Front Cover
 Front Matter

Group Title: Bulletin of the Florida Museum of Natural History
Title: Methods of assessing health and diet of Florida panthers (Puma concolor) using museum specimens /
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00101262/00001
 Material Information
Title: Methods of assessing health and diet of Florida panthers (Puma concolor) using museum specimens /
Physical Description: p. 73-108 : ill. ; 28 cm.
Language: English
Creator: Wilkins, Laurie
Allen, Julie M.
Coltrain, Joan
Flanagin, Shelly
Allen, Terry D.
Reed, David L.
Publisher: Florida Museum of Natural History, University of Florida
Place of Publication: Gainesville, FL
Publication Date: 2007
Copyright Date: 2007
Subject: Florida panther -- Nutrition   ( lcsh )
Florida panther -- Ecology   ( lcsh )
Puma -- Ecology -- Florida   ( lcsh )
Puma -- Nutrition -- Florida   ( lcsh )
Endangered species -- Ecology -- Florida   ( lcsh )
Endangered species -- Nutrition -- Florida   ( lcsh )
Genre: bibliography   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
non-fiction   ( marcgt )
Statement of Responsibility: Laurie Wilkins...et al.
Bibliography: Includes bibliographical references.
General Note: Cover title.
General Note: Bulletin of the Florida Museum of Natural History ; vol. 47, no. 3, pp. 73-108
 Record Information
Bibliographic ID: UF00101262
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 247965644
issn - 0071-6154 ;


This item has the following downloads:

PDF ( 7 MBs ) ( PDF )

Table of Contents
    Front Cover
        Front Cover
    Front Matter
        Front Matter
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
Full Text




PART I: Osteology as a Means of Assessing Florida Panther Health
Laurie Wilkins, Julie M. Allen, Joan Coltrain, Shelly Flanagin, Terry D. Allen,
& David L. Reed'

PART II. Stable Isotope Geochemistry: A Method to Evaluate the Diet
of Florida Panthers (Puma concolor) Using Museum Specimens
Julie M. Allen, Joan Coltrain, Laurie Wilkins, Shelly Flanagin, & David L. Reed

Vol. 47, No. 3, pp. 73-108




WILKINS, ALLEN and REED: Florida Panthers

The FLORIDA MUSEUM OF NATURAL HISTORY is Florida's state museum of natural history, dedicated to
understanding, preserving, and interpreting biological diversity and cultural heritage.

that publishes the results of original research in zoology, botany, paleontology, and archaeology. Address all inquiries
to the Managing Editor of the Bulletin. Numbers of the Bulletin are published at irregular intervals. Specific volumes
are not necessarily completed in any one year. The end of a volume will be noted at the foot of the first page of the
last issue in that volume.

Richard Franz, Managing Editor
Erika Simons, Production

Bulletin Committee
Richard Franz, C lhujl i % ,n
Ann Cordell
Sarah Fazenbaker
Richard Hulbert
William Marquardt
Larry Page
Tom Webber
Irvy R. Quitmyer
David Steadman, Ex officio Member

ISSN: 0071-6154

Publication Date: November 30, 2007

Send communications concerning purchase or exchange
of the publication and manuscript queries to:

Managing Editor of the BULLETIN
Florida Museum of Natural History
University of Florida
PO Box 117800
Gainesville, FL 32611-7800 U.S.A.
Phone: 352-392-1721
Fax: 352-846-0287
e-mail: drfranz@flmnh.ufl.edu



PART I: Osteology as a Means of Assessing Florida Panther Health
Laurie Wilkins, Julie M. Allen, Joan Coltrain, Shelly Flanagin, Terry D. Allen,
& David L. Reed'

PART II. Stable Isotope Geochemistry: A Method to Evaluate the Diet
of Florida Panthers (Puma concolor)
Julie M. Allen, Joan Coltrain, Laurie Wilkins, Shelly Flanagin, & David L. Reed

Key Words: methods, museum specimens, osteology, health, stable isotopes, diet, Florida panther,
Puma concolor


Preface ................ ..................... .................. 74
Part I. Osteology and Panther Health........................ 75
Abstract.................................. 75
Introduction .............................................. 75
M ethods........................... .......... ... .......... .. 77
Results and Discussion............................... 78
C conclusions ....................... ... ..... ..... .... 89
Acknowledgements...... .............. 90
Literature C ited........................... .............. 91
Part II. Stable Isotopes and Panther Diet................... 99
Abstract.................................. 99
Introduction .............................................. 99
M ethods........................... .......... ... .......... .. 99
Results and Discussion.................. ............ 102
C conclusions ...................... ......... ..... .... 106
Acknowledgements............................... 106
Literature C ited .............................. ............ 106
Epilogue ................................................................... 108

Wilkins, L, J. M. Allen, J. Coltrain, S. Flanagin, T. D. Allen, and D. L. Reed. 2007. Methods of assessing health and diet of
Florida panthers (Puma concolor) using museum specimens. Part I. Osteology as a method of assessing Florida panther
health. Bull. Florida Museum of Nat. Hist. 47(3), Pt. I: 73-97

Allen, J.M., J. Coltrain, L. Wilkins, S. Flanagin, and D.L. Reed. 2007. Methods of assessing health and diet of Florida panthers
(Puma concolor) using museum specimens. Part II. Stable isotope geochemistry: A method to evaluate the diet of Florida
panthers. Bull. Florida Museum of Nat. Hist. 47(3), PT. II: 98-108




Natural history museum specimens provide a valuable resource to examine aspects of the biology of the living animal long after
its death, which is particularly useful for large carnivores that are difficult to study or may be endangered. Voucher specimens
record changes in animal populations over time and increase in value as new data, new technologies, or new research applications
emerge. For example, animal pelts may now be analyzed for the environmental toxins they store, and skeletons provide a
permanent historic record of diet, illness, trauma, or past injury. These in turn reflect physical and physiological processes or
environmental conditions affecting the population, and they may be important indicators of dietary deficiencies, environmental
toxins, or inbreeding. Knowledge of such processes contributes to management and conservation decisions. Likewise,
museum specimens collected over time may show the positive effects of management activities on threatened populations.
The Florida Museum of Natural History (FLMNH) collection of Florida panther specimens (Puma concolor, previously
coryi) acquired through a cooperative salvage program with the Florida Fish and Wildlife Conservation Commission (FWC),
represents a unique and valuable assemblage of the large, endangered carnivore. Over 140 specimens of skins, skulls, and
skeletons exist from 1950 to the present, and they include every panther that has been recovered since the early 1980s the
beginning of the FWC capture and monitoring program (Belden et al. 1988). These provide a permanent historical record of
morphologic, genetic, and demographic changes in the panther population over the past 25 years. Specimens record biomedical
conditions that existed but were not necessarily detected in vivo by researchers. More importantly, the collection covers the
time period before and after genetic intervention and habitat management strategies began for this endangered species.
Previous studies of museum specimens facilitated the description of phenotypic traits unique to the Florida population of
puma, which was relatively unknown until a surviving population was discovered more than 25 years ago in south Florida
(Belden 1978). Among these unique characteristics are the highly inflated nasals of the cranium, or "roman nose," that had been
described by Young (1946) as a distinctive feature of the Florida population, and which later distinguished the panthers of the
Big Cypress and western Florida from the genetically distinct Everglades population (O'Brien et al. 1990; Wilkins et al. 1997). In
addition, museum specimens dating back to the late-1800s revealed a high frequency of kinked tails and cowlicks (whorls of hair)
found in the mid-dorsal region (Belden 1986; O'Brien et al. 1990; Wilkins et al. 1997). These anomalies have now become
accepted indicators of the highly inbred condition of the original Florida panther population, which together with a high
incidence of cryptorchid males, congenital heart defects, and elevated pathogen/parasite loads revealed a highly stressed and
at-risk population (Roelke et al. 1993). With the onset of the genetic out-breeding program in 1995, the frequency of kinks,
cowlicks and the more deleterious expressions of inbreeding in Florida panthers have decreased, but continue to be carefully
monitored and investigated (Land et al. 2005, Pimm et al. 2006, M. Roelke, pers.comm.).
Along with increased health risks due to inbreeding, there is evidence that panther health has varied through time among
regions in Florida, possibly due to diet and prey density (Roelke 1990). Likewise, past morphological studies have shown that
Florida panthers exhibit a greater-than-expected incidence of skeletal trauma, dental fracture, developmental abnormalities, and
osteopathologies when compared to a wild, presumably healthy population of puma from other regions of the country (Duckler
& Van Valkenburgh 1998a,b). External skeletal pathologies include arthritis, infection, trauma, and bone lesions of unknown
cause. A single internal morphological feature prominent in that analysis was the presence of Harris Lines (HLs), also known as
growth arrest lines, which are visible as transverse lines of bone deposition in radiographs of long bones. Generally, HLs can
occur in long bones of humans and other mammals, and they are believed to represent renewed growth after a period of
interruption (Park 1964; Wing & Brown 1979). Their appearance has been linked to nutritional deficiencies, infectious agents,
and/or environmental toxins, all of which are documented threats to panther survival (Roelke 1990; Roelke et al. 1993a,b;
Facemire & Guillette 1994; Facemire et al. 1995, Pimm et al. 2006).
Herein we present results of two studies of the FLMNH's collection of Florida panthers. In Part I, we expand earlier
investigations by Duckler and Van Valkenburgh (1998a,b) examining osteopathologies and HLs as health indicators. In Part II
we analyze stable carbon and nitrogen isotope ratios in Florida panther bone collagen to investigate dietary trends. The sample
includes specimens from a range of locations and time periods, and some animals born after the out-breeding initiative in 1995,
in which Texas females were introduced into Florida to supplement the existing Florida panther population. Citations for
references in the Preface are found in the Literature Cited section of Part I.

- Laurie Wilkins, Julie M. Allen, and David L. Reed, Florida Museum of Natural History.

WILKINS, ALLEN and REED: Florida Panthers


Laurie Wilkins', Julie M. Allen' 2, Joan Coltrain3, Shelly Flanagin', Terry D. Allen4,
and David L. Reed1

1Florida Museum of Natural History, University of Florida, P.O. Box 118700, Gainesville, Florida, 32611
(lauriew@flmnh.ufl.edu, shelly03@ufl.edu, dreed@flmnh.ufl.edu).
2Department of Zoology, 223 Bartram Hall, University of Florida, Gainesville, Florida 32611 (juliema@ufl.edu)
3/Department of Anthropology, University of Utah, 270 S. 1400 E., Salt Lake City, Utah 84102
4/Department of Sociology, University of Utah, 270 S. 1400 E., Salt Lake City, Utah 84102
(terry. allen@ soc.utah. edu).


Conservation efforts to reverse the negative effects of inbreeding in an isolated population of Florida panthers (Puma concolor
coryi) resulted in the release of eight Texas females into Florida in 1995 (Seal 1994; Johnson et al. 1998). Since that time, Florida
panthers have shown increased productivity, range expansion, and the reversal of a suite of deleterious morphological and
physiological effects of inbreeding. (Land et al. 2005, Pimm et al. 2006). Previously described bone pathologies in the Florida
panther may result from a compromised immune system due to inbreeding, poor health related to diet and nutrition, or the
presence of previously undetected pathogenic diseases. We examine the current collection of 140 post-cranial skeletons to
determine the frequency of trauma, infection, arthritis, and incidence of Harris Lines. Harris lines, visible from X-rays of long
bones, represent a cessation of growth due to a major episode of starvation or illness. We compare the population born before
and after 1995 to examine changes over a time line that includes genetic, biomedical, and management interventions. Our data
support earlier findings that there are idiopathic bone pathologies that exist in the Florida population, and we explore possible
causes. Multivariate analysis reveals that Harris Lines and osteopathologies increase with age, and those pathologies affect
males more than females, and both show increases after two years of age. There is a reduction in the number and severity of
osteopathologies in panthers born after 1995; however, the demographics of our population (as represented in the museum
sample) have shifted from an "aged" population to one that includes a disproportionately large number of young animals (<2
years old). It is likely that more than one biological process is operating to produce this result, and the study of osteological
material alone cannot provide definitive diagnoses. Advanced studies in the pathology of human arthritis offer intriguing
insights, and we expect at least some of their findings to have application in wildlife disease studies. Our results, together with
the rich resource of archival material, leads to new cooperative research opportunities between museums, wildlife biologists, and
wildlife veterinarians in the efforts to improve the conservation status of Florida panthers.

Post-mortem skeletal remains reveal a history of activ-
ity, injury or traumatic events in the life of an animal.
This is expressed by scars, malformations, unusual le-
sions, or excessive bone deposition as animals overcome
infections, disease, or injury in life. Unusual or abnor-
mal osteological features that have been described in
panthers include arthritis, evidence of infection, trauma
and bone lesions of unknown cause. Harris Lines (HLs),
internal osteological features, were also recorded in high
frequency in panthers. Duckler and Van Valkenburgh
(1998a) showed that 69% of Florida panthers (N=51)
exhibited at least one post-cranial osteopatholgy com-
pared to 46% in a sample of pumas (N=26) from other
locations. In the same study, the prevalence of HLs in
Florida panthers was significantly greater (56.9%) than

that of non-Florida pumas (27%). They conjectured that
the elevated incidence of HLs was due to more regular
episodes of poor nutrition, perhaps exacerbated by health
problems associated with inbreeding.
Harris Lines are visible in radiographs as dense
lines of bone deposition in long bones of humans and
other mammals (Fig. 1). They record episodes of ar-
rested bone growth in young individuals (Park 1964),
and have been experimentally induced in rats, rabbits,
pigs and dogs by starvation, selective nutrient deficiency,
and bacterial inoculation (Harris 1933; Wolbach 1947;
Park & Richter 1953; Platt & Stewart 1962; Mays 1995).
It was these experimental studies that linked the forma-
tion of HLs, with the occurrence of a physical stress
(Grolleau-Raoux et al. 1997). In humans, HLs have
been associated with episodes of childhood illness (for

Figure 1: X-ray of humerii of Florida panther 15 ,
UF24563) showing Harris Lines. Harris Lines are
transverse lines of bone density that represent a period
of arrested growth followed by renewed growth.

