Journal of Oral Rehabilitation 2001 28; 328334
The effects of chewing rates on mandibular kinematics
G. S. THROCKMORTON*, B. H. BUSCHANG*, H. HAYASAKI & T. PHELAN*
College of Dentistry, The Texas A&M University System Health Science Center, Dallas, TX, U.S.A. and Faculty of Dentistry, Kyushu University,
Fukuoka, Japan
SUMMARY The purpose of this study is to provide
basic understanding of how the speed of chewing
affects masticatory jaw kinematics. Twenty-six
healthy subjects (236 25 years of age) chewed a
standardized bolus of gum at fast (100 cycles s)1),
habitual and slow (50 cycles s)1) rates. The rates
were controlled with a metronome and the order of
rates was randomized for each subject. An optoelec-
trical system independently recorded head and jaw
movement. Special computer programs identied
representative cycles for each subject and computed
various aspects of jaw movement. Multilevel statis-
tical procedures were used to compare cycle varia-
bles among the three rates, estimate variability and
model jaw movements. Maximum ranges of antero-
Introduction
A number of studies have examined how chewing
cycles change when subjects alter their normal chewing
rate. When subjects are asked to chew at a faster than
normal rate not only are the total duration of the cycles
reduced, but vertical and lateral excursions of the
mandible are reduced (Morimoto et al., 1984; Plesh
et al., 1987, 1988, 1993). Such studies indicate that
cycle duration was reduced, at least in part, by reducing
the 3-D pathway taken by the mandible during the
chewing cycle.
Chewing cycle duration could also be reduced by
increasing mandibular velocity. However, previous
studies do not provide a clear indication as to whether
mandibular velocity increased during all or part of a
cycle. Morimoto et al. (1984) reported greater
mandibular velocities, but only for very fast cycles
(>3 Hz). Finally, there have been no studies that
characterize how chewing cycles change when
2001 Blackwell Science Ltd
*Baylor
posterior (AP), vertical and lateral jaw excursions
were signicantly less for the fast than the habitual
or slow rates. While the shape of 3-D pathway was
similar for the three rates, the perimeter of the
pathway was signicantly shorter for fast chewing
cycles. Maximum AP, vertical, lateral and total 3-D
jaw velocities were signicantly different among the
three rates. Between cycle variation in cycle dur-
ation and jaw excursion were least during fast
chewing and the greatest during slow chewing;
variability in maximum velocity was similar for
the three rates.
KEYWORDS : jaw kinematics, chewing rates, human,
chewing cycles
subjects are asked to chew more slowly than their
habitual rate.
The ability of subjects to control mandibular move-
ments during mastication has also received relatively
little study. Plesh et al. (1988) reported that mandibular
excursions were less variable in fast chewers than slow
chewers and that automatic chewing cycles were less
variable than voluntary chewing cycles. In addition,
subjects more accurately matched chewing frequencies
to the metronome when it was set close to their
habitual chewing rate. Within subject variability of
chewing cycles for rates other than the habitual rate
have not been evaluated.
The purpose of this study was to determine whether
subjects chew at different speeds by altering the length
of their 3-D pathway, by changing mandibular velocity
during all or part of the cycle, or both. In addition, we
examined whether chewing speed affects the within-
subject variability of cycle shape, cycle duration and
mandibular velocity.
328
1 EFFECTS OF CHEWING RATES ON MANDIBULAR KINEMATICS
Materials and methods
Subjects
Twenty-six healthy subjects (236 25 years of age)
were recruited among students, faculty and staff of Texas
A&M University System Health Science Center, Baylor
College of Dentistry. Half of the subjects were male and
half female. All subjects had a Class I normal occlusions
and no subjects had periodontal disease, signs or
3 symptoms of temparo-mandibular dysfunction (TMD),
extensive dental work, or were currently undergoing
