The Effects of Restraint Using Self-Locking Stanchions
on Dairy Cows in Relation to Behavior, Feed Intake,
Physiological Parameters, Health, and Milk Yield1,2
D. J. BOLINGER,3 J. L. ALBRIGHT, J. MORROW-TESCH,4
ABSTRACT
Holstein cows ( n = 64) ranging from peak to end
lactation were restrained in self-locking stanchions
(i.e., head locks) for approximately 4 h/d for four
periods in a modified switchback design. Milk yield,
milk fat percentage, somatic cell count, and dry mat-
ter intake and dry matter intake were unaffected by
restraint. Milk protein percentage was significantly
lower for cows that were restrained. Plasma cortisol
concentrations and the ratio of neutrophils to
mononuclear cells were not significantly different be-
tween restrained and unrestrained (control) cows. No
difference in the incidence of mastitis or other health
concerns was noted. Behaviorally, cows that were
locked in the stanchions spent significantly more time
lying after release from restraint. For cows that were
locked up, eating frequency over 24 h was signifi-
cantly reduced, but dry matter intake was not af-
fected. Total rumination frequency over 24 h was not
significantly different for cows that were restraubed;
however, cows that were restrained ruminated less
during the day following release. Grooming was consi-
dered to be a behavioral need and was significantly
increased during all times when cows were not locked
up. Grooming was also one of the first behaviors per-
formed following release. Acts of aggression were
elevated during all periods following restraint, but
oral behaviors, such as tongue playing and chewing
on objects, drinking behavior, and resting postures
were not affected. The use of self-locking stanchions
Received September 26, 1996.
Accepted April 9, 1997.
1Journal Paper Number 15224, Purdue University Agricultural
Research Programs.
2This investigation is part of the Indiana contribution to NC-119
Dairy Herd Management Strategies for Improved Decision Making
and NCR-131 Animal Care and Behavior.
3Current address: Clinton County Extension, 306 Elm Street,
St. John, MI 48879-2398.
4USDA-ARS Livestock Behavior Research Unit.
5Veterinary Clinical Sciences, School of Veterinary Medicine.
1997 J Dairy Sci 80:2411–2417
S. J. KENYON,5 and M. D. CUNNINGHAM
Department of Animal Sciences,
Purdue University, West Lafayette, IN 47907
did not appear to affect substantially the overall well-
being of the cow.
( Key words: behavior, physiological parameters, re-
straint, well-being)
INTRODUCTION
The facilities and equipment that are used to con-
duct daily dairy cattle management routines vary
widely. For instance, one method is to restrain dairy
cows to conduct routine management practices such
as artificial insemination, administration of treat-
ments such as bST, and veterinary checks. The de-
velopment of large herds coupled with free-stall hous-
ing and fence-line feeding systems has made obsolete
the traditional techniques of manual stanchions or
single chutes. To restrain a large number of cows
simultaneously, the self-locking stanchion at fence-
line mangers was developed in the 1970s. The utiliza-
tion of self-locking stanchions began in the far
western US, where large dairy herds are prevalent,
and has spread throughout the dairy industry in
North America.
Self-locking feed stanchions ease labor and save
time in handling cattle (11). Although no official
guidelines exist for the proper use of self-locking
stanchions, there are unofficial recommendations.
McFarland ( 8 ) suggested that self-locking stanchions
should be used for the brief restraint of a group of
cows for routine examinations such as pregnan-
cy checks, artificial insemination, and treatments.
Shipka and Arave ( 1 6 ) noted that group restraint for
4 h is an acceptable practice in mild climates when
the cows have healthy feet and feed is readily availa-
ble. According to Winchester and Morris (18),
prolonged restraint of any kind should be avoided
unless it is essential to the research objectives and
protocol.
The number of stanchions or feeding spaces
provided per cow varies greatly. McFarland ( 8 )
stated that feeding barrier openings are commonly
2411
2412 BOLINGER ET AL.
more numerous than free stalls, which permits
producers to overstock the free-stall space and still
have the capability to restrain all cows at once.
McFarland ( 8 ) also noted another common housing
design that provides three rows of free stalls per one
row of feeding or restraint stanchions and that results
in 25% more free stalls than stanchions. A concern
regarding this design is the inability of all cows to eat
at once, which they often attempt to do at the feeding
time and following milking.