review, see Garnm et al. 1968). As a result, they have
been used extensively in historical and archeological stud-
ies to characterize the health of human populations
(Macchiarelli etal 1994; McHenry 1968; Rathbun 1987).
Current interpretations differ on the significance
of HL formation. Some believe that HLs form as stress
lines during accelerated growth spurts rather than as
indicators of past illness, trauma, or malnutrition (Alfonso
et al. 2005). The alternative explanation-that HLs rep-
resent renewed bone growth upon recovery from a se-
rious illness-remains well supported (Mays 1995), al-
though it has been shown that there is often no one-to-
one correspondence between episodes of disease or
nutritional deficiency and HL formation (Marshall 1968).
Medical monitoring of Florida panthers during the 1980s
showed cats to be in variable health declining south-
ward towards the Everglades National Park (ENP) and
the Fakahatchee Strand State Preserve (FSSP), a con-
dition associated with the type and abundance of prey
(Roelke 1990), and supported by food habit studies. Scat
analysis showed the panthers living north of Alligator
Alley were killing predominantly large prey (white-tailed


deer and feral hogs), while those panthers living in the
FSSP were taking a large number of small prey (rac-
coon and armadillo) (Roelke et al. 1986, Maehr et al.
1990). A relationship between diet, health, and pres-
ence of HLs may exist in this stressed population, but it
is complex and requires further investigation.
Studies on the incidence of osteological abnormali-
ties in a wild mammal population are rare, and frequently
involve the description of a single or few specimens.
This is often true because post-cranial skeletons were
not archived, and are not available for study. The pan-
ther specimens in the FLMNH mammal collection to-
day number more than 140 individuals, including those
salvaged through mid-2006. This provides an unparal-
leled opportunity to expand the earlier study of morphol-
ogy conducted by Duckler and Van Valkenburgh
(1998a,b) and to further explore the nature of
osteopathology in natural populations. The literature on
bone pathology in mammals is vast, but often relates to
veterinary studies on domestic animals; terminology and
the diagnosis of skeletal abnormalities from skeletal as-
semblages is difficult and often confusing.
Arthritis is defined by the presence ofosteophytes,
which are bone spicules that develop around joint mar-
gins or sites of injury and represent the body's attempt
to repair an injury. Arthritis involves proliferation of bone
rather than loss of bone as in osteoporosis. The degree
of osteophytosis in humans and animals can vary con-
siderably-from single spicules, a ridge of new growth
along an area of muscle attachment, to a massive out-
growth of bone. The presence of osteophytes is a com-
mon indicator of osteoarthritis (OA) or degenerative often
age-related joint disease (DJD), which is common in
humans (Rogers & Waldron 1995), wild mammals (Fox
1939; Greer et al. 1977) and domestic animals (Jubb et
al. 1985). Degenerative joint disease is a progressive
condition in which the articular cartilage is slowly de-
graded and the surrounding bone reacts by producing
osteophytes. The disease affects many animal species,
and is common in domestic breeds of cats. However,
the presence of osteophytes may be associated with
other conditions, as 'arthritis' is a general term that in-
cludes a broad spectrum of disorders or diseases that
has been extensively studied in humans, and is begin-
ning to emerge as a field of study in other mammals, as
we will discuss in greater detail later.
In past studies, joint disorders have been broadly
separated into two categories, non-inflammatory and in-
flammatory. Osteoarthritis (OA), or DJD, including joint
disease resulting from trauma and developmental, meta-
bolic, or dietary causes is generally considered to be
non-inflammatory (Tumbull & Cowan 1999). Factors

WILKINS, ALLEN and REED: Florida Panthers

cited in pathogenesis of DJD include genetics, abnormal
joint alignment, excessive stress, trauma, local inflam-
mation, and hormonal influences (McKeag 1992; Lane
& Buckwalter 1993; O'Connor & Brandt 1993). Inflam-
matory arthritis includes a variety of disorders that typi-
cally involve reactive bone formation and fall under the
general term spondyloarthropathy (SpA). One particu-
lar form of SpA is reactive arthritis (ReA), not to be
confused with another type of arthritis, rheumatoid ar-
thritis (RA). In humans (ReA) is mediated by a variety
of infectious organisms to which panthers may also be
susceptible. Variable manifestations of SpA include
asymmetrical, pauciarticular (involving less than five
joints), peripheral (appendicular) joint erosions and fu-
sion, and axial (spine and pelvis) joint inflammations
(Resnick & Niwayama 1988; Rothschild & Martin 1993).
The use of the term DJD is widespread in the both
the wildlife and veterinary medicine literature but the
descriptive terminology is confusing as its symptoms over-
lap with that of SpA. Non-inflammatory disorders can
often lead to secondary inflammations, and inflamma-
tory joint disorders commonly result in secondary, often
severe degenerative changes (Turnbull & Cowan 1999).
According to Rothschild et al. (1998,2001), it is the more
inflammatory form of arthritis, (namely SpA), that is the
more likely condition in wild mammal populations, in-
cluding large cats, whereas DJD is more often associ-
ated with domestic breeds and zoo animals. SpA has
been described in hyenas, bears, canids, non-human pri-
mates, elephants, large felids, and mammalian and non-
mammalian fossils (Rothschild & Rothschild 1994;
Rothschild & Woods 1989, 1991, 1992; Rothschild et al.
1993, 1998,2001). Tumbull and Cowan (1999) described
synovial joint diseases in wild cetaceans that included
both degenerative and infectious manifestations.
Medical management, genetic augmentation, and
intervention to increase prey species may have contrib-
uted to improved health of Florida panthers over the years,
which could reduce the expression of health-related skel-
etal anomalies. Our objective was to review earlier stud-
ies, investigate a larger sample of cats from a longer
time period, and quantify the observed skeletal features
to determine if there have been any changes in the fre-
quencies found by Duckler and Van Valkenburgh (1998a).
Further, using multivariate statistical methodology, we
explore the interrelationship of HLs, skeletal anomalies
observed per individual, and the degree of severity of
those anomalies, and to what extent these vary by age,
sex and habitat. Any general deviation from normal bone
development is referred to as an Abnormal Osteological
Feature (AOF), to distinguish it from the many specific
types of pathologies that exist in nature and to have a

consistent and easy reference. We hope to learn more
about the causes of HLs and AOFs, or at least the con-
ditions under which they form, utilizing a large data set
and ample life history information about individual Florida
Considerable analogy with human arthritis research
was used for insight into panther bone pathology, as oth-
ers have done (Rothschild et al. 2001; Tumbull & Cowan
1999), to reveal new directions in arthritis research, and
to search for explanations for the high frequencies of
HLs andAOPs. The study of osteological material alone
cannot provide a definitive diagnosis; therefore this is a
first step to categorize an idiopathic disease process that
may exist in the panther population of Florida.

To document frequency of HLs in Florida panthers,
we X-rayed left and right humerii and femora of 69 males
and 43 females including animals that died as recently
as 2006. HLs were scored when the density line was
perpendicular to the long axis of the bone and extended
across the entire shaft. This is a more conservative
measure as often HLs need only extend one-fourth to
one-half way across the shaft of the bone. Ultimately,
only humeral HLs were counted, since femora showed
very few. Harris Lines do not always occur symmetri-
cally, so we totaled the number of HLs present in L and
R humerii. Age, sex, year of death, and use area were
recorded for each specimen (Appendix 1). Our X-rayed
individuals overlapped with those studied by Duckler and
Van Valkenburgh, but did not completely duplicate their

Major skeletal elements of Florida panther speci-
mens in FLMNH collections including 77 males 50 fe-
males, and one unknown (n=128) were inspected and
abnormal osteological features (AOFs), including healed
fractures, arthritis, infection, unusual lesions, malforma-
tions, and unknown pathologies were tabulated (Appen-
dix 1). We gained insight into bone pathology by refer-
encing studies in archaeology and forensics (Baker and
Brothwell 1980; Buikstra & Ubelaker 1994; Rogers &
Waldron 1995), paleopathology (Rothschild & Martin
1993), wildlife and veterinary science, and contempo-
rary studies in human arthritis, as well as consultation
with forensic specialists. Our results are reported and
compared to the earlier study by Duckler and Van
Valkenburgh (1998a,b).
We examined the cranial, axial, pelvic, and appen-
dicular skeleton including the feet, and scored any ab-

normality within each. We recorded the total number of
AOFs, being conservative in our estimate. For example,
an AOF associated with two or more bones, such as
tibia-fibula or humerus-radius-ulnajoint was only counted
once regardless of the number of affected bones. If there
were two lesions on one bone, both would be counted
only if they were qualitatively different, or located in a
different region of the bone, although we could never be
certain of the relationship between any two observa-
tions because of possible systemic involvement. The
skeleton records a lifetime of insults, and there is no
way to know for certain that this scoring system ad-
equately distinguished one event from another. No frac-
tures that were related to the cause of death were re-
corded; that is broken bones without evidence of bone
An overall severity score (S) was assigned to each
specimen, ranging from Si-Mild, S2-Moderate, to S3-
Severe, depending on the nature and extent of the
anomaly. Measures of severity suggest the greater or
lesser expression of a condition or the later phase of a
disease. While neither of these may have any basis in
clinical practice (Rogers & Waldron 1995), they do pro-
vide a basis for comparison. A severity score of S3
(the most severe case) was assigned only in cases where
bones were broken or the deformation from normal state
was extreme, and potentially crippling, the latter having
been noted either in life through observations and re-
ports, or during necropsy. Since many skeletons had
more than one incident that ranged in severity from S 1
to S3, the highest single score was selected, since that
one would have the greatest potential threat to the sur-
vival of that animal.

In many cases, the ages calculated for Florida pan-
thers were estimates, so age classes were established
based on several references, including degree of fusion
of cranial sutures and epiphyseal closure, direct com-
parison with animals of known age, and estimates pro-
vided in FWC annual reports (Land et al. 2005). Ani-
mals were grouped into five age classes (Class I = <1
year, Class II = 1-2 years, Class III = 2-4 years, Class
IV = 4-10 years, and Class V = > 10 years. To gain an
understanding of the relationship of health parameters
(HLs, AOFs) to regions of Florida (north or south), we
assigned a "Use Area"(UA) to each animal based on
published reports or, in the case of panthers not radio-
collared by the FWC, the location where they were found
dead. Use areas were defined UA1) north of Interstate
75 (former SR 84 or Alligator Alley), and UA 2) south of
Interstate 75 including southern Big Cypress National


Preserve (sBCNP), Fakahatchee Strand State Preserve
(FSSP), and the Everglades National Park (ENP)(see
map in Part II). Past health and diet studies of panthers
frequently delineated three distinct regions by consider-
ing ENP a separate region, but sufficient samples did
not exist, so ENP cats were grouped into UA2.

The dependent variables in the study are total num-
ber of abnormal osteological features (AOFs), severity
index (SI) of AOFs, and HLs. The independent vari-
ables were sex, age class and use area. The model
used to test the dependent variables included all three
independent variables and all possible interactions. A
Multivariate Analysis of Variance (MANOVA) was
performed simultaneously testing all three dependent
variables against the independent variables using SAS
(SAS Institute Inc., Cary, NC). Once significance was
established with the MANOVA, individual ANOVAs
were run on each independent and dependent variable.
A student's t was used to find differences between Age
Class groups whenever significance was established for
that variable. All of the individual ANOVAs were ana-
lyzed using JMP version 5.0 Statistical software for the
Macintosh (SAS Institute Inc., Cary NC).
The relationships between the three dependent
variables (HLs, AOFs, and SI) were analyzed using re-
gression analysis. Animals with no osteopathologies
were removed because these animals also had a scor-
ing of zero for the AOF severity index. Therefore, only
animals with at least one AOF were used to examine
trends between the dependent variables. To determine
the effect on HL or AOF frequency resulting from the
introduction of Texas cats in 1995, we compared the
prevalence of HLs and AOFs of animals that were born
prior to 1995 (HL n=63, AOF n = 79), to those born
since that time (HL n=49, AOF n = 49) using ANOVA.

The MANOVA testing the differences among total
AOFs, Severity of AOFs, and total HLs on sex, age
class, and use area shows a significant multivariate ef-
fect for sex and age class. This test fails to reveal any
significant multivariate effect for use area or any inter-
actions of the three independent variables (Table 1).

The average number of HLs increases with each
Age Class (AC) for all animals combined (Fig. 2). There
is a continuous increase in number of HLs with each
successive AC and overall significant differences were
found between the ACs (ANOVA; 4,10 = 4.82, p =

WILKINS, ALLEN and REED: Florida Panthers

Table 1: MANOVA results for Florida panthers from the Florida Museum of Natural History. The MANOVA
examined three dependent variables; total AOFs, severity of AOFs, and Harris Lines.


Wilks' Lambda

Age Class
Age Class Sex
Use Area
Use Area Sex
Age Class Use Area
Age Class Use Area Sex


F statistic




0.0013). A significant increase exists from AC-I to AC-
IV as well as from AC-IV to AC-V. There is no differ-
ence between males and females with respect to HLs,
nor was there any significant difference between the
average number of HLs for animals from the north,
above SR 84, and those from the south, including sBCNP,
FSSP and the ENP (MANOVA results; Table 1).
The apparent continual accumulation of HLs
throughout life is an unexpected result. Harris Lines
develop at the metaphases during growth before the
epiphyseal plate fuses to the long bone, at which point
bone growth stops. For panthers, this would be approxi-
mately between the ages of three and four years. Be-
cause bone is remodeled over time, it is also expected
that evidence of HLs would disappear with age as bone
continuously remodels, so an older animal should have
fewer HLs rather than more. One possibility for the
continued visibility of HLs in older animals is that previ-
ously formed HLs become easier to detect as cats age,
perhaps due to a reduction in the thickness of the bone
cortex. This has not been studied, however, it is pos-
sible to measure cortical bone thickness in individuals
g 2.5 b
2.0 be
. 0 bc T


Age Class

Figure 2: Mean number of Harris Lines for Florida
panthers at each age class. Letters represent groups
that are significantly different. Sample sizes are; Age
class I (n=13), II (n=23), III (n=27), IV (n=31), V (n= 18).