dental treatment. Informed consent, as established by
the Institutional Review Board at Texas A&M Baylor
College of Dentistry, was obtained from each participant.
For this within-subject design each subject served as
their own control as chewing speed was changed.
Recording jaw kinematics
A rigid eyeglass frame (415 g) with six infrared-emit-
ting diodes was strapped to the subject's head to track
head movements. A single infrared-emitting diode
(12 g) was taped to the subject's chin to track mandib-
ular movements during the chewing cycle. Movements
of the diodes during mastication were tracked in 3-D
space using an Optotrak computer system. X, Y and Z
coordinates of each diode were recorded at 100 Hz in
real time. After recording, movement of the head diodes
was subtracted from movement of the chin diode to
derive 3-D mandibular movements. A special probe was
used to digitize the positions of right and left tragus and
orbitale relative to the head rigid body, allowing all
mandibular movements to be oriented relative to the
Frankfort horizontal.
Each subject was seated in an upright position
approximately 2 m in front of the Optotrak cameras.
Subjects were allowed to accustom themselves to the
recording apparatus by chewing a piece of gum prior to
recording.
Chewing tasks
Gum was used as the test food because it produces more
consistent chewing patterns than do natural foods
(Jemt et al., 1979). The size of each piece of gum was
standardized by forcing it into a specially built mould
(900 mm3) producing a bolus weighing 25 g. Subjects
were asked to initially place the bolus of gum on his/her
2001 Blackwell Science Ltd, Journal of Oral Rehabilitation 28; 328334
329
tongue and then, when recording started, to chew the
gum naturally. Two chewing sequences of 2025 cycles
were timed for each subject and the duration of these
two sequences was then divided by the total number of
cycles to establish each subject's automatic habitual
chewing rate.
Each subject was then asked to chew six additional
sequences with the gum, two at their habitual rate, two
at a fast rate (100 cycles min)1) and two at a slow rate
(50 cycles min)1). The different sequences were con-
ducted in random order. During each sequence the
subject was asked to match their chewing rate to a
metronome. Jaw kinematics was recorded for 2025
cycles of each of the six chewing sequences.
Kinematic analysis
A special computer program written by one of us (H.H.)
used the chin marker's vertical coordinates to identify
chewing cycles within each chewing sequence. The
program identied all chewing cycles that were more
than 300 ms in duration and had a vertical excursion of
at least 3 mm. The rst cycle was discarded from further
analysis because transferring the gum from the teeth
results in an atypical cycle. The duration and maximum
excursive ranges for each of the remaining cycles
during each chewing sequence were determined,
allowing the calculation of average cycle duration and
mandibular excursive ranges in the anteroposterior
(AP), vertical and lateral directions for each sequence.
Each cycle was given a standard score based upon its
deviation from these averages and the 10 cycles with
the least deviation were selected as the most represen-
tative cycles for that chewing sequence. For each
subject further analysis was based upon the most
representative cycles from their six chewing sequences
(20 habitual, 20 fast and 20 slow cycles).
A second computer program (also by H.H.) calculated
the duration of each of the representative cycles, as
well as the length of the total 3-D pathway of the chin
marker, maximum excursive ranges along the AP,
vertical and mediolateral axes and maximum cycle
velocities. In order to compare the shapes of the cycles,
chin position was plotted for each 5% of cycle duration
during opening and each 5% of cycle duration during
closing, a total of 41 points.
The data were evaluated using multilevel statistical
models (Strenio et al., 1983; Goldstein, 1986, 1987).
The estimation procedures used iterative generalized
330 G . S . T H R O C K M O R T O N et al.

