free-roaming cats is considered to be an important
problem because of concerns about animal welfare,
wildlife predation, and zoonotic disease transmission.1
Methods for controlling populations of free-roaming
cats are controversial, in large part because of a lack of
data needed to assess the various options.1-3Domestic
cats are considered to be prolific breeders, with
females capable of bearing their first litter before 1
year of age and able to have multiple litters each year
thereafter.4,aHowever, estimates of the reproductive
he size of the free-roaming cat population in the
United States is unknown, but overpopulation of
capacity of female cats and the consequences of
unabated reproduction are often extrapolated beyond
scientific reliability, as they typically fail to use realis-
tic litter sizes or ignore kitten mortality rates.5,6The
purpose of this study was to determine reproductive
parameters of naturally breeding free-roaming cats.
For purposes of the present study, free-roaming cats
were considered to be cats that were not confined
when outdoors. Feral cats were considered to be a
subset of free-roaming cats.
Materials and Methods
Data for the study were collected from 2 sources.
Between May 1998 and October 2000, data were collected
on 71 sexually intact female cats in 9 managed feral cat
colonies in Randolph County, NC. The cats were being mon-
itored to assess the impact of a trap-neuter-return (TNR)
program on feral cat colony population dynamics. As each
colony was enrolled in the population dynamics study, all
cats in the colony were captured and pregnancy, lactation,
and estrus statuses of the female cats were determined.
Colonies were included in the study only if they had an
established caretaker who provided food and water on a reg-
ular basis and either owned the land on which the colony
resided or had the permission of the landowner to tend to
the cats; cats in the colony had access to adequate shelter,
such as a barn, storage shed, carport, basement, or crawl
space; the colony consisted of at least 10 adult cats (ie, cats
> 6 months old), with at least 3 adult male cats; the colony
was located in a rural or suburban residential area at least 1
km from the nearest 4-lane road; and the colony caretaker
agreed to random assignment of the colony to a treatment
group (control vs surgical sterilization), ear-tipping of all
cats for permanent identification, and regular visits to the
colony by the investigators for data collection. At the time of
inclusion in the study, cats in the colony were live trapped
and anesthetized with an IM injection of ketamine, tileta-
mine, zolazepam, and xylazine.7Female cats in 6 colonies (n
= 44) were surgically neutered. Female cats in the remaining
3 colonies (n = 27) were not surgically neutered. All cats in
all colonies were vaccinated against rhinotracheitis, pan-
leukopenia, calicivirus infection, FeLV infection, and rabies
and treated with ivermectin. Food and water were provided
daily. Cats were returned to their colony sites and monitored
for a 2-year follow-up period. During that time, census data
were collected on the colonies at least twice weekly by the
caretakers or principal investigator. Data collected included
parity, birth dates, litter sizes, and outcome of kittens. Parity
was estimated on the basis of whether the caretaker had
observed the cat to have been pregnant, lactating, or caring
for a litter previously and the reproductive status of the cat
at the time of enrollment in the study. Kittens that survived
to 6 months of age were trapped and enrolled in the popula-
tion dynamics study but were not enrolled in the present
study. Litter size data were collected on 61 litters produced
by the 27 control females during the 2-year study period.
Data were available on time of birth for all 61 litters, on lit-
ter-specific mortality rates for 59 litters, and on litter size for
50 litters. All cats were trapped, neutered, and vaccinated at
JAVMA, Vol 225, No. 9, November 1, 2004 Scientific Reports: Original Study1399
Reproductive capacity of free-roaming
domestic cats and kitten survival rate
Felicia B. Nutter, DVM; Jay F . Levine, DVM, MPH; Michael K. Stoskopf, DVM, PhD, DACZM
Objective—To determine reproductive capacity of
naturally breeding free-roaming domestic cats and kit-
ten survival rate.
Design—Prospective cohort and retrospective cross-
Animals—2,332 female cats brought to a trap-neuter-
return clinic for neutering and 71 female cats and 171
kittens comprising 50 litters from a cohort study of
feral cats in managed colonies.