Pirckelmann ( 1 3 ) reported that all feed barriers
and mangers should provide free access to feed
without risk of injury. Feed mangers should also pro-
vide undisturbed feed intake and minimize feed
wastage. The Midwest Plan Service ( 1 0 ) has recom-
mended that mangers with locking stanchions should
include the following features: 1 ) individual or group
release, 2 ) self-locking stanchions, and 3 ) quick
release for fallen cows. McFarland ( 8 ) has empha-
sized the necessity for quick and easy release of fallen
cows. The Midwest Plan Service ( 1 0 ) also recom-
mended that stanchions provide at least 62 cm of
space for mature Holstein cows and, more recently,
has suggested that 46 to 62 cm per cow appears to be
adequate when a TMR is available most of the day.
These recommendations must be continually evalu-
ated because barns are designed for higher yielding
cows.
Although the use of self-locking stanchions has
been widely accepted by the industry as a humane
method of restraint, there are few guidelines ( 5 ) for
their use and the scientific analyses of their impact.
The most likely misuse of properly constructed self-
locking stanchions is the inadvertent or intentional
restraint of cows for an excessively long time.
Various criteria have been proposed to identify in-
appropriate management and housing conditions for
agricultural animals. Animal well-being has both
physical and psychological components, and, thus, no
single objective measurement exists that can be used
to evaluate the well-being of an animal with regard to
a particular system of agricultural animal production
(3, 5).
The objective of this study was to analyze the
effects of prolonged restraint of lactating dairy cows
in self-locking stanchions on the parameters of perfor-
mance and well-being. Similar cooperative studies
were initiated at Michigan State University and Utah
State University. To comprehend fully the implica-
tions of this practice, it is important to understand
both normal and unusual behavior as well as the
standards for well-being for lactating dairy cows.
Journal of Dairy Science Vol. 80, No. 10, 1997
MATERIALS AND METHODS
General
At the Purdue University Dairy Research Center
near West Lafayette, Indiana, 64 lactating Holstein
cows were housed in free stalls with self-locking stan-
chions at the manger for 4 consecutive wk in Febru-
ary and March 1994. Cows were milked twice daily.
Milk yield of the cows ranged from 43.6 kg/d at peak
lactation to 19.2 kg/d near dry-off. Mean daily milk
yield at the beginning of the trial was 28.2 kg. Two
groups of 32 cows each, across the feed alley from
each other, were divided based on daily milk yield
(low vs. high), and TMR were balanced and fed in
groups relative to milk yield. The low and high milk
yield groups were then divided in half and blocked by
parity, days in lactation, and daily milk yield, result-
ing in a total of four treatment groups of 16 cows
each. Treatments were applied in a modified switch-
back design over four 7-d periods. Periods 1 and 3
were randomly assigned as control periods for one
high and one low yield group, respectively; periods 2
and 4 were assigned as control periods for the remain-
ing two groups. During control periods, the normal
protocol for dairy herd management at Purdue
University was followed. This protocol included short-
term lockups, not exceeding a total of 1.5 h per period,
for routine reproductive management practices such
as artificial insemination and veterinary examina-
tions. During treatment periods, cows were locked up
for 4 h/d beginning at 0700 h, which was the time of
feeding. This 4-h lockup met the maximum permissi-
ble time of restraint allowed by the Purdue Animal
Care and Use Committee. Cows were handled in ac-
cordance with the Guide for the Care and Use of
Agricultural Animals in Agricultural Research and
Teaching ( 5 ) .
Individual daily milk yields were recorded electron-
ically (Alfa Laval Alpro; Alfa Laval, Kansas City,
MO) in the milking parlor. A composite milk sample
from the a.m. and p.m. milkings was taken from each
cow on the 4th d of each period. The milk sample was
analyzed for milk fat percentage, milk protein per-
centage, and SCC by the Indiana State Dairy
Association Milk Laboratory (West Lafayette, IN).
Health disorders were recorded at the time of occur-
rence. Environmental temperature was noted by a
thermometer located in the free-stall barn at the time
of restraint.
Physiology
The amounts of feed offered and orts were recorded
daily for all treatment groups. Blood samples were
 RESTRAINT USING SELF-LOCKING STANCHIONS
drawn from each cow that was restrained in head
locks via coccygeal venipuncture during lockup on d 3
and 6 of each period. The unrestrained group was
held, bed during a 15- to 30-min period, and then
released. All blood samples were collected using
heparinized vacutainer tubes for later processing to
obtain plasma. Whole blood was used to make
differential leukocyte determinations. One-hundred
leukocytes were counted on a blood smear stained
with leukostat (Fisher Scientific, Pittsburgh, PA).