least squares. A three-level model, with variation Table 1. Total, opening and closing duration (ms) for slow,
hierarchically partitioned (i) between subjects, (ii) habitual and fast chewing cycles

between trials within subjects, (iii) between cycles

Total Opening Closing
within trials, was used to evaluate systematic rate Chewing
differences in cycle duration, excursive ranges and rate Mean s.e. Mean s.e. Mean s.e.
maximum velocities. Each analysis was based on a Slow 11490 210 5482 206 6000 172
sample of 520 (26 subjects 2 trials 10 repeats) data Habitual 9166 186 4257 125 4909 120
points. Eighth-order polynomials were used to sepa- Fast 6051 24 2698 96 3354 95
rately model the horizontal, vertical and lateral changes
s.e. standard error.
in jaw position as a function of cycle duration. The
polynomials were tted using four-level models, which
included the aforementioned three levels of variation
plus within cycle variation.

Results

Cycle duration

As expected there were signicant differences in total

cycle duration among the three chewing rates (Table 1;
Fig. 1). Duration of the fast chewing cycles was
approximately 30% less than duration of the habitual
cycles (605 24* versus 917 186* ms, respectively). Fig. 1. Average (s.e.) cycle duration for fast, habitual and slow
rates.
Duration of the slow chewing cycles was approximately
25% longer than the habitual cycles (1149 21* versus
917 + 186* ms, respectively). The small standard errors when chewing at the habitual rate (Table 2). Although
associated with each of these mean rates are because of slightly larger, the lateral excursive range during slow
the selection of the 10 most representative cycles and chewing did not differ signicantly from lateral excur-
the subjects who were chewing being controlled with a sions during habitual chewing.
metronome.

Cycle shape and the 3-D pathway

Cycle excursive range
The range of inferior chin movement was signicantly
less when chewing at the fast rate than when chewing
at the habitual and slow rates (Table 2). Maximum
inferior movement did not differ signicantly between
the slow chewing rate and the habitual rate. The range
of posterior movement of the chin showed a similar
pattern, but the differences were not statistically signi-
cant. Inferior and posterior excursive ranges are two
components of mouth opening. These results indicate
that subjects open less widely when chewing faster, but
open to the same size when chewing slowly or at their
habitual rate.
The lateral excursive range of the chin marker was
signicantly less when chewing at the fast rate than
* Standard error.
A comparison of jaw movements in the frontal plane
(Fig. 2) shows that cycles, when chewing gum, begin
with excursion towards the balancing side during
opening. At about 1 mm of opening, the chin begins
to move gradually towards the working side as opening
continues. At maximum opening the chin is located a
little less than 2 mm lateral to the midline in the fast
cycles and a little more than 2 mm lateral to the
midline during habitual and slow cycles. During clo-
sing, the chin continues its excursion towards the
working side until about 3 mm below its starting
position. During the remainder of the closing motion
the chin moves back towards the midline.
The pathway traced by the chin in the frontal plane is
essentially identical for slow and habitual chewing
cycles. During the rst 6 mm of opening, the pathways
are also essentially identical among all three chewing

2001 Blackwell Science Ltd, Journal of Oral Rehabilitation 28; 328334

1 EFFECTS OF CHEWING RATES ON MANDIBULAR KINEMATICS 331

Table 2. Inferior, posterior and lateral

Inferior Posterior Lateral Total 3-D
ranges of jaw excursion (mm) and total Chewing
3-D distances for slow, habitual and fast rate Mean s.e. Mean s.e. Mean s.e. Mean s.e.
chewing cycles
Slow 861 047 609 052 665 038 284 13
Habitual 856 032 609 035 648 024 276 09
Fast 752 043 547 047 552 032 239 12

Fig. 2. Frontal view of the jaw movements for habitual, slow and Fig. 3. Superior view of jaw movements for habitual, slow and
fast chewing rates. The y-axis represents inferior excursion of the fast chewing rates. The y-axis represents posterior excursion of the
chin in mm, the x-axis represents mediolateral excursions of the chin in mm, the x-axis represents mediolateral excursions of the
8 chin in mm. 9 chin in mm.