that vary in age to determine if this is the case. (A.
Falsetti, C. A. Pound Human Identification Lab, pers.
comm.). If it is assumed that HLs are indeed an indica-
tion of stress in the population, then a likely explanation
is that the older animals in our study experienced greater
health-related stress at a younger age than those in the
younger age classes, a possibility also noted by Duckler
and Van Valkenburgh (1998a). This might suggest that
health or living conditions have improved since the mid-
1980s. This is consistent with information from diet and
prey studies conducted in the mid-1980s in the sBCPR
and FSSP that reported that animals in the south were
eating fewer deer and more smaller prey, and that the
general health of animals living in the Fakahatchee was
poor (Roelke et al. 1986, Maehr et al. 1990, Roelke 1990).
This time period during the 1980s also reflects a
time when deer hunting in the FSSP was legal, but later
banned, and therefore panthers and hunters may have
been competing for the same large prey species with
the result that panthers in FSSP were eating small prey
rather than deer and hog (Roelke 1990). This too is a
complex issue because many factors affect prey den-
sity. Prior to 1980s there was no intentional manage-
ment on public lands to increase prey (such as deer) for
panthers. During the 1980s, actions were taken to re-
duce access to public lands, reduce hunting pressure
and harvest on white-tailed deer and hogs, reduce the
use of hunting dogs, and protect does and fawns
(Schortemeyer et al. 1991, Beier et al. 2003). Addi-
tional factors affecting deer densities are length and in-
tensity of hydroperiod in the ENP that might affect fawn
survival and variable quality of vegetation and soil types
(Land 1991, Fleming 1994). Improvements in prey base
throughout the range of panthers would contribute to a
healthier population.
Contemporary arguments exist that HLs do not
form as a result of pathological stress or disease, but
rather may reflect accelerated growth during specific

early periods of development (Alfonso et al. 2005). Many
of the HLs observed were faint, spatially clustered and
appeared more frequently in proximal humerii close to
the epiphyseal closure. The clustered nature of the HLs
may reflect many individual stresses associated with
rapid growth during a brief period of time. It is unlikely
that this study will clarify the debate on the cause of
HLs in vertebrates at least not without further research.
However, it should be noted that serious health incidents
nearly always result in transverse lines of considerable
density, and such lines tend to remain detectable despite
subsequent bone remodeling or advanced age (Park 1964,
Maat 1984).
Dense lines in panthers frequently occur in the
humerus anywhere from slightly distal to the midpoint
up to the upper proximal portion of the limb bone. No
dense lines in the lower quadrants were observed. If
HLs are caused by pathological stress or nutritional de-
ficiencies, then those events may not occur during the
earliest months after birth, but rather after cubs become
independent and begin to disperse at approximately 14
months of age. This activity might be more stressful,
especially for males who move a greater distance than
females, who must compete with resident males in other
locations (Maehr et al. 2002). Harris Lines formed in
very young animals are more likely to disappear as the
bone is continuously remodeled during the earliest months
of growth. As pointed out by Duckler and Van
Valkenburgh (1998a), HLs are more likely to be retained
in the adult skeleton when they form close to the time of
epiphyseal closure and growth termination.
Duckler and Van Valkenburgh (1998b) also showed
that pumas throughout their range have higher incidence
of HLs compared to other extant species, including mule
deer (Odocoileus hemionus), bobcat (Lynx rufus), and
gray wolf (Canis lupus). This may be due to the stress
to the bone brought about by one or a combination of
larger size (compared to the wolves), and the reckless
nature of their hunting style, in which leaping over boul-
ders, jumping from trees, and incredible spurts of speed
are legendary (Sunquist & Sunquist 2002). However,
why Florida panthers have more HLs than puma from
other geographic locations remains uncertain. A long
history of poor health as a result of diet, as well as del-
eterious consequences of inbreeding manifested as car-
diac defects, and high pathogen/parasite loads increased
the potential for illness or disease and HL formation
(Duckler and Van Valkenburgh 1998b). This remains a
plausible explanation.
A superficial review of the life history of several
young cats does not support the conventional view that
HLs form as a result of one particular stressor or star-


vation. For example, Florida panther 22 was captured
by FWC at six months old, subsequently abandoned by
her mother, and was later recaptured by FWC in a starved
and dehydrated condition. She was rehabilitated, re-
leased, and again in 3-4 months, as a result of another
FWC capture attempt, she was separated from her
mother and was recaptured again in a debilitated and
dehydrated condition. She was brought into captivity
and not released until 2 years later (M. Roelke, pers.
comm.). In spite of two traumatic separations from her
mother, two bouts of starvation, two capture events, and
a two-year period in captivity, there is no evidence of
HLs in this individual. Her sibling, FP 23 also captured
twice had distinct HLs. Florida panther 8, who later in
life was captured and found to be underweight and in
poor health, had HLs, but they were in the middle por-
tion of the limbs, indicating that the cause of the HL
occurred when she was very young. These examples
demonstrate that it is not possible to make a direct cor-
relation between episodes of starvation and the pres-
ence of Harris lines, particularly in older individuals.
HLs reportedly form as a result of a variety of
infectious diseases, as well as protein or vitamin defi-
ciencies. There is also evidence to suggest that they
may form in humans as a result of exposure to mercury
and other heavy metals (Raber 1999). A study examin-
ing the presence of HLs as they relate to mercury levels
recorded in individual panthers is underway (M.
Cunningham, pers. comm.).
The importance of HLs as indicators of pathologi-
cal stress has gone in and out of vogue over the past 50
years. Bone formation and homeostasis is a dynamic
process, and many physiological processes affect bone
health. Earlier research (Duckler and Van Valkenburgh
1998a) has shown that prior to 1995, Florida panthers
had significantly more HLs (56%) than non-Florida popu-
lations (27%) (Table 3). We are unable to compare our
results directly to theirs since we used total HLs for
each animal specimen rather than presence per side.
In addition, we did not have the opportunity to examine
X-rays of the non-Florida sample. In our results, an
ANOVA testing for HL differences between animals
born before and after 1995 showed no significant differ-
ence (F,110= 1.31 p = 0.2554) although there is a slight
trend showing a reduction in the average number of HLs
of post-1995 animals, (1.53) compared to the pre-1995
animals (1.92). Presence of HLs in Florida panthers
has not diminished over time in any significant way and
HLs continue to be abundant in all age classes. Future
research will investigate the relationship between HLs
and variables such as overall health, mercury levels, and
specific biomedical conditions reported for panthers.

WILKINS, ALLEN and REED: Florida Panthers



Figure 3: Mean Abnormal Osteological Features (AOFs)
for female (n = 50) and male (n = 78) Florida panthers
with error bars. Males have significantly more AOFs
than females (ANOVA; F 1,126 = 4.45, p = 0.0369).

The total number ofAOFs was significant for both
sex and age class, in the MANOVA analyses (Table 1).
Males had more AOFs than females (ANOVA; F 1,126 =
4.45, p = 0.0369; Fig. 3), and older animals had more
AOFs than younger animals with the total number in-
creasing with each successive age class (ANOVA; F 4,123
12.84; p < 0.0001; Fig. 4)
There were no significant results in the interaction
between any two variables and our dependent variables
in the MANOVA. However, when ANOVAs were run
for just severity of AOFs, use area, and age class (Fig
5), a significant interaction was found (F9,116 = 2.55 p =
0.043; Fig. 5). These results suggest that a trend exists,
but the original MANOVA is not sensitive enough to
detect this. Alternatively, it could represent a type I
error, but because results are significant it is necessary

3.5- a
0 2.5 b
< 2.0
S1.5- b
0.5 c
0.0 -1 -- 1 --------f------
Age Class

Figure 4: Mean AOFs for Florida panthers for each
age class with error bars. Letters represent significant
differences between groups. Sample sizes are: Age
Class I (n=14), II (n=27), III (n=34), IV (n=35), V (n=18).

Age Class


Figure 5: Interaction plot between Florida panthers from
the two different use areas. Each point is a mean for
each age class. Panthers in age class three show a
significant difference between the two use areas
(ANOVA F9,116 = 2.55 p = 0.043)

to mention the possible trend. The severity of the AOFs
in Use Area (UA) 2 are more variable than UA 1, but
generally animals in this group also have more severe
AOFs. Severity (and occurrence) increases between
AC-II and AC-III, between ages of approximately two
to four. A nearly significant interaction was also found
between sex and age class with severity index (SI) of
AOF's (F9,115 = 2.36 p = 0.057; Fig. 6). Males and
females have a similar low SI in AC-I and AC-II, but
males increase dramatically in the severity of AOF's
compared to females in AC-III and AC-IV. This corre-
sponds to the period of time that males are dispersing
and attempting to establish a territory and home range.



< 1.5




Age Class


Figure 6: Interaction plot for female and male Florida
panthers in each age class. Females and males from
the 3rd, 4th and 5th age classes are significantly different
from each other (ANOVA F 9,115 = 2.37 p = 0.057).

,/ -- Use Area 1
S ----- Use Area2

Table 2: Number of Florida panthers in age class for
animals born before and after 1995.

Age Class Pre-1995 Post-1995 Total

Maehr et al. (2002) describe the process as beginning at
approximately 14 months of age and continuing for 7.0
months for females and 9.6 months for males with males'
efforts frequently frustrated by insufficient vacant range
or range containing no individuals of the opposite sex.
Independence and dispersal of young cats can, and prob-
ably does, increase food stress as well as encourage,
especially in males, male-male conflicts. As males dis-
perse, they also have a greater chance of injury (and
death), thus accounting for both the increase in young
animals in our collection, and increased evidence of
trauma and AOFs.
Our sample consists of animals born prior to 1995
(n=79) and those born since 1995 (n=49; Table 2). There
has been a significant reduction of both the number
(ANOVA; F1,12 = 13.53, p = 0.0003; Fig. 7) and severity
(ANOVA; F1,124 = 6.97, p = 0.0093; data not shown) of
AOFs since 1995 in a sample that includes both males
and females. There is only one animal in AC- V that
was born after 1995 (n = 1; Table 2). This could bias
these results because AOFs also increase with age class,



O 1.2-

I 0.8-


Pre 1995 Post 1995

Figure 7: Mean AOFs and error bars for Florida
panthers born before 1995 (n = 79) and after 1995 (n =
49). Panthers born after 1995 have significantly fewer
AOFs (ANOVA; F1126 = 13.53, p = 0.0003).


but when we repeat the analysis after removing AC-V,
our results remain significant for total AOFs (ANOVA;
F ,108= 9.61 p = 0.0025). This leads us to conclude that
there is a trend towards a reduction of AOFs and their
severity since 1995.
Although no correlation exists between Harris
Lines and total AOFs (Fig. 8A,B), a correlation exists
between the number of AOFs and the severity of AOFs
(Fig. 8C). This correlation is expected given that an
increase in the number of AOFs will increase the prob-
ability that at least one of them is severe.
Panthers over the age of two born in Florida over
the last 50 years, show a range of mild-to-severe, and
potentially debilitating, osteological conditions (Appen-
dix 1). These include trauma (breaks and bite wounds),
arthritis, periostitis (disease of the periostium), enthesitis
(inflammation of osseous attachment of tendons, liga-
ments, synovium) and other unidentifiable infections fre-
quently observed in the feet, forearms, and axial skel-
The overall incidence of fractures and other trauma
remain high when compared to the population of non-
Florida panthers analyzed by Duckler and Van
Valkenburgh 1998. (Table 3). The overall prevalence of
all pathologies in our sample was 60.8% in the pre-1995
sample compared to 69% reported by Duckler and Van
Valkenburgh (1998a). However, our scoring protocol
differed in that we counted number of AOFs rather than
a type and presence/absence per side. This resulted in
more incidents of AOFs in our study. We also did not
consider HLs in our final totals ofAOFs, thereby poten-
tially lowering the score. We preferred to treat HLs
separately since there is disagreement as to whether
HLs are only associated with illness or injury. Further,
we hoped to learn more about the possible relationship
of one to the other through these analyses. Regardless
of how pathologies were counted, both studies found a
greater than 50% incidence of abnormal osteological
features in the population of Florida panthers. This is
far greater than any published study of abnormal bone
development in any wildlife species, although those stud-
ies are themselves very rare. Pathologies were less
frequent in the post-1995 sample (n=49) with 18 indi-
viduals (36.7%) exhibiting one or more pathology com-
pared to 48 (60.8% pre-1995). The fraction of the post-
1995 population with pathologies is significantly lower
than the pre-1995 sample. However, the proportions of
old versus young animals in the two samples vary con-
siderably. There is a 34% increase in animals under
two years of age in the post- 1995 sample, and it has
already been demonstrated that injuries and other or
AOFs accumulate with age (Table 3).

WILKINS, ALLEN and REED: Florida Panthers

6] A y = 1 95- 0 lOx
- R2= 0 01








0 1 2 3 4 5 6 7
Total AOFs

B y = 1 90- 0 03x
R= 001

-1 0 1 2 3 4 5 6 7
Total Harris Lines

C y= 128+0 21x
R- 0 13

-- -.

0 1 2 3 4 5 6 7
Total AOFs
Figure 8: Correlations between total number of Harris
Lines (A), total AOFs (B), and severity of AOFs (C)
for Florida panthers. Points were jittered (offset) by
adding a random number between -1 and 1 to each
point so that the points don't lay on top of each other.
Animals without any AOFs were removed from the
analysis because they also had a zero Severity Index,
which would influence the regression line.

The most abundant osteological expression was
that of arthritis, which was present in 53.2% (n=42) of
the sample of animals born before 1995, but only 30.6%
(n=15) of those born after 1995. It might have been
easy to dismiss the severe arthritis we observed as age-
related, since many of the specimens acquired in the
1980s and early 1990s represent an "aged population"
(Ballou et al. 1989; Roelke 1990; Duckler & Van
Valkenburgh 1998a). However, arthritis, as the accu-
mulation of osteophytes, even in an incipient form, was

present in cats of all ages, except the very youngest
(Age Class I). Young animals (those younger than two
years of age) did not show significant arthritic lesions,
and there are considerably more young animals (53.1%)
in our post-1995 sample. Age is a significant factor to be
considered in the interpretation of these results, as in-
sults to the skeleton continue to accumulate through life.
Some types of lesions were more prevalent,
whereas others were more notable because of their se-
verity. Among the most common AOFs were inflam-
mation and osteophytosis, an inflammatory arthritis, in
forelimbs, including elbow and wrist. A 3-year old male
(FP 89), reportedly small for his age, had extensive pro-
liferation of bone around the articular surfaces of both
the left and right humerus/radius/ulna joints, which was
also observed in life as an open wound at the elbow (M.
Cunningham, pers. comm.) (Fig. 9). A similar, but even
more extreme case is that of FP16 from Dade County.
Evidence of osteoarthritis was present in the spine and
feet of this 14-year old male, however, all bones of the
forelimb showed an advanced joint disease, possibly
caused by a broken right radius/ulna, which calcified
into a large mass in healing (Fig. 10A). Deep ebumations
visible in the trochleae of both humerii and the trochlear
notch of both ulnae reflect the loss of the protective
cartilage resulting in bone-on-bone contact (Fig. 10B).
Long-term instability and/or inflammation were indicated
by the excessive bone deposition around both joints sug-
gesting a debilitating and presumably painful condition.
In addition, the distal fibulae where they attach to tibiae,
both appeared to have been deformed or possibly bro-
ken, a condition that is difficult to explain (Fig. 10C).
While these were the most extreme cases, there
were other cats with less severe inflammations, but that
also involve the olecrenon process, the attachment site
of the complex triceps brachii muscle. Often the condi-
tion was expressed in both limbs, which might suggest a
chronic front limb dysplasia creating the potential for
instability, injury and inflammation. Some of these same
animals also showed osteophyte formation at distal ra-
dius/ulna above the level of the wrist, which may be a
related condition. These cases, particularly in the young
panther FP89, where there was no obvious evidence of
a wound or injury that could result in an infection, alerted
us to the possibility that there was a predisposition to
inflammation at certain sites. This led us to scrutinize
those sites (e.g., elbow joints) where we observed the
early signs of inflammation or distortion. One Florida
panther (FP 2; UF 20777), shows areas of inflammation
at the proximal ulnae, and, both distal left and right ulnae
show evidence of injury with subsequent osteophyte for-
mation (Fig. 11). FP205 (UF26520), approximately 1-1/