Blood was then centrifuged, and plasma was sepa-
rated, frozen (–20°C), and stored until assays were
conducted. A radioimmunoassay (Diagnostic Pro-
ducts, Los Angeles, CA) was used to quantify plasma
cortisol concentrations. The assay was sensitive to 0.2
mg/dl. Intraassay coefficients of variation averaged
4.1%, and the interassay coefficients of variation
averaged 8.4%.
Behavior
Four 24-h observation periods were conducted
starting at 0700 h on the 5th d of each period. At
20-min intervals each cow was observed until the
behavior was obvious. The location, posture, and
primary activity of each cow were recorded on data
sheets. Observations of location included manger
area, free stall, free-stall alley, water area, holding
yard, and milking parlor. Observations of posture
included standing, lying sternally recumbent, and ly-
ing laterally recumbent. Observations of activity in-
cluded ruminating, eating, drinking, walking, various
indicators of resting [e.g., lying sternally recumbent
on left or right side; lying laterally recumbent
(horizontal) on left or right side; head curled back on
neck, flank, or leg; and head resting on floor or pipe],
repetitive favoring or shifting of feet, as well as
grooming, agonistic (dominant and submissive) in-
vestigation, and oral (tongue playing, licking, chew-
ing objects) behaviors. Observations of sexual be-
havior (e.g., mounting, standing to be mounted, chin
resting) were recorded; however, these behaviors
were not considered because of the variation in
reproductive status among treatment groups. Obser-
vations of posture and activity were not recorded
during the milking time.
On d 5 of the final period, a videotape recorded the
cows under restraint at 3 h into the lockup. The cows
were also recorded for several minutes after their
release. The videotape was used to make only general
observations about the behaviors of the cows that
might not have been viewed at the 20-min intervals.
2413
TABLE 1. Production data by treatment group.
Restrained Control
X SE X SE
Milk yield, kg 28.9 0.7 28.9 0.7
Milk fat, % 3.67 0.23 3.68 0.23
Milk protein, % 3.19 0.12a 3.27 0.12b
SCC, cells/ml 158,000 14,000 196,000 14,000
DMI, kg/d 23.0 0.3 23.0 0.3
a,bMeans in rows with different superscripts differ ( P < 0.05).
Statistics
Statistical analysis was performed using SAS ( 1 5 )
and a modified switchback design; cows were blocked
by group location. Cow variation was removed by the
modified switchback design. Analysis of variance and
the Student-Newman-Keuls test were utilized as con-
servative methods to detect significance between
treatment means. Least squares means were used to
test for differences in levels of significance within
factors. Standard errors were used only as a method
of showing variability. All quantitative data analyses
were conducted on mean values by treatment group
for each period. Behavioral observations were grouped
to accommodate more accurate analyses and to
minimize confounding data. Because cows ate miner-
als a minimal number of times, observations about
minerals were combined with the consumption of the
TMR. The limited frequency of drinking from the floor
or drinking urine was added to the frequency of
drinking water for a total drinking category. Self-
grooming and grooming another were combined into
one grooming category. Dominant and submissive be-
haviors were grouped as agonistic activity. Tongue
rolling and chewing or licking objects were grouped as
oral activities. Location was removed from the data
file for analysis because it confounded many activi-
ties. The multiple postures for lying were combined
into one category. Time was analyzed based on total
observations for the 24-h period and the period follow-
ing lockup. Time following lockup was further divided
into time of release from stanchions to 1900 h and
1920 h until 0720 h the next morning.
RESULTS
Restraint of cows in self-locking stanchions for 4
consecutive h had no effect on the milk yield
parameters that were measured, except for the pro-
tein composition of the milk (Table 1). Milk yield was
unchanged by restraint. Mean milk yields of the
treatment groups were both equal to 28.9 kg, which
was highly unlikely. Mean milk fat percentage was
Journal of Dairy Science Vol. 80, No. 10, 1997
2414 BOLINGER ET AL.

TABLE 2. Mean plasma cortisol concentrations and the ratio of
neutrophils to mononuclear cells (N:M) by treatment group.
Restrained1 Control2
X SE X SE
Cortisol, mg/dl 12.7 1.7 15.5 1.7
N:M 0.77 0.03 0.75 0.03
1Blood samples were taken from each cow restrained in self-
locking stanchions for 4 h via coccygeal venipuncture on d 3 and 6
of each period.