Procedure—Data collected for all cats included preg-
nancy, lactation, and estrus status and number of
fetuses for pregnant cats. Additional data collected for
feral cats in managed colonies included numbers of
litters per year and kittens per litter, date of birth, kit-
ten survival rate, and causes of death.
Results—Pregnant cats were observed in all months
of the year, but the percentage of cats found to be
pregnant was highest in March, April, and May. Cats
produced a mean of 1.4 litters/y, with a median of 3
kittens/litter (range, 1 to 6). Overall, 127 of 169 (75%)
kittens died or disappeared before 6 months of age.
Trauma was the most common cause of death.
Conclusions and Clinical Relevance—Results illus-
trate the high reproductive capacity of free-roaming
domestic cats. Realistic estimates of the reproductive
capacity of female cats may be useful in assessing the
effectiveness of population control strategies. (J Am Vet
Med Assoc 2004;225:1399–1402)
From the Environmental Medicine Consortium (Nutter, Levine,
Stoskopf) and the Departments of Clinical Sciences (Nutter,
Stoskopf) and Population Health and Pathobiology (Levine),
College of Veterinary Medicine, North Carolina State University,
Raleigh, NC 27606.
Supported by the Morris Animal Foundation, the William and
Charlotte Parks Foundation, the College of Veterinary Medicine at
North Carolina State University, the Randolph County Humane
Society, and the North Carolina Zoological Society.
The authors thank Beth Chittick, Richard Ford, Michael Loomis,
Roger Powell, Kurt Sladky, and Chris Whittier for contributions to
the study design.
Address correspondence to Dr. Stoskopf.
the end of the population dynamics study
and again returned to their colony sites.
Data were also collected on a conve-
nience sample of 2,332 free-roaming female
cats trapped and brought by their caretakers to
a monthly TNR clinic in Raleigh, NC,b
between February 1996 and December 2001.
Information on living conditions of these cats
was not available; however, data on pregnancy
status (ie, identification of embryos or fetuses
visible without magnification), number of
fetuses per pregnancy, lactation status (ie, abil-
ity to express milk from teats), and estrus sta-
tus (ie, ovarian follicle development and uter-
ine status) were collected by veterinary techni-
cians and assistants at the time of neutering
and recorded on a standardized recording
sheet. Pregnancy status was recorded for 2,281
of the 2,332 cats, and 608 cats were confirmed
to be pregnant on the basis of identification of
embryos or fetuses in the uterus. Fetus counts
were recorded for 317 of the 608 pregnancies.
Lactation status was recorded for 2,205 cats,
and estrus status was recorded for 2,227.
Data from the population dynamics study
were used to determine litter sizes, litters per
year, kitten survival rate, and causes of death
for kittens that died. Descriptive statistics were
calculated, and associations between parity, litter size, kitten
survival rate, and litter order (first, second, or third per year)
were assessed with z tests. Commercial softwarecwas used for
all calculations; values of P < 0.05 were considered significant.
Distributions of fetus counts and litter sizes from live births
were significantly different (P = 0.008; Kolmogorov-Smirnov
test); thus, analyses were performed separately for each. Fetus
counts and litter size were compared with the Mann Whitney
U test. Distributions of pregnancy, lactation, and estrus status-
es were not significantly different between cats in the popula-
tion dynamics study and cats examined in the TNR program.
Therefore, data were pooled for further analysis.
Survival time for 169 kittens was evaluated by means of
the Kaplan-Meier product-limit estimate of the survivor
function.8Observations were right-censored at the end of 6
months (180 days). Survival times were compared by parity
of the queen, litter size, and litter order with the Peto and
Peto generalized Wilcoxon test for k samples with censored
data8; values of P < 0.05 were considered significant.
Six hundred twenty-five cats in the study (608 in the
TNR program and 17 in the population dynamics study)
were pregnant. Pregnancies were observed in all months
of the year, but the percentage of cats found to be preg-
nant was highest in March, April, and May and lowest in
November (Figure 1). Distributions of the percentages of
cats in estrus and the percentages of cats lactating had
similar pattens, with the peak in percentage of cats in
estrus preceding the peak in percentage of cats found to
be pregnant and the peak in percentage of cats lactating
following. Overall, 149 of 2,276 cats (131/2,205 cats in
the TNR program and 18 of 71 cats in the population
dynamics study) were reported to be lactating, and 295 of
2,298 cats (277/2,227 cats in the TNR program and 18 of
71 cats in the population dynamics study) were in estrus.