Table 3. Number and per cent of individuals with AOFs and Harris Lines in 2 populations of Puma concolor over time and a comparison of two studies of
Florida panthers. Total individuals (w/combined3Osteopathologies/ AOPs)

Sample (N) Arthritis Infection Trauma Unknown Total individuals Harris Lines4 S3 severity5 # aged 2 yrs

Previous Study' Florida
Skull 53
Skeleton 51



9 (18%)

3 (12%)

1 (2%)
4 (8%)

1 (2%)
4 (8%)

3 (0.7%) 3 (0.7%)
0 4 (15%)

2 (4%)
6 (12%)

8 (2%)
5 (19%)

4 (8%)
35 (69%)

14 (3%)
12 (46%)

29 (57%)

7 (27%)

This Study2 Florida
Skull +Skeleton AOP HL



42 (53.2%)
15 (30.6%)

1 (2.04%)

12 (15.2%)
5 (10.2%)

12 (15.2%) 48 (60.8%)
3 (6.1%) 18 (36.7%)
66 (51.6%)

42 (66.7%)
29 (59.2%)
91 (81.3%)

13 (31.0%)
5 (17.2%)

15 (23.8%)
26 (53.1%)

1 Duckler and Van Valkenburgh 1998a
2The two studies used different parameters in measuring pathology and are not directly comparable; our definition for arthritis was more inclusive of minor
skeletal changes
3-4 Harris Line count in current study is not included in Osteopathology (as it is in the Duckler & Van Valkenburgh study). We do not consider HLs to be
a function of pathology, see text.
5 The highest measure of severity represents broken bones, advanced debilitating arthritis, or other extreme deviations from normal bone development; the
percentage shown
is based on the number of individuals with a severe condition among the total number of number of individual animals expressing one or more pathologies.

WILKINS, ALLEN and REED: Florida Panthers

Figure 9: Example of severe inflammatory arthritis in ca. 3-year old Florida panther (FP89, UF30064) that died of intraspecific
aggression. Left are the right and left humerii, and corresponding ulnae are seen at the right. In life an open wound was
observed at the elbow.

2 years old, shows osteophyte formations at the
metaphases (Fig. 12). Note that FP205 was not included
in the statistical analyses, because he spent time in cap-
tivity. However, it is noteworthy that this animal died of
an infection due to esophageal laceration, and the osteo-
phyte formation visible in the photograph may be a sys-
temic reaction to that infection.
A second very distinctive osteopathology worth
noting was a severely degraded, porous, and rugose dis-
tal ulna in panther FP 10 (UF23986), who was killed by
another Florida panther. The remnants of a periostitis
infection, suggested by a layer of black film that cov-
ered most of the cleaned limb bones (Fig. 13) might have
been a contributing factor. There were several young
animals that exhibited a very porous state in the distal
radius/ulna region ofthe forearm. Without knowing more
about the developmental process we cannot say this is
abnormal, only that young cats were vulnerable to os-
teophyte formation and infections in the distal ulna.
Frequencies of fractures and other trauma are high
compared to the population of non-Florida panthers and

much higher than in other vertebrates such as deer and
wolf (Duckler & Van Valkenburgh 1998a,b). Few com-
parative studies exist of traumatic, degenerative or de-
velopmental lesions among wild carnivores or any mam-
mal species. Among the few is a study of wolves (Ca-
nis lupus) and coyote (Canis latrans) in Saskatchewan
(Wobeser 1992). Wolves showed a much greater num-
ber of broken bones (22.8%), but degenerative joint dis-
ease, involving the spinal column and limb joints similar
to that reported here, was found in only seven wolves
and two coyotes. As with panther FP 16 (UF29821), the
articular cartilage of a single wolf was eroded from the
condyles of both femurs with eburnation of the underly-
ing bone and extensive periarticular osteophytes were
present. One would expect a greater number of injuries
in carnivores, such as wolves and puma, than other mam-
mals because hunting is a dangerous activity. Although
the osteopathologies of non-Florida wild living puma were
high (n=12, 46%), they were not as high as those ob-
served in the Florida population (Duckler and Van
Valkenburgh 1998a). Now, ten years later with a sample


Figure 10: All bones of forelimb show advanced joint
disease in this 14-year old male (FP16, UF29821),
probably initiated by the broken right radius/ulna (A).
Deep eburnations visible in the trochleae of both humerii
and the trochlear notch reflect the loss of the protective
cartilage, resulting in bone-on-bone contact (B). Distal
fibulae, at the attachment site of tibiae, both appeared
to have been deformed or previously broken (C).

twice that of Duckler and Van Valkenburgh's study
(1998a), we do see a decline in trauma and arthritis.
However, once again, it is necessary to consider the
proportionate difference in age classes. Following

Duckler and Van Valkenburgh (1998a,b) we have docu-
mented a high incidence of injury, arthritis and other pa-
thology associated with the skeletons of Florida pan-
thers. We are not able to attribute them to any specific
type of arthritis, since so many have been described (see
following discussion of human arthritis studies) but it is
apparent that some level of infection is operating, and it
is likely that more than one causal agent exists. In sev-
eral specimens, the degree of osteophytosis alone,
whether initially due to an injury with secondary infec-
tion, would suggest there is a tendency for a systemic
and inflammatory condition to prevail. The consistent
involvement of the enthesis of the olecrenon process
and the distal radius/ulna suggests abnormal joint align-
ment or ligamentous instability, particularly because these
were often symmetric. This may also cause inflammation.
Another consideration is that this front limb joint is more
vulnerable to injury since it bears significant stress in
many locomotor activities (see McGonagle et al. 2001
below). Thirty-eight panthers have been killed by in-
traspecific aggression (Land et al. 2005); several of these
were severely debilitated.

Before we attempt a discussion of possible causes
AOFS, we would like to present a summary of the in-
tensive research in human arthritis during the last two

WILKINS, ALLEN and REED: Florida Panthers

decades. This provides new information and intriguing
insights into the pathogenesis the spondyloarthropathies
(SpA), also referred to as arthropathies. References to
SpA in wildlife studies are rare, but do exist (Turnbull
and Cowan 1999), and recall that they have been identi-
fied in museum specimens (Rothschild and his colleagues,
op cit.). Spondyloarthpathies include a diverse suite of
related conditions in humans, including ankylosing
spondylitis, reactive arthritis, psoriatic arthritis, and in-
flammatory, bowel-disease (Benjamin & McGonagle
2001). Collectively these arthropathies are character-
ized by inflammatory arthritis, extra-articular inflamma-
tion, preceding bacterial infection, seronegativity for rheu-
matoid factor, and a strong genetic association (HLA-
B27) (Dougados et al. 1991; Calin & Taurog 1998). In
humans, an important factor associated with the sus-
ceptibility of an individual to reactive arthritis is HLA-
B27. It appears that B-27 positive individuals are af-
fected more severely, although the pathogenesis is still
not fully understood (Toivanen and Toivanen 2004).
Today SpA is commonly referred to as enthesopathy
because of the involvement of the enthesis, the insertion
site of a tendon, ligament, or articular capsule into bone.
Enthesitis, the inflammation of an enthesis, is believed to
be a unifying concept for SpA (McGonagle et al. 1998).
Numerous enthesitic arthritic lesions, some incipient and
others grotesquely abundant, were observed at the
olecrenon process and distal radius ulna of several young
panthers. Enthesitis can accompany many disorders,
including traumatic, degenerative, inflammatory, endo-
crine, and metabolic conditions. In some cases, enthesitis
represents the initial or predominant manifestation of dis-
ease (Resnick & Niwayama 1983). Furthermore, one
aspect of SpA, namely ReA, (reactive arthritis) is now
considered in humans to be a disease, triggered by a
host of infectious agents including bacteria, parasites,
and viruses (Toivanen & Toivanen 2001a,b). The in-
flammatory expression is the interaction between the
infectious agent and the host immune response (Schoen
2000). A hypothesis in the study of psoriatic arthritis is
that enthesitis arises at sites of high shear and compres-
sion forces, with the additive interaction between me-
chanical stress, microtrauma, tissue repair mechanisms,
and bacterial molecules variably leading to inflammation
(McGonagle et al. 2001).
Whether it is possible to extend human research
to panthers is as yet unknown. The tendency for front
limb inflammation at the enthesis of the humero-radius
ulna joint may be due to the biomechanical stresses of
the front limbs associated with puma hunting behavior
and occasional injury. However, it is also well known
that panthers are immuno-compromised and vulnerable

Figure 11: Areas of unusual rugosity or inflammation at
the proximal ulnae and osteophyte formation at the distal,
possibly due to injury or instability caused by dysplasia
(FP 2, UF20777).

to a variety of pathogens (bacterial, viral, and parasitic)
possibly as a result of low genetic diversity (Roelke et
al. 1993a,b; Glass et al. 1994; Rotstein et al. 2000). Pan-
thers exhibit a variety of congenital abnormalities, such
as atrial septic defect and cryptorchidism related to low
genetic variability (Beier et al. 2003; Pimm et al. 2006).
Elbow dysplasia in dogs is influenced by multifactorial
processes including genetic predisposition, and if un-
treated is known to progress to crippling osteoarthrosis
(Pool 2002). Dysplasia and/or the increased suscepti-
bility to osteological inflammation as a result of injury or
infection could be another previously unsuspected mani-
festation of panther inbreeding. Further, inflammations
associated with SpA, such as those identified in the el-


Figure 12: 1.5 year old Florida panther (FP 205,
UF26520), shows early osteophyte formation at the
metaphysis. This animal died from an infection due to
esophageal laceration.

bow joint of panthers in this population, have been de-
scribed in other wild mammals, including cetaceans, and
are considered to be debilitating and a significant mor-
tality factor if allowed to progress (Tumbull and Cowan
Veterinary involvement with the panther began in
1983. Since that time, biomedical research to further
the understanding of disease, nutrition, and reproductive
physiology has been an integral part of the Florida pan-
ther recovery efforts (Roelke 1990, Dunbar 1994).
Among the many thousands of biological samples taken
over the years including blood, urine, skin biopsy, feces,
hair, saliva, viral and bacterial culture swabs, one is of
particular interest here; namely Feline Syncytia-forming
Virus (Roelke, pers. comm.). Feline Syncytia-forming
virus (FeSFV) is common in healthy as well as sick do-
mestic cats, but has been reported in conjunction with
chronic progressive polyarthritis. Pedersen et al. (1980)
reported that in a study 20 domestic cats with CPA be-

Figure 13: Distal ulna is abnormally porous and rugose;
possible cause is a perostitis, an infection of the
periostium, suggested by a layer of black film that
covered most of the cleaned limb bones (FP 10,

tween 1.5 and 5.0 years of age, only males were af-
fected, and two distinct forms of the disease were mani-
fested. The following is their actual account:
/ hc most prevalent form was characterized
by osteopenia and periosteal new bone
formation surrounding the affected joints. The
second form was characterized by severe
subchondral marginal erosions, joint instability,
and deformities. The periosteal proliferative
forms resembled Reiter 's arthritis of man, and
the deforming type resembled human rheumatoid
arthritis. The disease began as tenosynovitis
and synovitis, with subsequent changes in the
articular cartilage and periosteal bone.
Histopathologic changes in these cats were
similar to those occurring in both chronic
Reiter 's and rheumatoid arthritis of man.
Chronic progressive polyarthritis of cats was
not caused by identifiable bacteria or

WILKINS, ALLEN and REED: Florida Panthers

mycoplama, but was etiologically linked to
feline leukemia virus (FeLV) and feline syncytia-
forming virus (FeSFV) infections. "

Pedersen et al. (1980) postulated that polyarthritis
was an uncommon manifestation of FeSFV that occurred
in predisposed male cats. Feline leukemia virus may not
have been directly involved in the disease, but may have
acted in some way to potentiate the pathogenic effects
of FeSFV. Of all panthers tested for FeSFV from 1987-
1992, 59% were found to be infected with FeSFV (origi-
nal number not disclosed; Dunbar 1994). Sixteen of
those panthers reside in the FLMNH collection and some
of the males showed signs of severe arthritis, whereas
females were less affected. Seven males (FP7, 10, 12,
13, 16, 17, and 20), demonstrated the following arthritis
or unknown infections: SI of 3 (n=2), SI of 2 (N=4), and
SI of 1(N=1), and five of seven males had three or more
episodes of arthritis and infection that we would describe
as arthritic, infectious, or having a component that would
coincide with a description ofenthesititis inflammationn in
region of enthesis). Of the eight females represented in
our collection, most showed no evidence (N=2), or mild
arthritis (N=3), moderate infection on the rear feet (N=1),
and more extreme episodes were apparent in old fe-
males (N=2, FP18 and 21, 11 and 14 years of age, re-
spectively). It is difficult to come to any conclusions
based on this small sample, but males had both more
AOFs and more severe AOFs. Tests for FeSFV were
discontinued after 1994, however, archived tissue samples
exist and it may be worthwhile to explore this medium
as a possible contributing factor to the high incidences
of AOFs in Florida panthers.
Environmental contaminants such as methylmer-
cury might further compromise the animals' ability to
resist disease (Roelke et al. 1991). As well, they may
be a contributing factor in the formation of Harris Lines
and/or other osteological pathologies. Elevated levels of
mercury have been reported in Florida panthers
(Newman et al. 2005), and the death of at least one
animal (FP27, UF 24557) from the Everglades was sus-
pected to be the result of mercury toxicosis. Mercury is
known to interfere with bone metabolism and calcium
homeostasis (Suzuki et al. 2004), although it is not known
if it exacerbates the inflammatory process. Presence of
mercury in south Florida environments may also eventu-
ally explain the high incidence of Harris Lines. Dense
metaphyseal banding has many possible causal agents,
but heavy metals including lead and mercury, can induce
physiological changes that result in increased calcium
deposition (Raber 1999). An examination of whether
Mercury plays a role in the high number of

osteopathologies in Florida panthers is warranted.