2Nonrestrained cows were held, bled during a 15- to 30-min
period on d 3 and 6 of each period, and then released.
unaffected as well; the mean for the control cows was
3.68%, and the mean for the restrained cows was
3.67%. Mean protein percentage of milk was lower for
restrained cows (3.27% vs. 3.19%; P < 0.03). The SCC
exhibited limited significance; the mean value for
control cows was 196,000 somatic cells/ml, and the
mean value for restrained cows was 158,000 somatic
cells/ml as a result of lockup ( P < 0.10) (Table 1).
Only two clinical cases of mastitis occurred during the
study, one in each group. Mean daily DMI also
showed no significant change because of lockup. The
mean DMI for control cows equaled 23.0 kg, and the
mean DMI for restrained cows was also 23.0 kg (Ta-
ble 1).
Physiological indicators of well-being also re-
mained unaffected by restraint (Table 2). Mean
plasma cortisol concentrations equaled 15.5 mg/dl for
control cows and 12.7 mg/dl for restrained cows ( P <
0.27). The mean ratio of neutrophils to mononuclear
cells also was unaffected by restraint (Table 2). The
mean ratio of neutrophils to mononuclear cells was
0.75 for the control cows and 0.77 for the restrained
cows. Monocytes and lymphocytes were analyzed
together as mononuclear cells because these cells
could not be differentiated under the light microscope
(11).
Behavioral analyses (Table 3 ) showed that cows
that were locked up opted to rest more often ( P <
0.01) during the time that they were not locked up
than did the control cows during the same period. The
restrained cows also spent more time lying from 1920
to 0720 h. Stanchion restraint did result in a change
in time spent eating. The cows that were locked up
had a lower ( P < 0.01) eating frequency over 24 h
than did control cows. Ruminating behavior was
decreased ( P < 0.05) over all time periods after
lockup; however, ruminating behavior was unaffected
over 24 h. Time spent drinking was unaffected by
lockup in all applicable periods. Cows that were
locked up spent more ( P < 0.05) time grooming than
did control cows over all time periods. Agonistic be-
havior increased ( P < 0.01) during the remaining
hours of the day following lockup. A higher ( P < 0.05)
frequency in agonistic encounters was noted from
1920 to 0720 h as well. Oral activities were not
significantly affected by lockup. Restraint seemed to
increase ( P < 0.05) the frequency that cows rested
with their heads on their necks or flanks, but only
shortly following lockup. The numbers of cows that
rested with their heads on an object were unaffected
by lockup.

TABLE 3. Frequency (percentage) of postural and behavioral observations.

24-h Total 20 h Not locked up1 ≈7 h Following lockup2 1920 to 0720 h
Behavior Restrained Control Restrained Control Restrained Control Restrained Control
X SE X SE X SE X SE X SE X SE X SE X SE
Lying NA3 NA 56.0** 0.8 50.6 0.7 46.1** 0.9 37.7 0.9 65.9 1.2 63.6 1.2
Eating 28.2** 0.5 29.1 0.5 28.6 0.5 28.3 0.5 30.7 0.9 31.2 0.9 26.4 0.6 25.4 0.6
Ruminating 57.0 0.5 57.5 0.6 53.9** 0.6 58.2 0.6 50.5** 0.9 55.3 1.0 57.3** 0.8 61.0 0.8
Drinking NA NA 2.8 0.2 2.5 0.2 3.3 0.5 2.8 0.5 2.2 0.2 2.1 0.2
Grooming NA NA 6.8** 0.3 4.1 0.3 8.0* 0.6 4.9 0.6 5.7** 0.4 3.4 0.3
Agonistic activities
(aggression) NA NA 0.8** 0.1 0.4 0.1 0.8 0.2 0.3 0.2 0.8* 0.1 0.4 0.1
Oral activities NA NA 1.1 0.2 1.4 0.2 1.4 0.3 1.5 0.3 9.7 0.2 1.2 0.2
Head on neck
or flank NA NA 2.4 0.3 1.9 0.3 2.3* 0.3 1.4 0.3 2.5 0.4 2.4 0.4
Head on object NA NA 3.6 0.3 3.3 0.3 3.0 0.4 2.7 0.4 4.3 0.4 3.8 0.4
1End of lockup until 0720 h.
