Content uploaded by Carol Max Lang
Author content
All content in this area was uploaded by Carol Max Lang on Mar 18, 2014
Content may be subject to copyright.
208
Use of cage space by guineapigs
W.
J.
WHITE], M. W. BALK! and C. M. LANG2
Laboratory Animals
(1989) 23, 208-214
ICharles River Laboratories, 251 Ballardvale Street, Wilmington, MA 01887-0630, USA and 2Department of
Comparative Medicine, The Milton S. Hershey Medical Center, The Pennsylvania State University, PO Box 850
Hershey, PA 17033, USA
Summary
Cage space requirements for laboratory animals
have been established by Government Regulation
and Recommendations. In order to test the
adequacy of these space allocations, the use
of cage floor area by breeding groups of
guineapigs was studied. A computer-coupled
video tracking system capable of imaging in
low light intensity as well as total darkness
was used to determine the average per cent
occupancy by guineapigs in all portions of a
cage over 12-h light and dark cycles. Simul-
taneous time synchronized slow motion video
recordings permitted an analysis of activity
to be coordinated with cage use data. Results
of the study revealed that breeding groups of
guinea pigs utilize the periphery of the cage
almost to the total exclusion of the centre of
the cage. Approximately
75-85070
of all occupancy
in both the day and evening hours occurred
in
47070
of the cage floor area located along
the periphery. Analysis of video recordings re-
vealed that the animals remained active through-
out the day and night with no prolonged period
of quiescence that could be associated with
sleep. Results of this study suggest that while
guidelines for housing guineapigs based on area
allocation per animal can be formulated and are
easy to administer, they cannot be supported by
the behavioural characteristics of these animals
or careful quantitation of their pattern of cage
space utilization.
Keywords: Guineapigs; Housing; Spatial
behaviour; Activity
Received 4March 1988. Accepted 4November 1988
The environment of laboratory animals has been
shown to be an important variable in their well-
being (Besch, 1980). While many physical
parameters of the environment are amenable to
quantitation (e.g. temperature, humidity, and the
presence of gaseous pollutants) the relationship
of other factors to physiological responses is
difficult to assess. Of particular concern is the
adequacy of cage space.
When animals are brought into the laboratory,
limitations must be placed on the space available
to them. Their behavioural response to limited
space can be measured, however, the analysis of
such responses can be easily over-interpreted.
Certainly, when selectingcages for group housing
animals in a laboratory, two concerns are
paramount: (1) is sufficient space available to
minimize destructive behaviour? and (2) is the
space provided used effectively by the animals?
This second concern is important since animal
housing space is expensive to construct and
operate, hence effective use of that space can
directly impact on research productivity.
While destructive behaviour patterns such as
hair or ear chewing have been reported in
guineapigs, there is no evidence of any direct
correlation between the amount of available cage
space and their incidence. In" fact, guineapigs
appear to exhibit a strong herd of family
behaviour which is supported by observations of
feral animals (Rood, 1970; Rood, 1972). This
need for close group association contrasts
sharply with the aggressive territorial behaviour
of other rodents.
Little direct information is available on the use
of cage surface area by most laboratory animal
species. In the United States, guidelines have
Use of cage space by guineapigs
been established by both the USDA and NIH,
specifying the required floor area for group and
individual housing of most commonly used
species (USDA, 1976; NAS, 1985). These space
recommendations are divided into weight cate-
gories and reproductive status sub-categories,
and expressed in square inches per animal. The
surface area recommended per animal is usually
many times the resting area of the animal. No
information is given as to the source of the data
used to formulate these recommendations. There
is also no indication that space requirements for
species such as guineapigs with a communal
social structure have been adjusted to compen-
sate for behavioural patterns.
Of the few studies conducted on guineapigs
caged singly or in groups, none has addressed
space requirements. Nicholls and others have
concentrated on the activity of guineapigs
in a variety of environmental situations
(Nicholls, 1922). Their studies suggest that,
unlike many other animals, the guinea pig does
not have a true diurnal cycle of activity. Instead,
it is active throughout the day and though less
active during the night, it has no periods of
prolonged sleep. The activity of guineapigs does
not appear to be significantly influenced by age
or the presence of other guineapigs. Some slight
sex-related differences have been reported in the
amount of activity, but they are minor (Nicholls,
1922).