We present preliminary results of an exploration
into presence and causes of abnormal osteological fea-
tures, some of which may have a pathogenic basis, with
respect to a timeline that includes genetic out-breeding,
biomedical monitoring, and habitat and prey-base im-
provements. Our study demonstrates that HLs accu-
mulate over time throughout all age classes, but there is
no difference in the frequency of HLs between males
and females, no difference in the number of HLs ob-
served on cats living in the north versus the south, nor is
there any significant difference between pre-1995 and
post-1995 populations with respect to number of HLs.
Harris Lines remain common in Florida panthers, how-
ever their cause is enigmatic and somewhat controver-
sial. Food stress, illness, and infection cannot be elimi-
nated as possible causes for HLs. Exposure to heavy
metals, including Mercury, are known to cause HLs in
humans as well. We cannot eliminate the possibility
that at least some of the HLs are a response to acceler-
ated growth at certain times in the growth cycle. Inas-
much as HLs were also frequent in Puma from other
geographic regions, and in much higher frequencies than
in other large mammals in general. Another plausible
explanation for HLs might be high compression forces
operating on the forelimbs. As we are able to gather
further information about causes of AOFs, we may be
better able to understand HLs and their relationship to
bone pathology.
Osteopathology, and specifically joint disease, ex-
pressed through the formation of arthritic lesions in the
forelimbs, of Florida panthers appears to exist out side
the normal limits of a healthy population. Both AOF
numbers and severity increase with age, with males
showing a more dramatic increase than females, which
may correspond to the risks that dispersal places on
young animals, particularly males. The potential causes
of AOFs are multiple, complex, and probably interre-
lated. Multiple physiological inflammatory responses to
a broken bone are likely, as it presents an opportunity
for a pathogenic invasion (Woodard & Riser 1991). We
have seen that a break can create a systemic response,
with equally well developed and even grotesque arthritic
lesions on the opposite limb. Of course, a broken bone
creates instability, not only in the damaged area but on
other joints as well. The forelimb has shown a much
higher incidence of osteophytic development than other
limbs even though it is sometimes incipient and difficult
to detect. High compression forces operating at this
joint may account for the frequency of enthesitis or le-

sions observed at both the elbow and wrist joints.
Microtrauma can provide an opportunity for bacterial
involvement leading to inflammation and subsequent ar-
thritis. At least some of these osteophytic expressions
may be due to one or more of the spondeloarthropathies
described in the human medical literature. Collectively,
these arthropathies are characterized by inflammatory
arthritis, preceding bacterial infection, a strong genetic
predisposition, and a propensity for inflammation at the
sites where tendons, ligaments, and joint capsules at-
tach at the entheses (Dougados et al. 1991 Calin &
Taurog 1998). The role that FeSFV plays in the el-
evated incidents of AOFs is unknown, but provides at
least one link to the presence of polyarthritis in this popu-
There was a significant interaction effect between
Age Class and Use Area in the severity (and occur-
rence) of AOFs, with a greater severity occurring in
Use Area 2. This is consistent with earlier studies im-
plying that animals in the south were less healthy than in
the north, but many management strategies to improve
prey base have been implemented. We must consider
that we are looking at animals of different ages through
a time lens of perhaps 50 years, so it is difficult to pin-
point the specific time frame of the interaction.
The prevalence oftotalAOFs prior to 1995 (60.8%)
was significantly higher than in animals that were born
after 1995 (36.7%). Arthritis was the most common
AOF in both the pre- and post-1995 samples. There
has been a decrease in the incidence of pathology over
time. The presence of many more young animals in the
current population (relative to pre-1995 populations) may
bias the results. However, we cannot rule out, and we
can hope, that animals living today are healthier, less
genetically compromised, and therefore less likely to
develop severe pathologies. Whatever gains have been
made as a result of increased variability through out-
breeding or prey species management could be reversed
as the increase in panther numbers create greater com-
petition for habitat and prey, and offspring continue to
mate with each other.
Our initial goal of assessing the effect of out-breed-
ing on the health of Florida panthers vis-a-vis their oste-
ology was limited by sample sizes in the post-1995 higher
age classes. There was only one animal in AC-V and
nine in AC-IV of the post-1995 sample, compared to 17
and 26, respectively, in pre-1995 sample. Our sample
sizes in each group were also skewed with 79 pre-1995
and 49 post-1995. We cannot say with certainty that
the out-breeding initiative played an important role in the
improved health represented by our results as only eight
animals known to be the product of the genetic cross


between Texas and Florida cats were included in our
study. Most other panthers, with the exception of the
Everglades cats, showed at least one of the inbred pan-
ther characters of kinked tail or cowlick, and therefore
our sample included many more of the original inbreed
panther stock. Without the benefit of genetic diversity,
we are left with the same basic problems of an inbred
population: namely lowered disease resistance and re-
duced vitality (Roelke, et al. 1993a,b).
We have identified that a predisposition for joint
disease may exist in Florida panthers and offer this as-
sessment as a first step in developing a protocol to char-
acterize the nature of this disease. That bone is limited
in the way it can respond to any particular "insult" may
limit the ability to distinguish among diseases when study-
ing museum skeletons. Therefore, we recommend fol-
low-up with studies that include soft tissue analyses, his-
tology, bone density tests and radiology of joints includ-
ing the forearm and pelvis joints as the latter is often
involved in SpA.
There is new information revealed by this study
and the potential for new paths of discovery. Recom-
mendations by a panther advisory commission encour-
ages analyses to consider research in toxins, diseases,
and panther health, as well as how the prevalence of
abnormalities in panthers is correlated with, or interacts
with, genetic status (Beier et al. 2003). Our results,
together with the rich resource of archival material, leads
to new cooperative research opportunities between mu-
seums, wildlife biologists, and wildlife veterinarians in
the efforts to improve the conservation status of Florida

We would like to thank Geordie Duckler and Blair Van
Valkenburgh, University of California, Berkley, for their
advice and the use of data and photographs in a Florida
panther Symposium that gave rise to this paper. An-
thony Falsetti, forensic specialist, along with the staff
and students of the C. A. Pound Human Identification
Laboratory, University of Florida, provided invaluable
assistance with X-ray equipment and forensic informa-
tion. Mark Cunningham, Florida panther veterinarian,
and Bambi Ferree, and Darrell Land, all of the Florida
Fish and Wildlife Conservation Commission, responded
to many requests for panther historical and biomedical
information, as did Melody Roelke, National Institute of
Health. Forensic anthropologist Suzanne Abel, assisted
in our understanding of human and animal osteopathology.
Bruce Rothschild, Arthritis Center of Northeast Ohio,
and Donald Scott of the Southeastern Arthritis Center,
consulted with us on panther specimens, and Donald

WILKINS, ALLEN and REED: Florida Panthers

Scott reviewed the Osteopathology section of this manu-
script. Pre-veterinary student Christopher Gauthier, be-
gan the arduous task of documenting skeletal anoma-
lies. Photographs were taken by FLMNH photogra-
pher Jeff Gage. Anthropologist John Krigbaum advised
on the applicability Harris line analyses to Florida pan-
thers. We would like to thank three anonymous re-
viewers for their assistance in improving this manuscript.

Ackerman, B. B., Lindsey, F. G, & Hemker, T. P. 1986.
Predictive energetic model for cougars. Pp. 333-
352 in S. D. Miller and D. D. Everett, eds. Cats of
the World: Biology, Conservation, and Management.
Washington, D.C.: National Wildlife Federation.
Alfonso, M. P., J. L. Thompson, & V. G Standen. 2005.
Reevaluating Harris Lines -A Comparison between
Harris Lines and Enamel Hypoplasia. Collegium
Antropologicum, 29(2):393-408.
Baker, J., & D. Brothwell. 1980. Animal Diseases in
Archaeology, pp. 235. Academic Press, New York.
Ballou et al. 1989. Florida Panther, Felis concolor coryi,
population viability analysis and recommendations.
Captive Breeding Specialist Group, Species Sur-
vival Commission, IUCN, Apple Valley, Minnesota.
Beier, P., M. R. Vaughan, M. J. Conroy, & H. Quigley.
2003. An analysis of scientific literature related to
the Florida Panther: Final Report, Florida Fish and
Wildlife Conservation Commission, Tallahassee, FL.
Belden, R. C. 1978. Florida Panther investigation-a
1978 progress report. Pp. 123-133 in R. R. Odum
and L. Landers, eds. Proceedings of the rare and
endangered wildlife symposium. Georgia Depart-
ment of Natural Resources Technical Bulletin,
Belden, R. C. 1986. Florida Panther characteristics. Pp.
13. Recovery plan implementation, annual perfor-
mance report, Florida Game and Fresh Water Fish
Belden, R. C. 1986. Florida panther recovery plan imple-
mentation a 1983 progress report. Pp. 159-172 in
S.D. Miller and D.D. Everett, eds. Cats of the
World: Biology, Conservation and Management.
Proc. 2nd International Cat Symposium. The Na-
tional Wildlife Federation., Washington D.C.
Belden, R. C., W. B. Frankenberger, R. T. McBride, &
S. T. Schwikert. 1988. Panther Habitat use in
Southern Florida. Journal of Wildlife Management,
52 (4):660-683.
Benjamin, M., & D. McGonagle. 2001. The anatomical
basis for disease localization in seronegative
spondyloarthropathy at entheses and related sites.

Journal ofAnatomy, 199:503-526.
Buikstra, J. E., & D. H. Ubelaker eds. 1994. Standards
for data collection from human skeletal remains.
Arkansas Archeological Survey Research Series
No. 44.
Calin, A., & J. D. Taurog. 1998. The spondylarthritides,
2nd edn. Oxford: Oxford University Press.
Dougados, M., S. van der Linden, & R. Juhlin. 1991.
The European spondylarthropathy study group pre-
liminary criteria for the classification of
spondylarthropathy. Arthritis and Rheumatism,
Duckler, G L., & B. Van Valkenburgh. 1998a. Osteo-
logical corroboration of pathological stress in a
population of endangered Florida pumas (Puma
concolor coryi). Animal Conservation, 1:39-46.
Duckler, G L., & B. Van Valkenburgh. 1998b. Explor-
ing the health of late Pleistocene mammals: the use
of Harris lines. Journal of Vertebrate Paleontol-
ogy, 18(1):180-188.
Dunbar, M. R. 1994. Florida Panther Biomedical In-
vestigation 1 July 1990 30 June 1994. Final Per-
formance Report.
Facemire, C. F., & L. J. Guillette Jr. 1994. Reproduc-
tive impairment in the Florida Panther: Are con-
taminants a factor? Proceedings of the Florida
Panther Conference, Dennis B. Jordan ed. Florida
Panther Interagency Committee.
Facemire, C. F., T. S. Gross, & L. J. Guillette Jr. 1995.
Reproductive impairment in the Florida Panther:
nature or nurture? Environmental Health Perspec-
tive, 103(4):79-86.
Fleming, M. 1994. Distribution, abundance, and de-
mography of white-tailed deer in the Everglades.
Proceedings of the Florida Panther Conference, D.
B. Jordan, ed. Florida Panther Interagency Com-
Fox, H. 1939. Chronic arthritis in wild mammals. Trans-
actions of the American Philosophical Society,
Garn, S. M., F. N. Silverman, K. P. Hertzog, & C. G
Rohmann. 1968. Lines and bands of increased den-
sity. American Journal of Physical Anthropology,
Glass, C.M., R. G McLean, J. B. Katz, D. S. Maehr, C.
B. Cropp, L. J. Kirk, A. J. McKeiman & J. F.
Evermann. 1994. Isolation of pseudorabies
(Aujeszky's disease) virus from a Florida panther.
Journal ofWildlife Diseases, 30:180-184.
Greer, M., J. K. Greer, & J. Gillingham. 1977. Osteoar-
thritis in selected wild mammals. Proceedings of

the Oklahoma Academy of Science, 57:39-43.
Grolleau-Raoux, J., E. Crubezy, D. Rouge, J. Brugne &
S.R. Saunders. 1997. Harris Lines: A study of
age-associated bias in counting and interpretation.
American Journal of Physical Anthropology,
Harris, H. A. 1933. Bone Growth in Health and Dis-
ease. Pp. 46, Oxford University Press, London.
Johnson, K. A., E. D. Land & M. A. Lotz. 1998. Florida
panther genetic restoration. Annual performance
report, Florida Game and fresh Water Fish Com-
mission, Tallahassee, FL. 56pp.
Jubb, K. V., P. C. Kennedy, & N. Palmer. 1985. Pathol-
ogy of domestic animals, Pp. 574 Vol. 3. 3rd ed.
Academic Press Inc., New York, NY.
Land, D. 1991. Big Cypress deer/panther relationships:
deer mortality. Final Report. Florida Game and
Fresh Water Commission, Tallahassee, FL
Land, D., M. Cunningham, M. Lotz, & D. Shindle. 2005.
Florida Panther Annual Report 2004-2005, Bureau
of Wildlife Diversity Conservation, Pp. 77.
Lane, N. E., & J. A. Buckwalter. 1993. Exercise: a cause
of osteoarthritis? Rheumatic Disease Clinics of
NorthAmerica, 6:245-249.
Maehr, D. S., R. C. Belden, E. D. Land, & L. Wilkins.
1990. Food habits of panthers in southwestern
Florida. Journal ofWildlife Management, 54(3):420-
Maehr, D. S., E. D. Land, D. B. Shindel, 0. L. Bass, &
T. S. Hoctor. 2002. Florida panther dispersal and
conservation. Biological Conservation, 106 (2002):
Maat, G 1984. Dating and rating of Harris's lines. Ameri-
can Journal of Physical Anthropology, 63:291-299.
Macchiarelli, R., L. Bondioli, L. Censi, M. K. Hernaez,
L. Salvadeli, &A. Sperduti. 1994. Intra- and inter-
observer concordance in scoring Harris lines: atest
on bone sections and radiographs. American Jour-
nal of Physical Anthropology, 95:77-83.
Marshall, W. A. 1968. Problems in relating the presence
of transverse lines in the radius to the occurrence
of disease. Pp. 245-261 in Brothwell, ed. Skeletal
Biology of Earlier Human Populations, Society for
the Study of Human Biology. London: Academic
Press, Vol. 8.
Mays, S. 1995. The relationship between Harris lines
and other aspects of skeletal development in adults
and juveniles. Journal of Archaeological Science,
McGonagle, D., W. Gibbon, & P. Emery. 1998. Classi-
fication of inflammatory arthritis by enthesitis.
Lancet, 352:1137-40.