rates. However, at maximum opening, the chin is less
laterally and less inferiorly located during fast cycles
than it is during either slow or habitual cycles. During
the last 2 mm of closing the three chewing rates again
trace almost identical paths. However, during most of
the closing phase the chin is located closer to the
midline in the fast cycles than it is during slow or
habitual cycles. There is clearly less mouth opening and
less lateral excursion during closing in the fast cycles
compared with the other two rates.
A comparison of jaw movements from a superior
perspective (Fig. 3) shows a similar pattern. When
chewing gum the chin moves initially towards the
balancing side and posteriorly during opening, then the
chin moves towards the working side as posterior
excursion continues. At maximum opening, the chin is
about 1 mm further posterior during slow and habitual
cycles than it is during fast cycles. As the chin moves
back anteriorly during closing, it is less lateral in fast
cycles than in slow and habitual cycles.
These differences in pathways can be summarized
by comparing their total 3-D lengths (Table 2). The
total pathway is about 4 mm shorter when subjects
chewed fast than when they chewed habitually or
slowly. There was no signicant difference in the total
3-D pathway between chewing at a slow rate or the
habitual rate.
Mandibular velocities
Total 3-D velocity is the sum of the velocities along the
vertical, AP and mediolateral axes. When 3-D velocity

2001 Blackwell Science Ltd, Journal of Oral Rehabilitation 28; 328334

332 G . S . T H R O C K M O R T O N et al.

Table 4. Between cycle variation in total cycle duration (ms),

total distance (mm) and maximum 3-D velocity (mm s1) for slow,
habitual and fast chewing cycles

Duration 3-D distance 3-D velocity

Chewing
rate Mean s.e. Mean s.e. Mean s.e.

Slow 501 33 191 13 1084 72

Habitual 132 09 100 07 1129 74
Fast 23 01 40 03 1161 76

Chewing cycle variability

Fig. 4. 3-D velocities of jaw movements for habitual, slow and
fast chewing rates plotted as percentage of total cycle duration As shown in Table 4, subjects had the most difculty in
()100 represents the beginning of the opening phase; 0 represents matching cycle duration with the metronome when
maximum opening, the end of the opening phase and the
chewing slowly. Even when chewing at their habitual
beginning of the closing phase; and 100 represents the end of
the closing phase). rate, subjects showed considerable variance in the
duration of their chewing cycles. Variance in cycle
duration was least during fast chewing. A similar
is plotted against standardized cycle duration it is clear
pattern is seen in variation in length of the 3-D
that velocity increases rapidly at the beginning of the
pathway. As chewing speed increased variation in path
opening phase, gradually increases to a peak at about
length decreased signicantly. Interestingly, variation
75% of the opening phase and then decreases as
in maximum 3-D velocities was essentially identical
maximum opening is achieved (Fig. 4). The 3-D velo-
among the three chewing rates. Although the coef-
cities increase again to a peak at 2530% of the closing
cients of variation show that variability in maximum
phase and then gradually decrease until about 95% of
velocity decreases with increasing speed, these differ-
the closing phase, when there is a sharp decrease in
ences are not statistically signicant.
velocity until closing is complete.
Except for the rst 5% and last 5% of the cycle
duration, mandibular velocity is always greater during Discussion
the fast chewing cycles than during habitual cycles and
When subjects were asked to chew faster than their
mandibular velocity is always slower during the slow
normal rate they used a combination of a smaller 3-D
chewing cycles than in the habitual cycles. Differences
pathway and a higher mandibular velocity to achieve
are the greatest at the peak velocities of opening and
the shorter duration. In our study, the duration of the
closing. Even at maximum opening, where the vertical
fast cycles was about 30% less than the subject's
and AP velocities are zero, the lateral velocities (the
habitual rate. The 3-D pathway during fast cycles was
only velocity contributing to the 3-D velocity) are
about 10% shorter than the pathway of habitual cycles.
signicantly different among the three chewing speeds.
If these two chewing rates had used the same mandib-
Maximum vertical, AP, lateral and total 3-D velocities
ular velocity, the fast cycle duration would be expected
were all signicantly different among the three rates
to be only about 10% shorter. However, peak mandib-
(Table 3).
ular velocities during the fast cycles were also about
25% higher than those of the habitual cycles. The
Table 3. Maximum vertical, AP, lateral and total 3-D jaw
velocities (mm s)1) for slow, habitual and fast chewing cycles combination of a faster velocity and shorter 3-D
pathway are sufcient to explain the 30% shorter
Vertical AP Lateral 3-D duration of the fast cycles.
Chewing When subjects were asked to chew slower than
rate Mean s.e. Mean s.e. Mean s.e. Mean s.e.
normal their 3-D pathway was the same as the pathway
Slow 441 26 350 26 347 17 614 29 used at habitual chewing rates although the duration of
Habitual 511 23 406 25 401 17 710 27
the cycles was 25% longer. Their longer duration must
Fast 632 40 469 40 452 23 839 43
therefore be entirely because of a slower velocity during