2End of lockup until 1900 h.
3Not applicable; activity was affected by restraint.

*P < 0.05.
**P < 0.01.

Journal of Dairy Science Vol. 80, No. 10, 1997

 RESTRAINT USING SELF-LOCKING STANCHIONS
Environmental temperatures ranged from –15 to
3°C, and the weather was generally calm, but cold, for
the duration of the study.
DISCUSSION
The lack of a significant effect of the stanchion
restraint on milk yield and feed intake of the cows is
supported by the results of a study by Shipka and
Arave ( 1 6 ) who found no effect when cows were
locked in self-locking stanchions for 4 h. A second
trial was also conducted by Arave et al. ( 1 ) during
summer in Utah. Mean milk yield was much higher
( P < 0.0001) during weeks when cows were not re-
strained than during weeks when cows were re-
strained (37.2 ± 0.20 vs. 34.3 ± .20 kg/d per cow). In
that summer trial, feed intake was unaffected. Arave
et al. ( 1 ) concluded that extended lockup was more
detrimental during hot weather than during mild or
cold weather.
Although milk fat percentage was unaffected by
lockup, milk protein percentage was affected in this
study. The decline in milk protein percentage for
restrained cows is difficult to explain. No similar
research has been conducted that evaluates the effect
of restraint on milk components, and no previous
study exists with which to compare our results.
Health parameters appeared to be unaffected by
lockup. Only two clinical cases of mastitis were
documented during the study. Both cases occurred
during the third trial period, one case from the group
under control conditions and one from the group un-
der lockup. The SCC was not affected by restraint,
which is in agreement with the results of Shipka and
Arave (16). A single cow was removed from the trial
on d 4 of the first period and returned to her treat-
ment group on the 1st d of the second period. This cow
had symptoms of hardware disease and remained in a
hospital pen for 4 d. The cow completed the final
three periods of the trial with no further complica-
tions.
Physiological indicators of stress were unaffected
by restraint. In this study, cortisol and the ratio of
neutrophils to mononuclear cells showed no signifi-
cant change as a result of restraint. Ratios of neu-
trophils to mononuclear cells were within the accepta-
ble range for nonstressed animals as indicated by
Swenson (17), who recommended a ratio near 0.7. In
a previous study, cortisol concentration increased dur-
ing lockup by a mean of 5.3 ± 0.9 ng/ml during a
spring trial and by 9.1 ± 1.1 ng/ml during a summer
trial ( 2 ) . Corresponding means for cortisol concentra-
tions were 9.4 ± 0.9 versus 13.8 ± 1.1 ng/ml during
2415
lockup in spring and summer, respectively, and 14.7 ±
0.9 and 22.9 ± 0.9 ng/ml following lockup in spring
and summer, respectively. These differences were
highly significant ( P < 0.0001).
Cows that were locked up appeared to lie more
frequently ( P < 0.01) during the hours following
lockup than did the control cows. This elevated fre-
quency of lying for restrained cows apparently oc-
curred most between 1920 and 0720 h. This tendency
might have been delayed compensation for lying
deprivation during lockup. This result was similar to
the lying deprivation study of Metz ( 9 ) . Eating be-
havior appeared to have been significantly affected by
lockup. Restrained cows ate less frequently per 24 h
than did control cows, but eating frequency was un-
affected during the period when cows were unlocked.
This result suggests that frequency of eating was
reduced by restraint. Rumination also appeared to be
affected by the use of self-locking stanchions. Re-
strained cows ruminated less ( P < 0.05) throughout
the day following release from the stanchions;
however, there was no significant difference in rumi-
nation frequency for 24 h. This result suggests that
the restrained cows ruminated more during lockup
and compensated by ruminating less during the re-
mainder of the day.
Drinking behavior did not appear to be affected by
lockup. After release from the stanchions, the cows
simply might have consumed a greater volume of
water per drinking bout rather than drinking more
frequently. Arave et al. ( 1 ) found that restrained
cows were near water more often than were control
cows, but those researchers did not note an increase
in drinking frequency.
Even though self-locking stanchions physically
prevented grooming behavior, grooming behavior in-
creased ( P < 0.05) for restrained cows throughout all
periods following lockup. A possible explanation is
that less frequent grooming might be required for
maintenance than for compensation of deprived or
neglected grooming. Although grooming is practiced
infrequently, it is nevertheless important. Based on
these data and videotape observations following
lockup, grooming was one of the first behaviors per-
formed following release from the stanchions. These
results suggest that grooming is a behavioral need of
the cow.
