Cleide L.S. Oliveira1, BS; Maria L.F. Gomes2, BS; Julia
Aikawa1, BS; Graziella M. Silva2, BS; Luiz R. Francisco2, BS;
Rosana N. Morais1, DVM, PhD
Margays (Leopardus wiedii) and tigrinas (Leopardus tigrinus), as well as most small
cat species, have a very low breeding success in captivity which may partly be a consequence of poor captive environment
and/or husbandry.1 Difficulties to cope with threatening or aversive captive conditions as well as low stimulus
diversity environments may lead animals to experience stress and boredom, which in turn can compromise health and
reproduction. From the scarce data on the normal behavior of margays and tigrinas, either in the wild or in captivity, it
is assumed that they are nocturnal animals, spending most of the daytime resting or sleeping. Margays have more arboreal
habits than tigrinas, choosing branches of trees or platforms to rest upon. Both species are very secretive and shy,
spending large amounts of time out of sight or asleep when in exhibit. In previous studies, it was found that male margays
and tigrinas excrete significantly higher levels of corticoids in feces than ocelots in similar captive conditions, while
fecal androgen levels and sperm quality were significantly lower.2,3 Efforts to increase reproductive success
should include environmental enrichment to alleviate stress, stimulate activity, and provide animals with the perception
of security and confort. Thus the objective of this study was to evaluate the behavioral effects of three different
captive conditions in margays and tigrinas. Specifically we compared a baseline situation (condition 1) with enclosure
enrichment through addition of furnishings (condition 2) and enclosure enrichment plus providing live prey (condition 3).
Six adult margays (two males and four females) and four tigrinas (one male and three females) were
studied during a 90-day period. All tigrinas were singly housed and the two male margays were paired with females.
Throughout the experimental period, the cats were housed in off-exhibit, outdoor, wire-mesh enclosures (2.60 x 2.90 x 2.00
m) with a secluded hiding area (1.50 x 1.60 x 1.20 m) containing a dense box. During the baseline period (condition 1)
which lasted 30 days, the furnishings consisted of a raised (1.2 m) wooden platform (1.0 x 0.4 m) and a box containing
sand (0.8 x 0.4 x 0.3 m) on the floor, just below the platform. For the second 30-day period (condition 2), extra
enclosure furnishings were provided. One or two tree branches, placed diagonally from the floor or from the raised
platform to the top, a hollow log on the raised platform for tigrinas or hanging from the top over the platform for
margays, a vase (0.20 x 0.80 m) with vegetation at the front of enclosure on the same side of the platform, and, for
margays only, a rope hanging diagonally from the top (opposite direction to the branches) were placed within the exhibits.
During conditions 1 and 2, animals were fed a meat-based diet, supplemented with vitamins and minerals, offered once per
day in the afternoon. Early in the morning, freshly killed rats were offered at least twice per week as a supplement.
Water was available ad libitum from a small shallow pool on the front enclosure floor (opposite to the vase with
vegetation). During the last 30 days of the study (condition 3), enclosure conditions were maintained, however feeding
schedule was altered by providing live prey in alternate days instead of the supplement of rat carcasses. Each cat was fed
one adult live quail inside a small cardboard box with small holes, left on the floor of the enclosure before normal
feeding time. Behavioral observations were made from 1000 to 1500 hr, Monday-Friday, every week. This time frame was
chosen because of the regular husbandry and feeding routine which included cleaning the enclosures while the cats were
kept at the adjacent holding area. All cats were observed each day for a 30 min period in a rotational schedule with the
goal of obtaining observations which were representative of the cat's overall activity during the daily time frame.
Different behaviors were scored using instantaneous point sampling4 via direct observation at 1 min intervals
for 30 min periods and included one of the following major categories (sleeping, rest/alert, locomotion, stereotypic
pacing, and other) or the out-of-sight category, when the animals were hidden. Location in one of the four equal-sized
areas on the floor or at the hiding area, raised platform, branches, hollow log, or rope (only for margays) were also
recorded at 1 min intervals. For analysis, behaviors under each condition (excluding time while out-of-sight) as well as
location were expressed as daily percentages and comparisons between conditions were tested with a Kruskall-Wallis test.