2001 Blackwell Science Ltd, Journal of Oral Rehabilitation 28; 328334

1 EFFECTS OF CHEWING RATES ON MANDIBULAR KINEMATICS
the slow cycles. The peak velocities during slow cycles
were indeed 25% lower than the peak velocities during
habitual cycles. The velocity decrease is sufcient to
explain the 25% longer duration of the slow cycles.
Our results for faster than normal cycles are in close
agreement with those of Plesh et al. (1987, 1993), who
reported that as chewing speed increased, vertical gape,
the 3-D gape and the total 3-D pathway all decreased.
In addition, the amount of lateral excursion also
decreased as chewing speed increased. However, Plesh
et al. (1993) reported that the decrease in lateral
excursion was most pronounced during the opening
phase and that the entire cycle shifted towards the
working side. Our results indicate that the reduction in
lateral excursion occurs only late in the opening phase
and through much of the closing phase.
Plesh et al. (1987, 1993) based their pattern on all
cycles in the chewing sequences. Their cycles showed a
high degree of variability both within and between
subjects. They were unable to quantify the changes in
lateral excursions for statistical analysis. In this study,
selection of the 10 most representative cycles for each
chewing sequence greatly reduced within subject vari-
ability and the multilevel modelling allowed a more
precise comparison of cycle patterns both within and
between subjects. These techniques make it possible to
better quantify differences among the three chewing
rates and to locate where these differences occur in the
cycle.
The excursive ranges we reported for habitual chews
differ from jaw movements previously reported. Our
vertical range was 37 mm less, our posterior range was
1520 mm more, and our lateral range was 05
35 mm more than the maximum movements previ-
4 ously reported for gum chews (Neill & Howell, 1988;
Karlsson & Carlsson, 1989; Youssef et al., 1997; Snipes
5 et al., 1998). The differences are most likely methodo-
logical because our 3-D cycle distances compare well
with those reported by Karlsson and Carlsson (1989) for
gum and Kiliaridis et al. (1991) for peanuts. The vertical,
inferior and lateral components probably differed
because (i) we evaluated ranges of movements rather
than maximum movements, (ii) different starting points
for measuring excursions were used, (iii) we used the 10
most representative cycles, effectively eliminating
extreme cycles and (iv) differences in bolus size might
be expected to affect inferior and posterior movements.
No previous studies have examined cycle shape and
dynamic velocities when subjects were asked to chew
2001 Blackwell Science Ltd, Journal of Oral Rehabilitation 28; 328334
333
more slowly than normal. Normal adult chewing rates
generally centre around 1 cycle s)1 (1 Hz) (Shepherd,
1960; Beyron, 1964; Ahlgren, 1966; Atkinson &
Shepherd, 1967; Wictorin et al., 1968; Hedegard et al.,
6 1970; Gillings et al., 1973; Morimoto et al., 1984; Plesh
et al., 1988) although some faster rates have been
reported (Anderson, 1954; Mller, 1966). Plesh et al.
(1993) examined chewing rates of 100 and 160
cycles min)1, both faster than habitual chewing rates.
Morimoto et al. (1984) used chewing rates ranging from
2 to 7 Hz, all considerably faster than normal. Plesh et al.
(1987) examined chewing rates of 46 cycles min)1 but
did not include habitual-rate chewing cycles for com-
parison. In addition, their method of analysis did not
allow quantitative comparison of the slow cycles with
fast cycles (100 and 160 cycles min)1). Other than the