Information on fetus count was available for 317
cats in the TNR program and 17 cats in the population
dynamics study (1,401 total fetuses), and information
on litter size was available for 50 litters produced by
cats in the population dynamics study (171 total kit-
tens). Fetus count (median, 4; interquartile range [25th
to 75th percentile], 2 to 6; range, 1 to 10) was signifi-
cantly (P < 0.001) higher than litter size (median, 3;
interquartile range, 2 to 4; range, 1 to 6). Cats in the
population dynamics study produced a mean of 1.4 lit-
ters/y, with a maximum of 3 litters/y.
Survival data were available for 169 kittens.
Overall, 127 of the 169 (75%) kittens died (n = 87) or
disappeared (40) before 6 months of age. Median litter-
specific mortality rate was 75% (interquartile range,
20% to 100%; range, 0% to 100%). Kitten mortality
rate was not significantly associated with maternal par-
ity (P = 0.19), litter size (P = 0.10), or litter order (P =
0.38). Eighty-one of the 169 (48%) kittens died or dis-
appeared before they were 100 days old (Figure 2).
Median survival time was 113 days (10th to 90th per-
centile range, 24 to 180 days). Survival time was not
significantly associated with maternal parity (P = 0.12),
1400 Scientific Reports: Original Study JAVMA, Vol 225, No. 9, November 1, 2004
Figure 1—Percentages of free-roaming cats found to be pregnant, lactating, or in
estrus as a function of month of examination. Data are based on 2,332 free-roaming
female cats brought to a trap-neuter-return clinic for neutering and 71 female cats in
managed feral cat colonies.
Figure 2—Kaplan-Meier survival curve for 169 kittens born to
free-roaming cats. Kittens were observed for 180 days after birth.
litter size (P = 0.11), or litter order (P = 0.58). Causes
of death were determined for 41 of the 87 (47%) kit-
tens reported to have died. Thirty-seven of the 41
(90%) died as a result of trauma, with attacks by stray
and owned dogs (n = 18) and motor vehicle accident
(10) being the most common types of trauma. Other
types of trauma that resulted in > 1 death included falls
from haylofts (n = 2), being stepped on by horses or
people (3), and a suspected episode of infanticide (3).
Cause of death was not determined for 46 of the 87
(53%) kittens reported to have died, but many report-
edly had signs of disease, including upper respiratory
tract disease and diarrhea, prior to death.
For 10 female kittens born into control feral cat
colonies, ages at which they produced their first litters
were recorded. Median age at first parity was 10.5
months (interquartile range, 8 to 12 months; range, 6
to 15 months).
Results of the present study reinforce concerns
about the high reproductive capacity of free-roaming
domestic cats. Although cats are considered to be sea-
sonally polyestrous with a defined anestrus period
associated with day length,9,10pregnant cats were iden-
tified during all months of the year in the present
study, and similar findings have been reported previ-
ously.11However, only 15 pregnancies were identified
outside the spring and summer breeding season during
the 6 years of the present study. This would support a
hypothesis that seasonal births are dependent on opti-
mal environmental conditions.4
In the present study, the proportion of pregnant
cats peaked during the spring and late summer, which
is consistent with reported patterns in Florida,12
Australia,13and South Africa.14Proportions of the
queen population in estrus and lactation followed sim-
ilar seasonal patterns, with the percentage in estrus
peaking prior to the peak in the percentage that were
pregnant and the percentage lactating peaking after, as
expected. The proportions of the queen population in
estrus and lactation were lower than would be expect-
ed given the reported percentage that were pregnant,
most likely because of the difficulty of identifying
estrus and lactation, compared with identifying preg-
nancy. Also, estrus lasts a shorter time than either preg-
nancy or lactation, which would add to a bias for
detecting pregnancy during monthly TNR clinics.