The importance of grooming has been shown previ-
ously. In a study by Krohn ( 7 ) , nearly all cows
received social grooming; however, only about three-
quarters of the cows performed social grooming ( 7 ) .
Grooming among all individual cows, as well as the
low frequency of agonistic behavior in this group,
Journal of Dairy Science Vol. 80, No. 10, 1997
2416 BOLINGER ET AL.
illustrates the importance of affiliative behavior in a
well-established group of cows in an environment
with minimal competition for space ( 7 ) . Lying depri-
vation and social isolation were also shown to in-
crease the frequency of displacement grooming in
tethered heifers ( 6 ) and cows (12).
The frequency of aggressive behavior was higher
( P < 0.01) among restrained cows during the remain-
ing hours of the day following lockup. The level of
agonistic behavior was also elevated from 1920 to
0720 h, and perhaps indicated reduced well-being
because of the use of self-locking stanchions. Gonyou
( 4 ) suggested that increased aggression is an indica-
tor of frustration that reduces well-being.
Oral activities, including tongue rolling (or tongue
playing) and chewing or licking on objects, appeared
to be unaffected by lockup. Tongue rolling and some-
times chewing or licking on objects are considered
stereotypies ( 3 ) . The lack of an effect of lockup on the
frequency of oral behavior suggests that lockup is not
stressful enough to produce stereotypical oral be-
haviors.
Restrained cows tended to rest their heads on their
necks or flanks more frequently ( P < 0.05) than did
control cows during the hours immediately following
lockup. This tendency was in contrast to the data that
showed that restrained cows do not lie more during
the same time period; however, restrained cows did
ruminate less. Consequently, restrained cows spent
their lying time following lockup resting, and control
cows ruminated during that time. The resting of
heads on objects appeared to be unaffected by lockup.
This finding was different from that of Arave et al.
( 1 ) . Arave et al. ( 1 ) found that restrained cows spent
more time with their heads resting on the floor, and
control cows rested their heads curled back on their
flanks more often. Arave et al. ( 1 ) also noted that
control cows spent more time near the feed manger,
in free stalls, and in the exercise lot than did re-
strained cows.
Videotape recordings of the behavior of cows from 3
to 4 h after lockup and at the time of release revealed
several observations that would be difficult to analyze
fully under this protocol, but some general observa-
tions can be made. Several cows that had been stand-
ing for at least 3 h showed uneasiness by repetitively
shifting their BW between rear feet. The extent of
this shifting might not be accurately evaluated by the
24-h behavioral observations. The low frequency of
observations at 20-min intervals resulted in the ex-
clusion of a large proportion of this activity. Shipka
and Arave (16), who observed cows at 20-min inter-
vals, did not indicate an increase in feet shifting. The
Journal of Dairy Science Vol. 80, No. 10, 1997
videotape in our study also revealed that many of the
restrained cows stood with their heads hung low;
appearance was depressed, and cows had drooping
ears and glazed eyes. The cows might have been
showing signs of fatigue. The videotape also captured
many of the cows ruminating at 3 h into the lockup.
Videotape recording at the time of release of the
cows from the self-locking stanchions also provided
some insight into the deprivations to cows during
restraint. A substantial number of the cows engaged
in self-grooming behavior. Most of those cows licked
their sides or flanks, and a few rubbed against gates;
there was no known external parasite problem. Many
of the cows immediately walked to the water trough
or to the free-stall area. A large amount of congestion
occurred near the water area, which also served as a
passageway to the free-stall area. Because of the con-
gestion at the water area, some butting and other
signs of aggression were observed. Gonyou ( 4 ) has
suggested that aggression in pigs can be a sign of
frustration and discomfort and indicated poor well-
being. Few cows elected to lie down immediately in
the free stalls; however, within 20 min after release
from the stanchions, a majority of the cows were
lying. One cow backed into her free stall.
CONCLUSIONS
The use of self-locking stanchions for up to 4 con-
secutive h did not seriously affect the well-being of
lactating dairy cows. Yield parameters, specifically
milk yield and milk fat percentage, were unaffected
by this method of temporary restraint. Milk protein
percentage may be reduced by using self-locking stan-
chions; however, more research is needed to provide
explanation and further support for such a claim.
Feed intake on a DM basis and SCC were also un-
altered by lockup.