longer duration of slower cycles, they did not describe
characteristics of slower than normal cycles and did not
compare them with cycles at the subjects habitual rate.
Our results indicate that when subjects chew more
slowly than normal they follow their habitual 3-D
pathway of mandibular movement. Throughout this
pathway they use a lower mandibular velocity in order to
meet the cycle duration requirements set by the metro-
nome. However, controlling this slower chewing cycle
appears to be more difcult, as demonstrated by higher
levels of between-cycle variability. These results suggest
that slower chewing is controlled by voluntary slowing
of the central pattern generator. There appear to be no
physiological or neurological limits to slow chewing.
In contrast, when subjects chew faster than normal
they not only increased their mandibular velocity but
also decrease the distance the mandible travels. The
combination of a shorter pathway and higher velocity
results in the observed shorter cycle duration. The
reason why a shorter pathway is needed for faster
cycles is not known. Subjects have been able to chew
up to 7 cycles s)1 (Morimoto et al., 1984), more than
four times faster than the fast cycles in this study. If the
jaw muscles can be activated and contracted at these
high frequencies, then they should be able to contract
fast enough to generate chewing cycles at 100 Hz with
a more normal pathway.
One possible explanation might be that the inertia of
the mandible, tongue and hyoid apparatus have a
resonant frequency that tends to establish a subject's
habitual chewing frequency. Elasticity of perioral soft
tissues also contribute to this resonant frequency.
Different inertial masses and quality of perioral soft
334 G . S . T H R O C K M O R T O N et al.
tissues may help establish different resonant frequencies
for each individual. Inertia and elasticity will not limit
slow movements of the mandible and subjects would be
expected to maintain a normal cycle shape lower than
normal speeds. But inertia and elasticity may limit fast
movements of the mandible. As mandibular velocity is
increased the force needed to accelerate the mandible
also increases. In addition, elastic bres are often more
resistant to fast changes. If this theory is correct, then as
chewing speed is increased, muscular effort is not able to
accelerate the mandible to the same extent as at the
resonant frequency and the pathway taken by the
mandible is shortened. Further research is needed to
determine the validity of this hypothesis.
Reduced excursions seen in faster than normal cycles
might also result from interference in the pattern of
muscle contractions. The shape of the chewing cycle is
probably determined by the relative timing of muscle
contractions as well as the magnitude of those contrac-
tions. For example, there might be more overlap in the
activities of opening and closing muscles or overlap
between muscles producing balancing side excursion
and those producing working side excursion. None of
our subjects had any training in chewing at faster rates.
It is possible that subjects might learn to maintain their
habitual 3-D pathway while increasing their chewing
frequency. Further research is needed to test these
hypotheses.
Acknowledgements
This research was supported by NIDR trainee grant
DE07188 and the Center of Craniofacial Research and
Diagnosis, Baylor College of Dentistry.
References
AHLGREN , J. (1966) Mechanism of mastication: a quantitative
cinematographic and electromyographic study of masticatory
movements in children, with special reference to occlusion of
the teeth. Acta Odontologica Scandinavica, 24, 1.
ANDERSON , D.J. (1954) Studies in the dynamics of mastication.
Dissertation, University of London, London.
ATKINSON , H.F. & SHEPHERD , R.W. (1967) Masticatory movements
and the resulting force. Archives of Oral Biology, 2, 195.
BEYRON , H. (1964) Occlusal relations and mastication in Austra-
lian aborigines. Acta Odontologica Scandinavica, 22, 597.
GILLINGS , B.R.D., GRAHAM , C.H. & DUCKMANTON , N.A. (1973) Jaw
movements in young adult men during chewing. Journal of
Prosthetic Dentistry, 29, 616.
GOLDSTEIN , H. (1986) Multilevel mixed linear model analysis
using iterative generalized least squares. Biometrika, 73, 43.
GOLDSTEIN , H. (1987) Multilevel Model in Educational and Social
Research. Grifns, London.
HEDEGA
EDEGARD RD , B., LUNDBERG , M. & WICTORIN , L. (1970) Masticatory
function a cineradiographic study IV. Duration of the
masticatory cycle. Acta Odontologica Scandinavica, 28, 859.
JEMT , T., KARLSSON , S. & HEDEGA RD , B. (1979) Mandibular
EDEGARD
movements of young adults recorded by intraorally placed light-
emitting diodes. Journal of Prosthetic Dentistry, 42, 669.
KARLSSON , S. & CARLSSON , G.E. (1989) Recording of masticatory
mandibular movements and velocity by an optoelectric method.
International Journal of Prosthodontics, 2, 490.
KILIARIDIS , S., KARLSSON , S. & KJELLBERG , H. (1991) Character-
istics of masticatory mandibular movements and velocity in
growing individuals and young adults. Journal of Dental Research,
70, 1367.
MLLER , E. (1966) The chewing apparatus. Acta Physiologica
Scandinavica, 69 (Suppl. 280).
MORIMOTO , T., INOUE , T., NAKAMURA , T. & KAWAMURA , Y. (1984)
Frequency-dependent modulation of rhythmic human jaw
movements. Journal of Dental Research, 63, 1310.
NEILL , D.J. & HOWELL , P.G.T. (1988) A study of mastication
in dentate individuals. International Journal of Prosthodontics,
1, 93.
PLESH , O., BISHOP , B. & MC CALL , W.D. Jr. (1987) Mandibular
movements and jaw muscles' activity while voluntarily chewing
at different rates. Experimental Neurology, 98, 285.
PLESH , O., BISHOP , B. & MC CALL , W.D. Jr. (1988) Comparison of
automatic and voluntary chewing patterns and performance.
7 Experimental Neurology, 99, 326.
PLESH , O., BISHOP , B. & MC CALL , W.D. Jr. (1993) Kinematics of
jaw movements during chewing at different frequencies. Journal
of Biomechanics, 26, 243.
SHEPHERD , R.W. (1960) A further report on mandibular move-
ment. Australian Dental Journal, 5, 337.
SNIPES , W.B., THROCKMORTON , G.S. & BUSCHANG , P.H. (1998)
Normal masticatory function of girls and young women:
mandibular masticatory movements. American Journal of Human
Biology, 10, 53.
STRENIO , J., WEISBERG , H.I. & BRYK , A.S. (1983) Empirical Bayes
estimation of individual growth curve parameters and their
relationship to covariates. Biometrics, 39, 71.
WICTORIN , L., HEDEGa RD , B. & LUNDGREN , M. (1968) Masticatory
EDEGaRD
function a cineradiographic study III. Position of the bolus in
individuals with full complement of natural teeth. Acta Odon-
tologica Scandinavica, 26, 213.
YOUSSEF , R., THROCKMORTON , G.S., ELLIS , E. III & SINN , D.P.
(1997) Comparison of habitual masticatory patterns in men and
women using a custom computer program. Journal of Prosthetic
Dentistry, 78, 179.
Correspondence: Dr Peter H. Buschang, Department of Orthodontics,
Baylor College of Dentistry, The Texas A&M University System Health
Science Center, 3302 Gaston Avenue, Dallas, TX 75246, U.S.A.
E-mail: phbuschang@tambcd.edu
2001 Blackwell Science Ltd, Journal of Oral Rehabilitation 28; 328334