The Mann-Whitney test was used for comparisons between species in each condition. Space use was evaluated using the spread
of participation index (SPI), as previously described for small cats.4
The two species greatly differed in relation to total daily activity and space utilization during the
baseline period, as well as to the effects of the captive modifications made under conditions 2 and 3. Tigrinas spent
98.5% of the time out of sight during condition 1, which did not differ (P > 0.05) from conditions 2 (97.3%) and
3 (95.0%). However during baseline, animals stayed hidden most of the time (98.5%) at the holding area, changing to the
hollow log (76.4%) during condition 2 and using almost equally the two locations (holding area, 44.3% and hollow log,
54.0%) during condition 3. During the rare occasions when tigrinas were visible, they were displaying stereotypic pacing
(74.5% at condition 1), which was reduced during condition 2 (13.6%) and 3 (42.9%), although differences were not
significant (P > 0.05). Pacing was replaced by increasing rest/alert and locomotion behaviors in condition 2
(81.4% rest/alert and 5.7% locomotion) and 3 (35.0% rest/alert and 35.7% locomotion). Margays on the other hand, were
visible most of the time (79.6%) during baseline condition and changes in the captive environment significantly (P
< 0.05) increased this percentage to 91.2% and 93.6% with the addition of new furnishings and live prey in conditions 2
and 3, respectively. During condition 1 however, animals spent most of the time sleeping (43.0%) or rest/alert (39.7%)
with very low motion activity (9.2%), including some stereotypic pacing (3.0%). At condition 2 and 3, animals spent more
time awake (P < 0.05), with sleeping time reduced, respectively, to 23.4% and 18.1%. Alertness was higher
(P < 0.05) either at condition 2 (62.2%) or 3 (70.0%) in relation to baseline. Locomotion was not affected,
representing around 8% of total activity throughout the experiment. The locations preferred by margays at baseline were
the raised platform (69.1%) and the holding area (20.0%). With new furnishings, time spent at holding area was reduced to
9.0% and at platform to 52.8%. Cats stayed at the hollow log 28.8% of the time with some incursions to the branches
(1.3%). This pattern was maintained at condition 3, with time divided between the platform (44.3%) and the hollow log
(35.0%). On a few occasions, margays were seen playing on the ropes (0.1%), however not quantified, keepers observed this
behavior more frequently. Space use, as indicated by SPI value, did not change significantly between conditions for both
species. Tigrinas has an SPI of 0.98 throughout the study (cats spent most of the time in one area). Margays had a SPI
around 0.91, which differed from tigrinas (P < 0.05), although both values are high, indicating minimum
utilization of the area. Margays tended to be on the ground with a higher frequency (8.8%) than tigrinas (1.1%) (P
In conclusion, these results indicate that, at least for margays, providing the enclosure with more
furnishings and hiding places can significantly reduce the time animals spent sleeping, an effect even more consistent
with the simultaneous inclusion of live prey. Although basal locomotion was not increased, apart from the short term
effects immediately following live prey presentation (data not included), alertness was increased and, even for tigrinas,
the identification of the animals outside in the hollow log may be interpreted as an increase in well-being and a
reduction of fear. Perhaps the low levels of locomotion is related to the lack of sufficient cover from keepers and zoo
personnel or to the nocturnal habits of these species. We are currently measuring fecal corticoids for these same animals
to evaluate impact of these behavioral changes in stress level.
This work was supported by a grant from Sector of Biological Sciences, Federal University of Parana
(PROSEAPEX), and Course of Specialization in Physiology (UFPR). The authors thank the zoo personnel for enclosure
furnishings and assistance with sample collection.
1. Mellen J. 1991. Factors influencing reproductive success in small captive exotic felids
(Felis sp.): a multiple regression analysis. Zoo Biol. 10:95-110.
2. Morais RN, N Moreira, W Moraes, RG Mucciolo, O Lacerda, MLF Gomes, WF Swanson, LH Graham,
JL Brown. 1996. Testicular and ovarian function in South American felids assessed by fecal steroids. Proc. Am. Assoc.
Zoo Vet. Pp. 561-565.
3. Morais RN, RG Mucciolo, MLF Gomes, O Lacerda, W Moraes, N Moreira, WF Swanson, JL Brown.
1997. Adrenal activity assessed by fecal corticoids and male reproductive traits in three South American felid species.
Proc. Am. Assoc. Zoo Vet. Pp. 220-223.
4. Shepherdson DJ, K Carlstead, JD Mellen, J Seidensticker. 1993. The influence of food
presentation on the behavior of small cats in confined environments. Zoo Biol. 12:203-216.