While the density of caged guineapigs may
have some bearing on their 'perception of well-
being', other factors such as cage cleanliness or
the location of food and water may be equally
important. Since it is impossible to define, let
alone quantitatively assess, the relative impor-
tance cage space has on an animal's own per-
ceived well-being, careful study of how cage
space is used may be the only way to develop
standards for laboratory housing of these
animals.
In such a study, a working hypothesis would
be that as cage floor surface decreases, use of
the remaining floor area should proportionally
increase, as animals seek to avoid unwanted
interactions. However, if herding or family
209
grouping is important to a particular species,
shrinking floor space will not result in pro-
portionally higher use of space.
The current study was designed to compare the
use of a cage by two different group sizes of
guineapigs, in order to determine if the amount
and pattern of use might change in ways that
could be directly associated with the number of
animals housed in the cage. It was not the intent
of this study to assess or define stress in cage
housed guineapigs which could be subjective at
best (Levine, 1985).
Materials and methods
Animals
A total of 42 outbred Albino Hartley (Cr1:
(HA)BR) guineapigs (Charles River Laboratories
Inc., Wilmington, MA) were used in the study.
The animals ranged in weight from 723 to 885 g,
with a mean weight of 795
±
48 g. The animals
were randomly assigned to 6 breeding groups of
one male and 6 females. The animals were
maintained in mating groups within plastic tubs
only slightly larger than the study cage. In those
studies involving one male and 6 females, an
entire mating group was used. The order of
selection of the groups to be studied during each
observation period was made at random and
only 6 groups were studied under each set of
conditions (group size and time of day). In the
case of those observations of groups of one male
and 3 females, a male from one of the breeding
groups as well as three females from the same
group selected at random were used. Once
constituted, study groups were allowed to
acclimate for at least 14 days before the study.
In order to provide for acclimation to the study
cage, the animals were placed in the cage 30 min
prior to the beginning of each study period.
In order to allow the study to focus on
the
effects of group size in the presence of normal
sexual behaviour, conception had to be pre-
vented. While the presence of animals in varying
stages of gestation would not have been a
problem, the possibility of birth during observ-
ation periods of varying numbers of young
would have confounded acquisition of data and
210
interpretation of results. Since such variables
could be more appropriately addressed in future
studies of the effect of pregnancy and young on
the use of cage space, it was decided to
vasectomize the males used in the study.
Caging and environment
The animals were grouped housed in plastic tubs
containing pine shaving bedding. Each tub
measured 0·61 x 0·91 xO·23 m (length x width x
height) and each contained two water bottles for
ad libitum access to water. The animals were fed
a commercial guineapig chow (Continental Grain
Co., Chicago, IL) ad libitum. They were
maintained in an animal holding room with an
air exchange rate of 15 changes per hour of 100070
fresh air. The temperature was maintained at
22 ±1 °C with a relative humidity of 50 ±10%.
The light cycle was 12 h light (0700-1900 h), 2 h
dark (1900-0700 h) without twilight.
Experimental design
Mating groups of two sizes were studied in the
same size cage. Groups were composed of one
male and
3
females or one male and
6
females.
The cage area selected for study was
4636
cm2
which corresponds to the amount of floor area
required by 4 breeding animals, according to
USDA guidelines. It is also only slightly larger
than the space requirement for a group of 7
animals not designated as breeders but in the
same weight range as breeding animals.
The study cage was constructed of non-glare
black plastic and measured O· I x 0,6 x 0·23 m
(width x length x height). A J-type stainless steel
feeder was mounted in one corner next to a
470 ml water bottle. Groups of animals were
studied in the cage for 12 consecutive hours, after
which they were removed from the study cage
to allow for cleaning. The study cage did not
contain bedding, as this would alter the contrast
required for the optical quantitation of floor
space and activity.
During the day and evening illumination
conditions, the cage surface temperatures and
air temperatures were monitored using a mul-
tiple channel electronic thermometer with air
White, Balk
&
Lang
temperature and surface temperature probes. No
significant differences in surface temperatures
were noted between the centre of the cage floor
or peripheral sections of the cage wall. No
differences were noted between these measure-
ments and the ambient air temperature which
remained at 22
±
0,9
°C.
Equipment and measurements
The study employed a Videomex-III Multiple
Zone Motion Monitor (Columbus Instruments
International, Corp., Columbus, OH) con-
sisting of a computerized video digitalizer and
computer-controlled data acquisition and pro-
cessing system. A low-light television camera
(Videocon- TCIOS/N, RCA Closed-Circuit Video
Equipment, Lancaster, PA) capable of recording
in visible and infrared light, as well as in light
of low intensity, was substituted for the video
camera standard to the system. An 8 mm wide-
angle lens with an automatic iris was used.
Output from the camera was split to allow for
computerized acquisition and digitalization of
images during recording of non-digitalized images
on a lO-speed slow motion video recorder (Time
Lapse Video Recorder TC3920, RCA Closed-
Circuit Video Equipment, Lancaster, PA). The
storage of images on video tape allowed real-time
viewing (30 frames per second) of gated images
(one frame per second) and comparison of cage
space use patterns quantitated by the computer
with actual visualization of the activity through-
out the study period.
The video camera was mounted over the study
cage and the system focused on a computer-
generated grid superimposed on the cage. The
grid was divided into 15 grid subunits subdivided
into 1400 pixels each with each subunit corres-
ponding to 309 cm
2
of floor space. The food
and water containers were noted on the grid.
During operation, the system digitalized the
video image of the cage, calculating the area
occupied by animals in each subunit. Data was
accumulated over one-hour periods and ex-
pressed as the average number of pixels occupied
per grid subunit per hour. This measurement is a
direct expression of the average area of occupancy
Use of cage space by guineapigs 211
Fig. 1. Statistical comparisons were made by grouping data
from grid subunits. Comparison A compared peripheral
subunits to three central subunits while Comparison B
examined differences between several peripheral subunits and
the remaining eight.
per grid subunit. Illumination without shadows
was provided during daytime studies by overhead
fluorescent lighting supplemented by two 20-W
fluorescent fixtures mounted O' 86 m above and
to either side of the study cage. Nighttime
illumination was provided by four
IS-W
infrared
darkroom fixtures that emitted no light in the
visible spectrum. These fixtures were mounted
O'61 m above the four corners of the study cage.
All data acquired was converted to percentage
occupancy by dividing the average area of
occupancy for individual subunits by the total
area of occupancy of the cage
floor.
This
procedure corrected for variation in light intensity
between nighttime and daytime observations. A
correction was also made for sidewall reflection
caused by spherical aberrations in the lens
A
B
ABC
L M N
K
J
system. The system was automatically corrected
for reflections caused by the watering device and
feeder by calibration at the beginning of each
observation period
Following review of the data and videotape
records, data was combined across time and
between groups of grid subunits (Fig. 1).
Statistical comparisons were made by Chi square
analysis, with probability equal to or less than
0,05
being considered significant. Two groups
of comparisons were made. First, occupancy of
3 central grid subunits (representing 3/15 of a
floor area of the cage) was compared to
occupancy of the 12 peripheral grid subunits.
Second, as shown in Fig. I, 7 adjacent peripheral
grid subunits (representing
470/0
of the available
area) were compared to the remaining 8 subunits.
The effect of group size as well as time of day
was analysed.
Results
The mean percentage occupancy for the day and
nighttime observations as well as the combined
day and night observations for the two group
sizes of guineapigs in the 4636 cm
2
cage are
summarized in Table 1. The expected frequency
of occupancy for each grid subunit (assuming
that activity was evenly dispersed across all
surfaces of the cage) was approximately 6'7%.
The number of grid subunits scoring 6'
7070
or
greater occupancy ranged from 5 to 7 with the
remainder scoring less than that amount.
As can be seen, most of the occupancy
occurred in the cage periphery, primarily in
subunits B to H. As food and water occupied a
portion of grid subunit A, occupancy measure-
ments there may actually be higher considering
the availability of space in that subunit. Subunits
in the centre of the cage (M, Nand
0)
had
uniformly low percentages of occupancy. Grid
subunits L, K,
J
and I also had low occupancy,
with only a few exceptions. This pattern was
consistent in daytime and nighttime observations.
Statistical comparison of the average percentage
occupancy of peripheral and central subunits
(Table 1) confirmed the observations that
occupancy in the centre was significantly lower
212
Table 1. Guineapig space utilization by cage sector
Group composition Day Night
(12 h) (12 h) 24 h
3 females +I male
Comparison A'
Periphery (A-L) 98'6
c
,d 96'7d97'5d
Centre (M-O) 1·4 3,3 2'5
Comparison B
b
Areas (B-H) 74'9d87'6d82'2d
Areas (I
-0
+A) 25·1 12·4 17·8
6 females +I male
Comparison A
Periphery (A-L) 97'8d93'7d95'3d
Centre (M-O) 2·2 6,3 4,7
Comparison B
Areas (B-H) 74'9d87'6d82'2d
Areas
(1-0
+A) 25·6 21·6 23·2
Studies were conducted in a cage with 4636 cm
2
of floor
area, subdivided for scoring purposes into 15 subunits of
equal size (309' 7 cm
2).
Location of lettered subunits is
outlined elsewhere in the text. Cage sector refers to a group
of subunits spatially related to each other.
'Comparison A was between the mean total percentage
occupancy over the time periods listed in the peripheral 12
subunits (1st sector) to the 3 subunits comprising the centre
of the cage (2nd sector).
bComparison B was between the mean total percentage
occupancy over the time periods listed between 7 peripheral
contiguous subunits (3rd sector) and the remaining 8 subunits
in the cage (4th sector).
cMean percentage occupancy was calculated by dividing the
total area of a subunit occupied by guineapigs by the mean
total surface area of guineapigs in the cage multiplied by 100.
Area measurements were made in pixels.
dThe mean total percentage occupancy was significantly
different
(P5,
0.05) between sections of the comparison
within the time period.
than that in the periphery during the daytime and
nighttime. This also held true when the total 24 h
observation period was examined.
There was no statistically significant difference
in the pattern of peripheral vs central cage
occupancy when groups of 4 vs 7 guineapigs were
compared. In all cases, the combined observed
percentage occupancy in subunits
B
to H ranged
from
75lTJo
to
88lTJo
although their combined area
represented
47lTJo
of the total space.
Upon being placed in the study cage the
animals underwent a brief period of exploratory
behaviour that lasted from 30 to 45 mins in which
all areas of the cage were visited by the animals.
Marking of the corners of the cage by urination
and defecation also occurred as did exploration
White, Balk
&
Lang
of the food and water sources. Following this
period the animals reverted to the activity and
cage use patterns noted during the rest of the
study period.
Analysis of the video recordings revealed that
during day and night observation periods,
guineapigs in both groups tended to use only a
portion of the area available to them. In most
cases they avoided the centre of the cage,
preferring instead to back up against the cage
walls. They usually sought each other's company
and often lined up parallel to each other along
one wall. There was no prolonged period of
quiescence that could be associated with sleep.
The animals remained active throughout the day
and night, with individual periods of inactivity
lasting at most only a few minutes.
Social interactions apparently occurred during
both the day and night. No overtly aggressive
behaviour was noted during any study period
although behaviour with sexual implications was
noted throughout the entire day.
Eating and drinking occurred throughout both
the day and night observation periods, as did
defecation and urination. The animals generally
tended to urinate and defecate in one spot. This
was usually in the corner of subunit E but also,
less frequently, in the corners of subunits G and
K. Urination and defecation near the feeding and
watering devices occurred less frequently than in
the corners. The amount of food and water
consumed, as well as the amount of urination
and defecation, was not quantitated. Hence,
differences between these parameters during day
and night hours could not be assessed.
Discussion
This study supports previous observations that
both laboratory and wild guineapigs actively seek
out one another and prefer to form groups. The
results confirm and extend Nicholls' observations
of caged-housed guineapigs, i.e. that guineapigs
are active throughout the day and night and
have no apparent nocturnal or diurnal rhythm
(Nicholls, 1922). Activity was broken by small
rest periods averaging less than 5 min. During
these periods, the guineapigs were seen to lie on
Use of cage space by guineapigs
their sides with feet extended, in a very quiescent
state. This occurred most frequently during the
night.
Our data suggest that groups of guineapigs use
very little of the space available to them. They
prefer the sides of the cage and like to be close
to each other during activity and rest. The
observation that 75-88070 of cage occupancy
occurs in seven contiguous grid subunits along
two adjacent cage walls (an area representing
only 47% of that available) suggests under
utilization of the available cage floor area.
While it is difficult to explain the low
occupancy rate (between 12% and 25%) of the
rest of the cage, the consistency in this pattern
between groups of 4 and 7 guineapigs is
remarkable. Since the amount and pattern of
cage space use did not vary significantly with the
size of the group, one can conclude that
4636 cm
2
of floor area provides adequate space
for a breeding group of 7 guineapigs, even
though this is 40% less than current USDA
guidelines. It remains to be determined if further
reductions in floor area would alter behavioural
patterns and whether other modifying factors,
such as the presence of young or the availability
of shelter, will modify the use of cage floor area.
It could be argued that the study cage environ-
ment used in this investigation represented a
novel environment to the guineapigs since it was
smaller, black in colour, and contained no
bedding material compared with their home
cages. Certainly the initial exploratory behaviour
noted at the beginning of each study period
supports this idea but the rapid adaption to a
regular activity pattern that was remarkably
similar between study groups and conditions
argues against the notion that the animals were
reacting to it in a fashion that could be
interpreted as stressful or abnormal when
References
Besch EL (1980) Environmental quality within animal
facilities. Laboratory Animal Science 30, 385-398
213
compared with their home cage. Moreover,
frequent but unquantified observations of the
same groups of animals in their home cages
revealed similar cage use patterns. These observ-
ations also correspond to anecdotal reports by
others engaged in breeding and maintenance of
guineapigs.
The question of whether the patterns and
amount of space use within a cage of a given size
resembles that of guineapigs in their natural
habitat is difficult to answer and is perhaps
not germane. Guineapigs are reared in cages
and purpose-bred for laboratory research. The
amount of area available in the wild cannot be
provided in a laboratory setting. Moreover, we
do not know whether laboratory-reared guinea-
pigs exhibit the same behavioural patterns as wild
members of the genus Cavia.
The findings of the present study suggest that
the current guidelines for guineapig housing
based on area allocation per guineapig, cannot
be supported by the behavioural characteristics
of these animals or careful quantitation of their
patterns of cage space use. Additional studies
must be conducted for guineapigs, as well as
other laboratory animals, to develop more
accurate cage space requirements for each
species.
Acknowledgments
This work was supported in part by Grant
RR00469 from the USPHS, NIH as well as by
Charles River Laboratories, Inc. The authors
thank Mary Bartholomew for assistance with
statistical analysis.
This study was conducted in accordance
with the Animal Welfare Act of 1966 as
amended, and The Guide for the Care and
Use of Laboratory Animals DHHS 1985 (NIH
Publ. No. 86-23).
Code of Federal Regulations (February 1976) Title 9,
Animals and Animal Products, Sub-Chapter A-Animal
214
Welfare, Parts I, 2 and 3. United States Department of
Agriculture
Committee on Care and Use of Laboratory Animals, Institute
of Laboratory Animal Resources, National Research (1985)
Guide for the Care and Use of Laboratory Animals.
Washington, DC: National Academy of Sciences
Levine S (1985) Definition of stress? In Animal Stress (ed.
OP Moberg), pp. 51-70.
White, Balk
&
Lang
Nicholls EE (1922) Study of the spontaneous activity of the
guinea pig. Comparative Psychology 2, 303-330
Rood JP (1972) Ecological and behavioral of three genera
of argentine cavies. Animal Behavior Monographs 5,
1-83
Rood JP
&
Weir B (1970) Reproduction in female wild
guinea pigs. Journal of Reproduction Fertility 23,
393-409