Reported values for mean litter sizes for free-roam-
ing, laboratory-raised, and cattery cats vary from 2.1 to
5 kittens/litter, with ranges from 1 to 10 kittens/litter
having been reported,11,14-22and litter sizes in the pre-
sent study were consistent with these values. Litter size
was significantly smaller than fetus count in the pre-
sent study, which may be an indication of late gesta-
tional or early neonatal losses that were not directly
observed. Litters of kittens could not always be located
immediately after birth, and kittens were typically first
counted at 3 to 4 weeks of age, when they began to
visit the colony feeding site. This has been the only
method used by some researchers to determine litter
sizes18and, on the basis of our findings, results in con-
servative estimates of actual reproduction.
On average, cats in the present study gave birth to
1.4 litters/y, although 2 cats had 3 litters in a single
year. Production of multiple litters a year has been neg-
atively associated with survival of kittens in the first
litter in other studies,23,24but we did not find a clear
association between those variables in our data.
However, the 2 females that each produced 3 litters in
a single year did have 100% mortality rates for at least
1 of the first 2 litters in that year. This association
makes intuitive sense but requires a larger data set to
appropriately interpret the relationship. Of 10 female
cats born into control feral cat colonies and closely fol-
lowed to determine age at first parity, 9 produced their
first litters at < 1 year of age, with 1 cat giving birth at
6 months of age. This young age at first reproduction
combined with the potential to produce multiple litters
a year contributes to the perception of cats as prolific
High neonatal and juvenile mortality rates are
widely reported for domestic cats. Reported percent-
ages of kittens that die in the early neonatal period (ie,
up to 6 or 8 weeks of age) range from 12.8% to
48%.22,25,26In 1 study,26up to 90% of kittens died before
6 months of age. Similarly, 81 of 169 (48%) kittens in
the present study had died or disappeared before they
were 100 days old, and 127 (75%) had died or disap-
peared before they were 6 months old. Trauma
accounted for the death of most kittens for which
cause of death was confirmed. Causes of kitten death
may be highly dependent on a variety of environmen-
tal variables, and considerable variation in these data
should be expected between study sites, making gener-
alization difficult. Variations are also likely within
causes of death. For example, single or multiple stray
dogs were responsible for deaths of kittens in 2
colonies in the present study, whereas a caretaker’s
dogs were responsible for the deaths of multiple kittens
in a third colony. It is likely that both motor vehicle
accidents and dog attacks were overrepresented as
causes of death in the present study because the noise
or graphic visual evidence associated with these causes
of death is likely to draw attention. Cats that become
debilitated often seek hiding places, making it less like-
ly that cats that die of illness or disease will be identi-
fied. Predation of kittens by other animals, such as rap-
tors, foxes, and coyotes, likely resulted in the disap-
pearance of some kittens in the present study but was
not recorded as a cause of death, likely because the car-
casses were consumed. Causes of kitten death and the
relative rank of contribution to the overall mortality
rate were reported in a study24of farm cats in Ithaca,
NY; however, relative rankings were different from
rankings in the present study, likely because of differ-
ences in study design and environmental conditions of
the kittens, such as human population density, road
density, road proximity, and climatic conditions.
Examined out of context, our data would tend to
reinforce the popular notion that kittens born to free-
roaming cats live a marginal existence and that their
mortality rate is unreasonably high. However, reported
kitten mortality rate was consistent with reported rates
for similarly sized wild carnivores,27,28suggesting that
the living conditions of free-roaming cats are compara-
JAVMA, Vol 225, No. 9, November 1, 2004 Scientific Reports: Original Study 1401
ble to those of other wildlife. It also suggests that the
assessment and management of feral cat colonies with
methods developed for studying other small wild car-
nivores are appropriate. Results of the present study
provide information needed to develop reliable esti-
mates of the impact of reproduction by sexually intact
free-roaming domestic cats in rural and suburban
regions of the southeastern United States.
aLiberg O. Predation and social behavior in a population of domestic
cats: an evolutionary perspective. PhD dissertation, Department of
Animal Ecology, University of Lund, Lund, Sweden, 1981.
bOperation Catnip Inc, Raleigh, NC.
cStatView 5, SAS Institute Inc, Cary, NC.
1. Patronek GJ. Free-roaming and feral cats—their impact on
wildlife and human beings. J Am Vet Med Assoc 1998;212:218–226.
2. Mahlow JC, Slater MR. Current issues in the control of stray
and feral cats. J Am Vet Med Assoc 1996;209:2016–2020.
3. Slater MR. Community approaches to feral cats: problems, alterna-
tives, and recommendations. Washington, DC: Humane Society Press, 2002.
4. Deag JM, Manning A, Lawrence CE. Factors influencing the
mother-kitten relationship. In: Turner DC, Bateson P , eds. The domestic
cat: the biology of its behavior. Cambridge, UK: Cambridge University
5. Olson PN, Johnson SD. New developments in small animal
population control. J Am Vet Med Assoc 1993;202:904–909.
6. Luoma J. Catfight. Audobon 1997;Jul–Aug:84–91.
7. Williams LS, Levy JK, Robertson SA, et al. Use of the anes-
thetic combination of tiletamine, zolazepam, ketamine, and xylazine
for neutering feral cats. J Am Vet Med Assoc 2002;220:1491–1495.
8. Lee ET, Lee ETT, Wang JW. Statistical methods for survival
data analysis. 3rd ed. New York: John Wiley & Sons, 2003.
9. Hurni H. Daylength and breeding in the domestic cat. Lab
10. Scott PP , Lloyd-Jacob MA. Reduction in the anestrous period
of laboratory cats by increased illumination. Nature 1959;184(suppl
11. Prescott CW. Reproduction patterns in the domestic cat.
Aust Vet J 1973;49:126–129.
12. Scott KK, Levy JK, Crawford CP . Characteristics of free-roam-
ing cats evaluated in a trap-neuter-return program. J Am Vet Med Assoc
13. Jones E, Coman BJ. Ecology of the feral cat, Felis catus (L),
in South Eastern Australia. II. Reproduction. Aust Wild Res 1982;9:
14. van Aarde RJ. Reproduction and population ecology in the
feral house cat, Felis catus, on Marion Island. Carnivore Genetics
15. Ekstrand C, Linde-Forsberg C. Dystocia in the cat: a retro-
spective case study of 155 cases. J Small Anim Pract 1994;35:
16. Kane E, Allard RE, Douglass GM. The influence of litter size
on weight change during feline gestation and lactation. Feline Pract
17. Lawler DF , Monti KF . Morbidity and mortality in neonatal
kittens. Am J Vet Res 1984;45:1455–1459.
18. Mirmovitch V. Spatial organization of urban feral cats (Felis
catus) in Jerusalem. Wildl Res 1995;22:299–310.
19. Povey RC. Reproduction in the pedigree female cat. A sur-
vey of breeders. Can Vet J 1978;19:207–213.
20. Robinson R, Cox HW. Reproductive performance in a cat
colony over a 10-year period. Lab Anim 1970;4:99–112.
21. Root M, Johnston SD, Olson PN. Estrous length, pregnancy
rate, gestation and parturition lengths, litter size and juvenile mor-
tality in the domestic cat. J Am Anim Hosp Assoc 1995;31:
22. Scott FW, Geissinger C, Peltz R. Kitten mortality survey.
Feline Pract 1978;8(1):31–34.
23. Ewer RF . The carnivores. London: Weidenfeld & Nicholson,
24. Wolski TR. The life of the barnyard cat. Feline Health Perspect
25. Jemmett JE, Evans JM. A survey of sexual behavior and
reproduction in female cats. J Small Anim Pract 1977;18:31–37.
26. van Aarde RJ. Population biology and the control of feral
cats on Marion Island. Acta Zool Fennica 1984;172:107–110.
27. Cypher BL, Warrick GD, Otten MRM, et al. Population
dynamics of San Joaquin kit foxes at the Naval Petroleum Reserves
in California. Wildl Monthly 2000;145:1–43.
28. Fritts SH, Sealander JA. Reproductive biology and popula-
tion characteristics of bobcats (Lynx rufus) in Arkansas. J Mammal
1402 Scientific Reports: Original Study JAVMA, Vol 225, No. 9, November 1, 2004