McGonagle D., L. Stockwin, J. Isaacs, & P. Emery.
2001. An enthesitis-based model for the pathogen-
esis of spondyloarthropathy. Additive effect of
microbial adjuvant and biomechanical factors at
disease sites. Journal of Rheumatology, 28:2155-
McHenry, H. M. 1968. Transverse lines in long bones
of prehistoric California Indians. American Journal
ofArcheological Science, 29:1-18.
McKeag, D. B. 1992. The relationship of osteoarthritis
and exercise. Clinics in Sports Medicine, 11:471-
Newman, J., E. Zillioux, E. Rich, L. Liang, & C.
Newman. 2005. Historical and other patterns of
monomethyl and inorganic mercury in the Florida
Panther (Puma concolor coryi). Archeology En-
vironmental Contaminants Toxicology, 48(1):75-80.
O'Brien, S. J., M. E. Roelke, N. Yuhki, K. W. Richards,
W. E. Johnson, W. L. Franklin, A. E. Anderson, 0.
L. Bass, R. C. Belden, & J. S. Martenson. 1990.
Genetic introgression within the Florida Panther
Felis concolor coryi. National Geographic Re-
search, 6:485-494.
O'Connor, B. L., & K. D. Brandt. 1993. Neurogenic
factors in the etiopathogenesis of osteoarthritis.
Rheumatic Disease Clinics of North America,
Park, E. 1964. The imprinting of nutritional disturbances
on the growing bone. Pediatrics, 33:815-862.
Park, E. A., &C. P. Richter. 1953. Transverse lines in
bone: the mechanism of their development. Bulle-
tin ofthe Johns Hopkins Hospital, 93:234-248.
Pedersen, N.C., R. R. Pool, & T. O'Brien. 1980. Fe-
line chronic progressive polyarthritis. American
Journal of Verterinary Research, 41(4):522-35.
Pimm, S. L., L. Dollar, & 0. L. Bass, Jr. 2006. The
genetic rescue ofthe Florida panther. Animal Con-
servation., 9(2)115-122.
Platt, B. S., & R. J. C. Stewart. 1962. Transverse tra-
beculae and osteoporosis in bones in experimental
protein-calorie deficiency. British Journal of Nu-
trition, 16:483-495.
Pool, R. R. 2002. Canine elbow dysplasia. ACVS Sym-
posium Equine and Small Animal Proceedings,
October 1, 2002.
Raber, S. A. 1999. The dense metaphyseal band sign.
Radiology, 211:773-774.
Rathbun, T. A. 1987. Health and disease at a South
Carolina plantation: 1840-1870. American Journal
of Physical Anthropology, 74:239-253.
Resnick, D., & G. Niwayama. 1983. Entheses and
Enthesopathy. Radiology, 146:1-9.

WILKINS, ALLEN and REED: Florida Panthers

Resnick, D., & G. Niwayama. 1988. Diagnosis of Bone
and Joint Disorders. Philadelphia, W. B. Saunders.
Roelke, M.E., 1986. E.R. Jacobson, GV. Kollias, &
D.J. Forrester. 1986. Medical management and bio-
medical findings on the Florida panther (Felis
concolor coryi). Annual performance report, En-
dangered Species Project E-1-10 7506. Florida
Game and Fresh Water Fish Commission. 65 pp.
Roelke, M.E. 1988. Florida Pather Biomedical Investi-
gation: health and reproduction. Annual perfor-
mance report, Endangered Species Project E-1-12
7506. Florida Game and Freshwater Fish Commis-
sion. 47 pp.
Roelke, M. E. 1990. Florida Panther biomedical investi-
gation. Final Performance Report, Florida Game
and Fresh Water Fish Commission, Tallahassee,
Roelke, M. E., D. P. Schultz, C. F. Facemire, & S. F.
Sundlof. 1991. Mercury contamination in the free-
ranging endangered Florida panther (Felis concolor
coryi). Proceedings of the American Association
of Zoo Veterinarians 1991:273-283.
Roelke, M. E., J. S. Martenson, & S. J. O'Brien. 1993a.
The consequences of demographic reduction and
genetic depletion in the endangered Florida Pan-
ther. Current Biology, 3:340-350.
Roelke, M. E., D. J. Forrester, E. R. Jacobson, G. V.
Kollias, F. W. Scott, M. C. Barr, J. R. Evermann,
& E. C. Pirtle. 1993b. Seroprevalence of infec-
tious disease agents in free-ranging Florida Pan-
thers (Felis concolor coryi). Journal of Wildlife
Diseases, 29(1):36-49.
Rogers, J., & T. Waldron. 1995. A Field Guide to Joint
Disease in Archaeology. John Wiley and Sons Ltd.,
West Sussex, Pp. 119.
Rothschild, B. M., & L. D. Martin. 1993. Paleopathology:
Disease in the Fossil Record. Boca Raton, CRC
Press, Pp. 386.
Rothschild, B. M., X. M. Wang, & R. Cifelli. 1993.
Spondyloarthropathy in Ursidae: a sexually trans-
mitted disease. National Geographic Research
Exploration, 9:282-284.
Rothschild, B. M. & C. Rothschild. 1994. No laughing
matter: spondyloarthropathy and osteoarthritis in
Hyaenidae. Journal of Zoo and Wildlife Medicine,
Rothschild, B. M., & R. J. Woods. 1989.
Spondyloarthropathy in gorillas. Seminars in Arthri-
tis and Rheumatism, 18:267-276.
Rothschild, B. M., & R. J. Woods. 1991. Reactive ero-
sive arthritis in chimpanzees. American Journal of
Primatology, 25:49-56.

Rothschild, B. M., & R. J. Woods. 1992.
Spondyloarthropathy as an Old World phenomenon.
Seminars in Arthritis and Rheumatism, 21:306-316.
Rothchild, B. M., C. Rothschild, & R. J. Woods. 1998.
Inflammatory arthritis in large cats: an expanded
spectrum of spondyloarthropathy. Journal of Zoo
and Wildlife Medicine, 29 (3):279-284.
Rothschild, B. M., C. Rothschild, & R. J. Woods. 2001.
Inflammatory arthritis in canids:
spondyloarthropathy. Journal of Zoo and Wildlife
Medicine, 32(1):58-64.
Rotstein et al. 2000. Prevalence of Bartonella henselae
antibody in Florida Panthers. Journal of Wildlife
Diseases, 36:157-160.
Schoen, R. T. 2000. Infectious arthritis and immune dys-
function. Current Opinion in Rheumatology,
Schortemeyer, J.L., D.S. Maehr, J.W. McCown, E.D.
Land & P.D. Manor. 1991. Prey management
for the Florida panther: a unique role for wildlife
managers. Transitions of the Fifty-sixth North
American Wildlife and Natural Resources Confer-
ence: 512-526.
Seal, U.S., ed. 1994. A plan for genetic restoration and
management of the Florida panther (Felis concolor
coryi). Report to the Florida Game and Fresh
Water Fish Commission. Conservation Breeding
Specialist Group, Apple Valley, MN. 22 pp.
Sunquist, M. & F. Sunquist. 2002. Wild Cats of the World,
University of Chicago Press, Chicago, Ill.
Suzuki, N. et al. 2004. Both mercury and cadmium
directly influence calcium homeostasis resulting from
the suppression of scale bone cells: the scale is a
good model for the evaluation of heavy metals in
bone metabolism. Journal of Bone and Mineral
Metabolism, 22(5):439-446.
Toivanen A., & P. Toivanen. 2001a. Reactive Arthritis.
The Israel Medical Association Journal, 3:681-
Toivanen, A., & P. Toivanen. 2001b. Reactive arthri-
tis. Current Opinion in Rheumatology, 12(4):300-
Toivanen, A. & P. Toivanen, 2004. Reactive arthritis.
Clinical Rheumatology, 18(5)689-703.
Turnbull, B. S., & D. F. Cowan. 1999. Synovial joint
disease in wild cetaceans. Journal of Wildlife Dis-
eases, 35(3):511-518.
Wilkins, L., J. M. Arias-Reveron, B. Stith, M. E.
Roelke, & R. C. Belden. 1997. The Florida Pan-
ther Puma concolor coryi: a morphological in-


vestigation of the subspecies with a comparison
to other North and South American cougars. Bul-
letin of the Florida Museum of Natural History,
Wing, E. S., & A. B. Brown. 1979. Paleonutrition.
Academic Press, New York, Pp. 202.
Wobeser, G 1992. Traumatic, degenerative, and de-
velopmental lesions in wolves and coyotes from
Saskatchewan. Journal of Wildlife Diseases,
Wolbach, J. B. 1947. Vitamin A deficiency and excess
in relation to skeletal growth. Journal of Bone
Joint Surgery, 29:171-183.
Woodard, J. Carroll & W. H. Riser. 1991. Morphol-
ogy of fracture nonunion and osteomyelitis. Vet-
erinary Clinics of North America: Small Animal
Practice, 21(4):813-844.
Young, S. P. 1946. History, life habits, economic sta-
tus and control. Pp. 1-173 in S. P. Young & E. A.
Goldman, eds. The puma, mysterious American
cat. Washington, D. C. :American Wildlife Insti-
tute, Reprinted by Dover Publications, Inc., New
York (1946).

WILKINS, ALLEN and REED: Florida Panthers

Appendix I. Florida Panthers listed by FLMNH number (UFID), Fish and Wildlife Conservation Commission number (FP#),
age class, pre or post 1995 group, use area, total number of abnormal osteological features (AOFs), severity of AOFs, specific
AOFs of Arthritis (AT), Infection (IN), Unknown Cause (UN), Gross Trauma (GT), and number of Harris Lines.

CLASS 19951 AOFs



9789 3
10424 UCFP06 3
11915 2
11927 UCFP05 2
14390 UCFP02 2
14699 R.ALLEN 3
16374 6 4
18798 3 4
18847 14 5
18944 PCO 047 3
19077 CB 17 3
19090 4
19096 1 5
20777 2 5
20957 UCFP 13 2
20958 UCFP12 4
20973 UCFP14 2
22409 7 4
22529 4 5
23849 UCFP34 2
23986 10 2
24042 Volusia Co.4
24096 13 4
24160 4
24267 8 5
24268 PCO059 4
24314 20 4
24315 25 4
24316 24 4
24557 27 3
24561 3
24563 15 4
24595 33 3
24611 35 1
24621 30 2
24644 39 3
24645 UCFP 19 1
24646 17 4
24928 18 5
24929 41 3
24931 37 4
25908 76 3
25914 84 2
25922 PCO 192 1
26083 43 3
26157 28 4
26159 29 3


AT 1



AT 1

AT 1
AT 1

AT 1
AT 1

GT 1

GT 2
GT 0


UN 0
GT 4

UN 0





CLASS 19951 AOFs


26161 22 4
26840 47 2
26841 44 3
26842 53 1
26843 50 3
26844 34 4
26845 UCFP21 1
26856 UCFP22 3
26938 UCFP23 1
26939 26 4
27148 31 5
27370 38 4
27616 12 5
27618 52 3
27700 42 4
28713 UCFP30 2
28802 58 3
28980 40 5
29199 UCFP25 3
29242 UCFP26 3
29250 68 4
29262 45 4
29263 51 4
29273 72 3
29370 46 4
29532 64 3
29566 UCFP33 1
29567 74 3
29621 36 5
29819 63 4
29821 16 5
29826 80 4
30022 UCFP35 2
30023 UCFP36 2
30064 89 3
30178 90 2
30366 UCFP39 1
30367 UCFP38 2
30374 UCFP43 3
30391 11 5
30393 23 5
30398 UCFP40 1
30399 UCFP41 2
30430 UCFP29 3
30431 97 2
30433 UCFP27 2
30434 105 4
30935 49 5
30936 UCFP46 1
30937 UCFP45 3
30938 96 2
30948 98 4
30957 111 5


GT 0


AT 1
AT 1

AT 1

AT 1
AT 1


CT 2
GT 3
CT 2
GT 1
GT 1
CT 2
GT 4
CT 0

WILKINS, ALLEN and REED: Florida Panthers

CLASS 19951 AOFs





'Designates animal born before (1) or after (2) 1995
2Panther used areas to the North or South of Alligator Alley
3Measure of Severity (1) slight (2) moderate (3) represents broken bones, advanced debilitating arthritis, or other extreme


AT 1
AT 1
AT 1
AT 1

GT 2


WILKINS, ALLEN and REED: Florida Panthers



JulieM. Allen2, Joan Coltrain3, Laurie Wilkins', Shelly Flanagin', and David L. Reed1

1Florida Museum of Natural History, University of Florida, Dickinson Hall, Museum Road and Newell
Drive, Gainesville, Florida, 32605

2Department of Zoology, 223 Bartram Hall, University of Florida, Gainesville, Florida 32611.

3Department of Anthropology, University of Utah, 270 S. 1400 E., Salt Lake City, Utah 84102.


Many of the methods used to determine the diet of large carnivores are costly, time-consuming and may not give a complete
picture of diet. Here we discuss the potential for stable-isotope analysis to provide a more complete picture of long-term intake
with a pilot study of Florida panthers. We use 20 Florida panthers (Puma concolor) from the Florida Museum of Natural History.
Our results are consistent with a diet based primarily on deer and hog as other studies have suggested. However, male and
female isotope values differ significantly suggesting distinctly different dietary intake. In addition, contrary to findings from
studies employing other techniques, no differences are apparent in the isotope signatures of animals from northern versus
southern areas. Although the isotope data presented here are preliminary, stable isotope analysis appears to be a useful tool for
assessing differences in diet across demographic parameters and will be useful in dietary studies contributing to conservation

Determining the diet of an animal is costly and time-
consuming, particularly with large carnivores that are
difficult to track. Although many of the tools used to
determine diet provide useful information, most cannot
provide a complete picture of long-term dietary pattern-
ing. For example, the analysis of scat is biased toward
indigestible dietary components. Furthermore, scat
analysis examines the most recent meal but does not
reveal seasonal or long-term dietary trends unless col-
lection occurs over a long period of time. Another com-
mon method, kill-site analysis, requires that the home
range of an animal be accurately determined then
searched until a carcass is located. This method also
tracks short-term dietary intake and is biased towards
the carcasses of larger prey taxa that are easier to lo-
cate. Finally, observational studies require significant
time depth, particularly when studying large nocturnal
carnivores, since predation events are difficult to ob-
serve. Alternatively, stable isotope analysis does not re-
quire a multi-year study, but does provide a long-term
dietary signal and can be used with other methods to
give a more complete picture of carnivore diets. Here
we present stable isotope analysis of museum speci-

mens as a potential source for information that can aid
in management strategies of wild carnivores.
Stemming from the adage, 'you are what you eat,'
the stable isotopes of carbon and nitrogen are incorpo-
rated into animal tissues in predictable ways represen-
tative of the isotopic signature of prey taxa (Hobson &
Schwarcz 1986; Kelly 2000). Because tissues are cre-
ated at different rates, an animal's isotope chemistry
can track both long- and short-term dietary patterning.
For example, hair grows much faster than bone, so ex-
amining the isotopic signature of a hair shaft will give a
more recent signature of diet than bone collagen, which
turns over very slowly providing an average of long-
term intake. Thus stable carbon and nitrogen isotope
values are commonly used to reconstruct the diets of
large carnivores, such as seals and wolves, and other
animals (Hobson et al. 1996; Szepanski et al. 1999; Urton
& Hobson 2005, Kelly 2000). These values represent
the ratio of the heavy to light isotopes for carbon and
nitrogen (13C/12C, 15N/14N) written in delta notation
(813C, 815N) as parts per million (%o) difference from
an internationally recognized standard (PDB for carbon
and atmospheric air for nitrogen). The standard is as-
signed by definition a value of 0%o and computed as


