species of Dendrocopos wood-

peckers. Auk 85:203-215.
1970. Behavior and breeding biology of the Red-cockaded Woodpecker. Auk 87:

255-278.
Lahrl, H. 1956. Der Star als Bruthohlenkonkurrent. Die Vogewelt 77:47-50.
MacArthur, R. H. 1958. Population ecology of some warblers of northeastern coniferous forests.

Ecology 39:599-619.
, J. M. Diamond, and J. R. Karr. 1972. Density compensation in island faunas. Ecology

53:330-342.
and J. W. MacArthur. 1961. On bird species diversity, Ecology 42:594-598.

, and E. Pianka. 1966. On optimal use of a patchy environment. Natl. Aead. Sci.
Proc. 51:1207-1210.

, H. Recher, and M. Cody. 1966. On the relation 1)etween habitat selection and species
diversity. Amer. Natur. 100:319-327.

and E. 0. Wilson. 1963. An equilibrium theory of island zoogeography. Evolution 17:
373-387.

, and E. 0. Wilson. 1967. The theory of island bio-geography. Mono. Papul. Biol. No. 1,
Princeton Univ. Press, Princeton, N. J. 203 p.

Meyer ae Schauensee, R. 1964. The birds of Colombia. Narberth, Pa., Livingston Pub. Co.
430 p.

Morse, D. H. 1968. A quantitative study of foraging of male and female spruce-woods warblers.
Ecology 49:779-784.

Murray, B. G. 1971. The ecological consequence of interspecific territorial behavior in birds.
Ecology 52:414-423.

Newton, I. 1967. The adaptive radiation and feeding ecology of some British finches. Ibis
190:33-98.

Nice, M. M. 1941. The role of territory in bird life. Amer. Midl. Natur. 26:441-487.
Nolan, V., Jr. 1959. Pileated Woodpecker attacks Pilot Black Snake at tree cavity. Wilson Bull:

71:381-382.
Odum, E. P., and E. J  Kuenzler. 1955. Measurement of territory and home range size in

birds. Auk 72:128-137.
Odum, H. T., and R. F. Pigeon (eds). 1970. A tropical rainforest. A study of irradiation and

ecology at El Verde; Puerto Rico. Natl. Tech. Info. Serv., Springfield, Va., 1678 p.
Orians, G. H., and M. F. Willson. 1964. Interspecific territories of birds. Ecology 45:736-745.
Perkins, L. G. 1970. Woodpeckers. Gosse Bird Club, Broadsheet 15: 14-16.
Peters, J. L. 1948. Check-list of the birds of the world. Cambridge, Mass., Harvard Univ.

Press, Vol. 6,259 p.



204 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 4

Preston, F. W. 1962. The canonical distribution of commonness and rarity: Part 1. Ecology 43:
185-215; Part II. Ibid. 43:410-432.

Pynn6nen, A. 1939. Beitrage zur kenntnis der hiologie Finisher Spechte. Pt. 1. Ann. Zool.
Soc. Zoologicae-Botanicae-Fennicae Vanamo 7:1-66.

Ridgway, R. 1902-16. The birds of North and Middle America. Parts 2-7. U. S. Nati. Mus. Bull.
No. 50.

Salt, G. W. 1957. An analysis of the avifaunas iii the Tetoh Mountains and Jacksoit Hole,
Wyoming. Condor 59:373-393.

Schoener, T. W. 1967. The ecological significance of sexual dimorphism in size iii the lizard,
Anolis conspersus. Science 155:474-477.

Selander, R. K. 1966. Sexual dimorphism and differential niche utilization in birds. Condor
68:113-151.

, and D. R. Giller. 1959. Interspecific relations of woodpeckers in Texas. Wilson
Bull. 71:107-124.

Sielmann, H. 1959. My year with the woodpeckers. London, Barrie and Rockcliffe, 139 p
Skutch, A. 1945. The most hospitable tree. Sci. Monthly 60:5-17.

1948. Life history 6f the Golden-naped Woodpecker. Auk 65:225-260.
1969. Life histories of Central Ainericaii birds, Part III. Cooper Ornithol. Soc.,.580 p

Snow; D., and B. K. Snow. 1964. Breeding seasons and annual cycles of Trinidad landbirds.
Zoologica 49:1-40.

Spring, L. W. 1965. Climbing and perching adaptations in some North American woodpeckers.
Condor 67:457488.

Stickel, D. W. 1962. Predation on Red-hellied Woodpecker nestlings hy a Black Rat Snake.
Auk 79:118-119.

1965. Territorial and breeding habits of Red-bellied Woodpeckers. Amer. Midi.
Natur. 74:110-118.

Stolpe, M. 1932. Physiologisch-anatomishe Untersitchungen fiher die hintere Extremitat der
V6gel. J. F. Ornith. 80:161-247.

Wallace, A. R. 1880. Island life. London, Macmillan.
Williams, C. B. 1964. Patterns in the balance of nature and related problems in quantitative

ecology. New York, Academic Press.
Williamson, P. 1971. Feeding ecology of the Red-eyed Vireo (Vireo olicacnis) and associated

foliage-gleaning birds. Ecol. Monogr. 41: 129-152.
Woodwell, G. M., and R. H. Whittaker. 1968. Primary production in terrestrial ecosystems.

Amer. Zool.-8:19-30.
Woolfenden, G. E., and S. A. Rohwer. 1969. Breeding birds in a Florida suburb. Bull. Florida

State Mus., Biol. Sci. 13(1):1-83.
Zusi, R. C. 1969. Ecology and adaptations of the Trembler on the island of Dominica. The

Living Bird 8:137-164.



Contributigns to'the BULLETIN OF THE FLORIDA STATE MUSEUM, BIOLOGICAL SCI-
ENCES SERIES, may be in any field of biology. Manuscripts dealing with natural history of
systematic problems involving the southeastern United States or the New World tropics are
solicited especially. Manuscripts should be of medium length-circa 35 to 200 pages (10,500-
16,000 words). Examination for suitability is made by an Editorial Board.

The BULLETIN is distributed Worldwide through institutional subscriptions and exchanges.
It is considered the responsibility of the author to distribute his paper to all interested individ-
uals. To aid in this the author(s) receive(s) 50 copies free, and he may purchase additional
separates at cost if ordered when page proof is returned. The author is also responsible for any
charges incurred for alterations made by him on galley or page proofs. The Museum will send
an invoice to the author for this amount upon completion of publication.

PREPARATION OF MANUSCRIPT

Contributors should consult recent numbers of the BULLETIN for preferred style and format.
Highly recommended ara guide is the CBE Style Manual, 3rd Edition, 1972 (Washington, D.C.,
Amen Inst. Biol. Sci.)

MSS must be submitted in duplicate (please no onionskin) and satisfy the following minimal re-
quirements: They should be typewritten, double-spaced (especiall!/ tables, figure captions, and
"Literature Cited"), on one side of numbered sheets of standard (8-1/2 x 11 in.) bond paper,
with at least one-inch margins all around. Figure legends and tables should be typed on separate
sheets. All illustrations are referred to as figures. They must comply with the following stan-
dards: Photographs should be sharp, with good contrast, and printed on glossy paper. Drawings
should be made with dense black waterproof ink on quality paper or illustration board. All il-
lustrations should have a cover sheet. All lettering will be medium weight, sans-serif type (e.g.
Futura Medium, News Gothic) in cutout, dry transfer, or lettering guide letters. Make allow-
ance so that after reduction no lowercase letter will be less than 1 mm high (2 mm is preferred)
nor any capital letter greater than 5 mm high. The maximum size for illustrations is 8-5/8 x 14 in.
(twice typepage size); illustrations should not be less than typepage width (4-5/16 in.). Designate
the top 6f each illustration and identify on the back with soft pedcil by author's name, MS title,
and figure number.

All manuscripts not submitted in BULLETIN format will be returned to the author for retyping.

Manuscripts and all editorial matters should be addressed to:
Managing Editor of the BULLETIN
Florida State Museum
Museum Road
University of Florida
Gainesville, FL 32611