With no effect on plasma cortisol, the ratio of neu-
trophils to mononuclear cells, or yield parameters and
with few significant changes in behavior, the use of
self-locking stanchions for up to 4 consecutive h does
not appear to compromise the well-being of the cow.
These same effects, however, could indicate the abil-
ity of cows to compensate for the effects of lockup.
More investigation into the 4-h lockup period needs to
be completed. Many short-term effects of lockup, such
as shifting of feet, have yet to be studied fully.
ACKNOWLEDGMENTS
The authors express sincere appreciation to Erika
Berg for data entry and for reading blood slides and to
Carol Reisert for statistical analyses. We thank Rod
Allrich and Jim Callahan for their suggestions on the
manuscript.
 RESTRAINT USING SELF-LOCKING STANCHIONS
REFERENCES
1 Arave, C. W., D. Bolinger, M. P. Shipka, and J. L. Albright.
1996. Effect of extended lock-up of lactating cows on milk
production, feed intake, and behavior. J. Anim. Sci. 74(Suppl.
1):43.(Abstr.)
2 Arave, C. W., M. L. Shipka, J. Morrow-Tesch, and J. L. Al-
bright. 1996. Changes in serum cortisol following extended lock-
up time of lactating cows. J. Dairy Sci. 79(Suppl. 1):
191.(Abstr.)
3 Fraser, A. F., and D. M. Broom. 1990. Farm Animal Behaviour
and Welfare. 3rd ed. Bailliere Tindall, Philadelphia, PA.
4 Gonyou, H. W. 1994. Why the study of animal behavior is
associated with the animal welfare issue. J. Anim. Sci. 72:2171.
5 Guide for the Care and Use of Agricultural Animals in Agricul-
tural Research and Teaching. 1997. Fed. Am. Soc. Food Anim.
Sci., Savoy, IL. (in press).
6 Jensen, M. B. 1995. The effect of age at tethering on behavior of
heifer calves. Appl. Anim. Bahav. Sci. 43:227.
7 Krohn, C. C. 1994. Behavior of dairy cows kept in extensive
(loose housing/pasture) or intensive (tie stall) environments.
III. Grooming, exploration and abnormal behavior. Appl. Anim.
Behav. Sci. 42:73.
8 Metz, J. H. 1984. The reaction of cows to a short-term depriva-
tion of lying. Appl. Anim. Behav. Sci. 13:301.
9 McFarland, D. F. 1995. Freestall housing guidelines: design
details to enhance management. Page 81 in Proc. 2nd Western
2417
Large Herd Dairy Manage. Conf., Las Vegas, NV. Dairyman
Magazine, Corona, CA.
10 Midwest Plan Service. 1985. Dairy Housing and Equipment
Handbook. 4th ed. Iowa State Univ. Press, Ames.
11 Midwest Plan Service. 1995. Dairy Freestall Housing and
Equipment. 5th ed. Iowa State Univ. Press, Ames.
12 Munksgaard, L. 1994. Methods for assessment of stress in
cattle: behavior, pituitary-adrenal-reactivity and growth hor-
mone concentration. Ph.D. Thesis, Royal Veterinary and
Agricultural Univ., Denmark.
13 Pirckelmann, H. 1995. Feed bunk and feeding equipment de-
sign for cattle. Page 136 in Proc. Anim. Behav. Design Livest.
Poultry Systems Int. Conf., Northeast Reg. Agric. Eng. Serv.,
Ithaca, NY.
14 Roth, J. A. 1985. Cortisol as a mediator of stress-associated
immunosuppression in cattle. Page 225 in Animal Stress. G. P.
Moberg, ed. Am. Physiol. Soc., Waverly Press, Bethesda, MD.
15 SAS User’s Guide Statistics. Version 6, 4th ed. 1989. SAS
Inst., Inc., Cary, NC.
16 Shipka, M. P., and C. W. Arave. 1995. Influence of extended
manger lock-up on cow behavior and production factors in dairy
cattle management. J. Anim. Sci. 73(Suppl. 1):310.(Abstr.)
17 Swenson, M. J., ed. 1977. Duke’s Physiology of Domestic
Animals. 9th ed. Cornell Univ. Press, Ithaca, NY.
18 Winchester, C. F., and M. J. Morris. 1956. Water intake rates of
cattle. J. Anim. Sci. 15:722.
Journal of Dairy Science Vol. 80, No. 10, 1997