Equation 1:

13C or 615N = Rsample Rstandard x 1000 %o

where R = 13C/12C or 15N/14N

Stable carbon isotope ratios in bone collagen monitor
the ratio of carbon-13 to carbon-12 in the amino acid
sequences that make up collagen fibrils, and are repre-
sentative of diet for the following reasons. When CO2 is
taken up during photosynthesis, metabolic processes al-
ter or fractionate the ratio of 13C/12C, depleting plant
tissues in 13C relative to atmosphere (-7.7%o). The de-
gree of fractionation associated with photosynthesis co-
varies with the kinetic properties of carbon uptake and
enzymatic processes of carbon fixation (Farquhar et al.
1989). In terrestrial plants, two photosynthetic path-
ways, the C3 or C4 pathway, have distinct carbon iso-
tope signatures. Fractionation is contingent upon the
photosynthetic pathway the plant uses to metabolize at-
mospheric CO2. Cool season grasses, trees, tubers and
most bushy plants utilize the C3 photosynthetic pathway
which discriminates heavily against 13C, expressing a
mean 613C value of -26.7 2.7%o (n=370) (Cerling et
al. 1998). A small set of forbs and all warm-season
grasses such as corn (Zea mays), common to regions
where daytime growing-season temperature exceeds
220C and precipitation exceeds 25 mm (Ehleringer et al.
1997), use C4 photosynthesis resulting in less discrimi-
nation against 13C and an average 613C value of -12.5 +
1.1%o (n=455) (Cerling et al. 1998). 613C values are
passed from producer to consumer leaving a diagnostic
signature in both hard and soft tissues that does not co-
vary with the skeletal element analyzed or sex of the
sample independent of differences in feeding ecology
(Hobson & Schwarcz 1986; Lovell et al. 1986). Frac-
tionation between plants and herbivores is ca. 5%o and
isotopic enrichment between herbivores and each suc-
cessive trophic level is about 1%o.
Stable carbon isotope analysis is also used to esti-
mate reliance on marine resources in diets lacking a C4
component. Kinetic processes governing bicarbonate
(HCO3-) formation in seawater fractionate marine bi-
carbonates approximately 7%o relative to atmosphere,
placing seawater 613C values near 0%o (Chisholm et al.
1982; Tauber 1981). Because submerged marine plants
employ a C3 photosynthetic pathway and derive carbon
primarily from seawater bicarbonates, they express mean


613C values of -16 to -18%o, approximately 7%o more
positive than terrestrial C3 plants giving them a distinc-
tive marine label.

Nitrogen isotopes are also incorporated into the
food web in predictable ways. Most terrestrial plant taxa
obtain nitrogen from soil ammonium (NH4 ) or nitrate
(NO3-) and have mean 815N values of 3-6%o with a 0-
9%o range contingent upon temperature and aridity (Pate
1994). Independent of effective moisture, legumes and
some epiphytes have symbionts that fix atmospheric ni-
trogen and accordingly express mean 815N values of
1%o, with a ca. -2 to 2%o range (Evans & Ehleringer
1994; Pate 1994). Herbivores in temperate climates
typically exhibit 815N values of 6-9%o while arid-land
species and non-obligate drinkers, those that recycle urea,
reflect their water-conservation strategies in more posi-
tive 815N values (Ambrose 1991).
Stable nitrogen isotope analysis follows from the
understanding that 15N/14N increases by approximately
2-4%o with each increase in trophic level due primarily
to fractionation during urea production, enriching the iso-
tope signature of nitrogen available for protein synthesis
(Schoeller 1999). For example, a carnivore eating a diet
consisting mostly of herbivorous prey such as deer will
exhibit systematically lower nitrogen isotope ratios than
a carnivore in the same ecosystem subsisting primarily
on omnivorous animals such as pigs and raccoons.

The Florida panther (Puma concolor) is suitable
for dietary studies using stable isotope analysis for sev-
eral reasons. First, the Florida Museum of Natural His-
tory houses a large collection of panther, over 140 speci-
mens, providing an excellent resource for bone tissue
from animals representing various geographic ranges and
temporal periods. In addition, Florida panther diet has
been studied using scat and kill-site analysis providing a
comparative database. Finally, Florida panthers are en-
dangered, and an understanding of their predation strat-
egies is central to successful management of a declining
Maehr et al. (1990) demonstrated using scat and
kill site analysis that the main prey of panthers inhabiting
the public lands of southwest Florida, including Florida
Panther National Wildlife Refuge (FPNWR), Big Cy-
press National Park (BCNP), Big Cypress Seminole
Indian Refuge (BCSIR), and Fakahatchee Strand State
Preserve (FSSP), was wild hog (Sus scrofa) and white-

WILKINS, ALLEN and REED: Florida Panthers

FL Panthe
NWR 1-75

~2 "

iee Strand


Figure 1: Map of Southern Florida showing the range of the Florida panther and the northern and southern maj or use
areas. The northern use area includes Florida Panther National Wildlife Refuge (FPNWR), Big Cypress Seminole
Indian Refuge (BCSIR) not shown, and northern Big Cypress National Park (nBCNP). The southern use area
includes southern Big Cypress National Park (sBCNP), Fakahatchee Strand State Preserve (FSSP) and Everglades
National Park (ENP). Northern and southern BCNP are separated by 1-75. Map was provided by D. Land from

tailed deer (Odocoileus virginianus), followed by rac-
coon (Procyon lotor), armadillo (Dasypus
novemcinctus), and rabbit (Sylvilagus spp.) in order
of their abundance in the diet. However, frequencies of
both large and small prey varied considerably between
the northern and southern portions of BCNP (nBCNP
and sBCNP; Figure 1), which crosses a soil boundary.

The soils of nBCNP are thought to be better drained
and more fertile (Leighty et al. 1954). Here, hogs domi-
nated panther diet making up 58.7% of the biomass con-
sumed, whereas in sBCNP hogs made up merely 22.7%
(Maehr et al. 1990, Table 1). In contrast, smaller spe-
cies (raccoon, armadillo, and rabbit) were present in
higher frequencies in the diets of panthers from sBCNP


Table 1: Percent biomass of prey animals consumed by Florida panthers. Values were taken from Maehr et al.
(1990), and Dalrymple and Bass (1996) calculated from scat and kill-site analysis.

Percent Biomass consumed
Deer Small Mammals

Other Study

North Big Cypress National Park 58.7
South Big Cypress National Park 22.7
Everglades National Park 0.7


(Maehr et al. 1990). Shortemeyer (1994) reported that
south of 1-75 deer and hog densities rarely exceed one
animal per 100 acres. In Everglades National Park
(ENP), deer represented 78.4% of biomass consumed,
in part because no hogs are present, (Dalrymple and
Bass 1996) but a 67% decrease in deer densities since
1959 has been reported (Fleming et al. 1994) suggesting
that deer are more difficult to hunt in ENP. Conse-
quently, the biomass of smaller prey in ENP was 18.2%
higher than that of nBCNP (Table 1).
Veterinary studies based on physical condition, body
weight, reproductive status, and hematologic and serum
values conducted between 1986 and 1990 identified a
north/south "health cline." Panthers utilizing land north
of State Road 84, including private ranches, the FPNWR,
and the Bear Island unit of the BCNP, were in better
condition than those in the ENP or FSSP (Roelke 1990).
The panthers in sBNCP were also found to be in poorer
physical condition, have larger home ranges and lower
reproductive output than panthers in the north (Maher
et al., 1989a; Maher et al. 1990). McCown (1991) found
that panthers in the cypress-dominated southern portion
of sBCNP were smaller and less productive than in the
Bear Island unit located in nBCNP. Differences in pan-
ther diet leading to a north to south gradient of declining
health has been attributed to a shift from reliance on hog
in the north to deer and smaller prey in the south (Beier
et al. 2003). In the most southern part of ENP panther
mean body size is small which is thought to be due to
low deer densities. One danger the panthers face when
feeding on small more aquatic based prey is the poten-
tial for mercury ingestion. (Maehr et al. 1990; Iriate et
al. 1990; Roelke 1990; Dalrymple & Bass 1996).
Here we report the bone collagen stable carbon
and nitrogen isotopes ratios of 20 Florida panthers and
25 prey indivuals. Panthers were chosen across the
north/south diet cline to determine if their stable isotope
chemistry reflects the apparent dietary differences dis-
cussed above. The northern use area included FPNWR,
nBCNP and FPSIR; the southern use area was com-
prised of sBCNP, FSSP and ENP (Figure 1). Only adult


Maehr et al., 1990
Maehr et al., 1990
Dalrymple and Bass 1996

panthers were selected for study, and we chose animals
with the greatest amount of natural history information
in order to relate isotope values to known home ranges
and diets. All of the animals were radio-collared and
tracked by FWC.
Stable isotope values were also obtained from bone
collagen of the five primary prey taxa of Florida pan-
thers gathered from three sources; the FLMNH mam-
mal collection, actual prey remains in panther scats from
ENP, and samples provided by Deborah Jansen from
panther scats/kill sites in the BCNP. We sampled 10
white-tailed deer (Odocoileus virginianus), 11 raccoon
(Procyon lotor), three rabbit (Sylvilagus palustris), two
armadillo (Dasypus novemcintus), and two feral hogs
(Sus scrofa).

Approximately 500 mg of cortical bone was
cleaned of surface contaminants then demineralized
whole in 0.6N HCI at 4o C. Sterile ddH2O was used
throughout and glassware was sterilized by heating to
5500 C for 3h. The resultant collagen was then rinsed to
neutrality, treated with 5% KOHto remove organic con-
taminants, rinsed to neutrality, and lyophilized. Approxi-
mately 100 mg of lyophilized collagen was gelatinized in
5 ml of water (pH 3) for 24 hours at 1200 C. Water-
soluble and -insoluble phases were separated by filtra-
tion and the former was lyophilized and weighed to de-
termine the collagen yield. 613C and 685N values were
determined by flash combustion to produce CO, and N,
and measured against the appropriate reference gas on
a Finnigan Delta Plus mass spectrometer with Carlo Erba
EA118 CHN interface. Stable isotope measurements
and weight percent C and N values were obtained from
single sample combustion. All samples met well-estab-
lished preservation criteria (Ambrose 1990). Analytical
precision is 0.1%o for carbon and 0.2%o for nitrogen.

Stable isotope values for panthers and their prey are
reported in Table 2. Mean carbon and nitrogen isotope


WILKINS, ALLEN and REED: Florida Panthers

values for panthers in the northern and southern use
areas are not significantly different (613C%o, t-test, t=-
1.7, df 11, p=0.10; 815N, t-test, t=-0.97, df =13, p =0.35;
Fig. 2A). However, animals in the north have a wider
range of 813C%o and 815N%o values indicative of a more
varied diet.
When sorted by sex, male and female panthers
differ significantly in carbon (t-test, t = -3.7, df=17, p =
0.002), and nitrogen isotope values (t =-4.4, df= 15, p =
0.0005; Fig. 2B) indicating likely reliance upon a differ-
ent suite or combination of prey taxa. Also, the range of
male panther isotope values is greater than that of fe-
males suggesting a more varied diet in males.

Carbon (813C%o) isotope signatures increase by
1%o with each increase in trophic levels above the pri-
mary consumer. 813C%o values of deer are consistent
with a high C3 browsing diet and ranged from -24.1 %o to
-20.4%o with an average of -21.9 1.2%o. Feral hog
813C values are nested within the range for deer making
them indistinguishable isotopically, at -23.2%o and -21.6%o
with an average of -22.4 1.1%o. Raccoon 813C%o
values range from -23.9%o to -16.7%o with an average

of -19.9 2.4%o indicating a wide range of dietary ele-
ments including some with a marine component. Rabbit
613C%o values are -21.6%o to -20.4%o, consistent with a
diet of C plant foods. The two armadillos sampled have
widely varying 813C%o isotope values of -19.9%o and -
15.7%o suggesting that isotopic diversity is likely high
within this species, and that a larger sample of armadil-
los, as well as raccoon, is needed (Table 2, Fig. 3).
Armadillo prey items (invertebrates) may feed on a wide
variety of plant material, generating considerable varia-
tion in 613C%o values.
Male panther 813C%o isotope values range from -
20.1%o to -15.6%o with an average of -18.6 1.5%o and
females from -21.0%o to -19.1%o with an average of -
20.3 0.7%o (Table 2, Fig. 3). Three males from the
northern use area are outside the range of deer and hog
values suggesting a large portion of their diet included
other prey taxa, possibly raccoon and armadillo. Two
Florida panthers (FP 15 and FP 16) were tracked by
Dalrymple and Bass (1996) and have carbon numbers
of-19.6%o and -19.9%o, respectively. Both animals sub-
sisted primarily on deer and hog but it was noted that FP
16 also killed eight raccoons during the tracking.

Table 2: Stable carbon and nitrogen isotope values for (A) Florida panthers and (B) their prey species. Florida
panthers are listed by Florida Museum of Natural History identification number (UFID) as well as a FP number
assigned by the Florida Fish and Wildlife Conservation Commission. The prey taxa are listed by the species, county
collected, and UFID for animals sampled from the Florida Museum of Natural History mammal collection.

A. Florida Panthers




Use Area



Table 2 continued:

B. Prey Species

Odocoileus virginianus (White-tailed Deer)

UFID 24907

UFID 9512
UFID 22583
UFID 22582
UFID 9508


Procyon lotor (Raccoon)
UFID 25949

UFID 22569
UFID 10192

UFID 24012

UFID 10089
UFID 4611

Sus scrofa (Feral Hog)


Sylvilagus palustris (Rabbit)


Dasypus novemcintus (Armadillo)




















WILKINS, ALLEN and REED: Florida Panthers


0 0

* North
o South

-22 -21 -20 -19 -18

0 -

c 8.5 -


-17 -16 -15

* .4

* Male
* Female

-22 -21 -20 -19 -18
o "C%o

-17 -16 -15

Figure 2: Stable carbon (613C%o) and nitrogen ( 15N%o) isotope ratios for 20 Florida panthers from the Florida
Museum of Natural History. There was no significant difference between animals in the north (n = 9)and southern
(n =11) use areas (A). A significant difference was found between male (n = 12) and female (n = 8) panthers (B).

When nitrogen isotope signatures (615N%o) are
plotted (Fig. 3), panthers have higher 815N%o signatures
than most of their prey taxa as expected. Nitrogen iso-
tope signatures increase 2-4%o with each step up the
food web, and values for animals at each trophic level
met this expectation. Rabbits from the Everglades have
the lowest nitrogen values, with a range of 1.2%o 2.9%o,
n=3), perhaps due to the use of agricultural fertilizers
(SWFMD 1992). Fertilizers are depleted in 15N with
615N%o values ranging from -0.5 to 0.3%o (Schwertl et
al. 2005). It is also possible, that vegetation in the diets
of the sampled rabbits was depleted by high levels of
nitrogen-fixing bacteria in agricultural runoff (Craft and
Richardson 1998). If so, further sampling of rabbits may
show distinct differences between rabbit isotope values
in various micro-environmental zones within southern
Deer 615N%o isotope signatures are highly vari-

able ranging from 3.7%o 8.8%o with an average of 5.6
+ 1.4%o (n=10); the lowest value (3.7%o) is from a Dade
County specimen. Feral hog 615N%o values again fell
within the range of deer at 4.8%o and 6.5%owith an av-
erage of 5.7 1.2%o (n=2) making their influence on
panther 615N%o indistinguishable isotopically from that
of deer. Armadillo 615N%o values are 7.1%o and 7.2%o
(n=2) slightly higher than deer (5.6%o) as expected given
their insectivorous diet (Taber 1945). Raccoon 615N%o
values range from 6.5%o 10.7%o with an average of
8.4 1.1%o (n=ll) representing a varied omnivorous
diet (Whitney 1931). Enriched raccoon 61 5N%o values
suggest the input of marine resources such as sea turtle
eggs, crustaceans, bivalves and other marine inverte-
Male panther 615N%o values range from 8.3%o -
9.8%o (n=12) with an average of 8.9 0.5%o and female
panther 615N%o values from 7.9%o 8.5%o (n=8) with

*. *.4 : *
N .
0 ft

E 0 1



4 -



* Raccoon
a Deer
A Rabbit
* Armadillo
xFeral Hog
. Female Panther
* Male Panther

25 -23 -21 -19 -17 -15

-25 -23 -21 -19 -17 -15

Figure 3: Stable carbon (613C%o) and nitrogen (65"N%o) isotope ratios for 20 Florida panthers and common prey
species. Part A includes all data points and B the averages; bars represent one standard deviation.

o 9

. 8.5-





4 0


E Deer
A Rabbit
X Feral Hog
o Female Panther
Male Panther

an average of 8.1 + 0.2%o. These values are consistent
with a diet primarily based on deer and hogs, as reported
in earlier dietary studies (Maehr et al. 1990; Dalrymple
& Bass 1996). Everglades panthers 15 and 16 (UF24563
and UF29821) were tracked by Dalrymple and Bass
(1996) who reported that FP 15 ate 14 deer, one marsh
rabbit and one opossum during the follows; FP 15 had a
nitrogen value of 8.1 %o. FP 16 had the most varied diet
in the Dalrymple and Bass study, seven deer, five hogs,
11 alligators, eight raccoons, one marsh rabbit, one otter,
one bobcat and exhibits a nitrogen isotope signature of
9.3%o. This nitrogen value is indicative of a diet high in
high- trophic-level prey taxa, such as alligator, raccoon
and bobcat, in keeping with the findings of Dalrymple
and Bass (1996). Bone collagen isotope values average
dietary intake over several years, perhaps the lifetime
of an adult panther, and tracking is likely to capture pre-
dation events over a narrower window of time. These
results then suggest that FL panther 15 had a diet high in
trophic level taxa for a long period of time.

Our stable isotope analyses are consistent with the ex-
pectation that deer and feral hog are primary dietary
elements. Our results have also revealed interesting
differences in male and female panther diets, but fur-
ther sampling of panthers and prey taxa is required to
substantiate sex-based dietary patterning. The most
likely prey base for female panthers is deer, hog and
rabbits with an emphasis on deer and hog indicated by
their nitrogen values. Male panthers had wider ranges
of 613C%o and 615N%o isotope values indicating a more
varied diet. It is possible that the more varied diet in
males is related to the fact that males have larger home
ranges (Maher et al. 2002; Jansen pers. com.). It is
possible that males may wander out of suitable deer or
hog habitat in defense of a large home range necessitat-
ing the addition of other prey species. Also, the simple
act of moving across large geographic distances may
cause male panthers to come into contact with a more
diverse assemblage of prey species. Alternatively, it is
possible that defending large home ranges makes it in-
efficient to remain at kill-sites of very large prey (deer
or hog) as long as females. The pronounced differ-
ences in male and female panther isotope values should
be investigated further.
In particular, male panthers in the north have car-
bon values that appear more representative of raccoon
and armadillo values. A dietary cline has been reported
with panthers in the north feeding primarily on feral hog
whereas those in the south feed primarily on deer and
small mammals (Beier et al. 2003). We found no differ-


ences isotopically between the diets of northern and
southern panthers, and show no significant difference
between deer and hog isotope signatures, thus cannot
speak to this issue in greater detail. A larger sample of
deer and hogs may reveal differences in their isotopic
signatures, given the omnivorous diet of feral hogs.
Three male panthers in the northern study area appear
to have relied heavily on small mammals, a result not
previously found in other studies. Two of the three, FP
45 (UF29262) and FP 63 (UF29819), were likely killed
by other panthers. These results suggest heavy compe-
tition between panthers in the north for prey or mating
opportunity. The third panther, FP 34 (UF26844), died
of a bacterial infection. All three panthers had Harris
lines and other osteopathologies, which are suggestive
of overall health problems (Wilkins et al. 2007). It is
possible that stable isotope chemistry can assist in iden-
tifying sick animals that were not able to hunt efficiently
and were thus forced to eat smaller prey. Similarly, stable
isotopes may reveal areas in which prey densities are
low and panthers are competing heavily for food. Sam-
pling many more animals in each area may help to dif-
ferentiate between these two issues. Our results, although
preliminary, suggest that stable isotopes may reveal im-
portant dietary differences among Florida panthers and
prove to be a useful management tool.

Anthropologist John Krigbaum advised on the applica-
bility of isotope analyses to Florida panthers. We would
like to thank Deborah Jansen for providing important
samples for this research. This work was supported in
part by the Florida Fish and Wildlife Conservation Com-

Ambrose, S. H. 1990. Preparation and characterize
tion of bone and tooth collagen for isotopic analy-
sis. Journal ofArchaeological Science, 17:431-451.
Ambrose, S. H. 1991. Effects of diet, climate and physi-
ology on nitrogen isotope abundances in terrestrial
foodwebs. Journal of Archaeological Science,
Beier, P., M. R. Vaughan, M. J. Conroy, & H. Quigley.
2003. An analysis of scientific literature related to
the Florida panther: Final Report, Florida Fish and
Wildlife Conservation Commission, Tallahassee, FL.
Cerling, T. E., J. R. Ehleringer, & J. M. Harris. 1998.
Carbon dioxide starvation, the development of C4
ecosystems, and mammalian evolution. Philosophi-
cal Transactions of the Royal Society of London B,

WILKINS, ALLEN and REED: Florida Panthers

Chisholm, B. S., E. D. Nelson, & H. P. Schwarcz. 1982.
Stable-carbon isotope ratios as a measure of ma-
rine versus terrestrial protein in ancient diets. Sci-
ence, 216:1131-1132.
Craft, C. B., & C. J. Richardson. 1998. Recent and long
term soil accretion and nutrient accumulation in the
Everglades, Soil Science Society of America Jour-
nal, 62(3):834-843.
Dalrymple, G. H., & 0. L. Bass. 1996. The diet of the
Florida panther in Everglades National Park, Florida.
Bulletin of the Florida Museum of Natural History,
Ehleringer, J. R., T. E. Cerling, & B. R. Helliker. 1997.
C4 photosynthesis, atmospheric CO2, and climate.
Oecologia, 112:285-299.
Evans, R. D., & J. R. Ehleringer. 1994. Water and nitro-
gen dynamics in an arid woodland. Oecologia,
Farquhar, G. D., J. R. Ehleringer, & K. T. Hubick. 1989.
Carbon isotope discrimination and photosynthesis.
Annual Review of Plant Physiology and Molecular
Biology, 40:503-537.
Fleming, D.M., Schortemeyer J., &Ault J. 1994. Distri-
bution, abundance and demography of white-tailed
deer in the Everglades. Pp. 247-274 in D. Jordan,
ed. Proceedings of the Florida Panther Conference,
Fort Meyers, Florida. U.S. Fish and Wildlife Ser-
vice, Gainesville, Florida.
Hobson, K. A., D. M. Shell, D. Renouf, & E.
Noseworthy. 1996. Stable carbon and nitrogen iso-
topic fractionation between diet and tissues of cap-
tive seals: implications for dietary reconstructions
involving marine mammals. Canadian Journal of
Fisheries and Aquatic Sciences, 53:528-533.
Hobson, K. A.,& H. P. Schwarcz. 1986. The variation
in 13C values in bone collagen for two wild herbi-
vore populations: Implications for paleodiet stud-
ies. Journal ofArchaeological Science, 13:101-106.
Iriarte, J.A., Franklin W.L., Johnson W.E., & Redford
K.H. 1990. Biogeographic variation of food habits
and body size of the America puma. Oecologia,
Kelly, J. E. 2000. Stable Isotopes of carbon and nitrogen
in the study of avian and mammalian trophic ecol-
ogy. Canadian Journal of Zoology, 78:1-27.
Leighty, R. G, M. B. Marco, G A. Swenson, R. E.
Caldwell, J. R. Henderson, 0. C. Olson, & G C.
Willson Jr. 1954. Soil Survey (detailed reconnais-
sance) of Collier County, Florida. Pp 72, Florida
Agriculture Experimental Station, University Florida,
Lovell, N. C., D. E. Nelson, & H. P. Schwarcz. 1986.

Carbon isotope ratios in palaeodiets: Lack of age
and sex effect. Archaeometry, 28:51-55.
Maehar, D.S., Land E.D., Roof J.C., & McCown J.W.
1989a. Early maternal behavior in the Florida pan-
ther. American Midland Naturalist 122:34-43.
Maehr, D. S., R. C. Belden, D. E. Land, & L. Wilkins.
1990. Food Habits of panthers in Soutwest Florida.
Journal ofWildlife Management, 54(3):420-423.
Maehr, D. S., D. E. Land, D. B. Shindle, 0. L. Bass, &
T. S. Hoctor. 2002. Florida panther dispersal and
conservation. Biological Conservation, 106:187-197.
McCown, J. W. 1991. Big Cypress deer/panther rela-
tionships: deer herd health and reproduction. Final
Report, Florida Game and Fresh Water Fish Com-
mission, Tallahassee, Florida.
Pate, F.D. 1994. Bone chemistry and paleodiet. J. Ar-
chaeology Method Theory 1:161-209.
Roelke, M. E. 1990. Florida panther biomedical investi-
gation. Final Performance Report, Florida Game
and Fresh Water Fish Commission, Tallahassee,
Schoeller, D. A. 1999. Isotope fractionation: Why aren't
we what we eat? Journal of Archaeological Sci-
ence, 26:667-674.
Schortemeyer, J.L. 1994. Habitat management for pan-
thers in South Florida An overview. Pp. 460-466
in D. Jordan, ed. Proceedings of the Florida Pan-
ther Conference. Fort Meyers, Florida. U.S. Fish
and Wildlife Service, Gainesville, Florida, USA.
Schwertl, M., K. Auerswald, R. Schaufele, & Schnyder.
2005. Carbon and nitrogen stable isotope composi-
tion of cattle hair: ecological fingerprints of pro-
duction systems? Agriculture, Ecosystems and En-
vironment, 109:153-165.
South Florida Water Management District. 1992. Sur-
face water improvement and management plan for
the Everglades: Supporting information document.
SFWMD, West Palm Beach, FL.
Szepanski, M.M., Ben-David, M., & Van Ballenberghe
V. 1999. Assessment of anadromous salmon
resources in the diet of the Alexander Archipelago
wolf using stable isotope analysis. Oecologia,
Taber, F. W. 1945. Contribution on the life history and
the ecology of the Nine-Banded armadillo. Journal
of Mammalogy, 26(3):211-226.
Tauber H (1981) 13C evidence for dietary habits of pre-
historic man in Denmark. Nature 292:332-333.
Urton, E. J., & K. A. Hobson. 2005. Intrapopulation
variation in gray wolf isotope (d (15)N and d (13)C)
profiles: implications for the ecology of individuals.
Oecologia, 145(2):317-26.


Wilkins L., Allen J.M., Coltrain J. Flanagin S., Allen, TD.,
& Reed, D.L. 2007. Part I: Osteology as a means
of assessing Florida panther health. In Methods of
assessing health diet of Florida panthers (Puma
concolor) using museum specimens. FLMNH Bul-

Natural history museums are charged with the mission of archiving the world's biodiversity. As threats to the world's
biodiversity increase it is becoming apparent that the only place to view or study certain organisms will be within
natural history collections. Cooperation between natural history museums and the agencies that monitor threatened
wildlife can facilitate the archival of specimens from seriously threatened species. In the case of Florida panthers, the
Florida Museum of Natural History has cooperated with the Florida Fish and Wildlife Conservation Commission, The
National Park Service, and US Fish and Wildlife Service to recover, process, and archive skeletal, genetic, and other
material from every dead Florida panther recovered during the last 20 years. In so doing, the FLMNH now has an
impressive collection of over 140 specimens of Florida panther.
Archival material such as that maintained in natural history collections provides a unique opportunity to examine
a population over time and link health parameters and dietary shifts to life history parameters such as age, parentage,
genetic profile and/or habitat. In the case of small, highly inbred populations like Florida panthers, it is important to
note that we cannot foresee all of the health threats to the population without the benefit of having many specimens
to examine. Similarly, we cannot foresee all of the tools and technologies that will be available to study these museum
specimens. As such, it is important that we maintain a safely archived, permanent repository for Florida panthers
and other critically endangered organisms.
In studying the osteology and osteopathology of over 140 individual Florida Panthers, Wilkins et al. (2007, Part
I) were able to not only examine the correlates of bone abnormalities that have been well documented, but they also
noticed new findings relating to systemic infections that have yet to be studied in detail. Allen et al. (2007, Part II)
examined the diet of Florida panthers with stable isotope geochemistry, which has not yet been applied to Florida
panthers. This study was able to show significant differences in the diet of male and female panthers, which had
never been noted before in the literature. Both studies demonstrate the value of natural history collections to the
conservation and maintenance of critically endangered species.

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs