ORIGINAL ARTICLE

Numerical simulation of canine retraction by

sliding mechanics
Yukio Kojimaa and Hisao Fukuib
Nagoya, Japan

Background: Bone remodeling laws have been used to simulate the movement of a single tooth, but the
calculations for simulating the movement of several teeth simultaneously are time-consuming. The purpose
of this article is to discuss a method that allows the simulation of more complex tooth movements. Methods:
A 3-dimensional finite element method was used to simulate the orthodontic tooth movement (retraction) of
a maxillary canine by sliding mechanics and any associated movement of the anchor teeth. Absorption and
apposition of the alveolar bone were produced in proportion to the stress of the periodontal ligament.
Results: In a reference case, the canine was retracted by a 2N force with 0.016-in square wire. The frictional
coefficient between wire and bracket was 0.2. The movement of both the canine and the anchor teeth could
be calculated with the elastic deformation of wire. The canine tipped during the initial unsteady state and then
moved bodily during the steady state. It became upright when the orthodontic force was removed. The
anchor teeth moved in the steady state and tipped in the mesial direction. The decrease in applied force by
friction was about 70%. The tipping of the canine decreased when the wire size was increased or when the
applied force was decreased. Conclusions: Simple assumptions were used in this calculation to simulate
orthodontic tooth movements. The calculated results were reasonable in mechanical considerations. This
method might enable one to estimate various tooth movements clinically. However, precise comparisons
between calculated and clinical results, and the improvement of the calculation model, are left for a future
study. (Am J Orthod Dentofacial Orthop 2005;127:542-51)

bone.5-7 In these methods, applying bone remodeling

W
hen an orthodontic force is applied to a
tooth, stresses and strains are produced in
the periodontal ligament (PDL) and the
surrounding bone. Absorption of alveolar bone occurs
in a region of compressive stress, whereas apposition of
the alveolar bone occurs in a region of tensile stress. By
this bone remodeling action, the teeth move. To plan
tooth movement, a clinician must understand the force
to be applied to the tooth and the stress distribution in
the PDL.
Many calculations have been carried out by using
the finite element method on the stress distribution in
the PDL. These results show that the moment to force
(M/F) ratio controls the pattern of tooth movement,1,2
ie, tipping or bodily movement. These results have been
used for appliance design.3,4 Some trials have been
done to estimate long-time orthodontic tooth movement
considering absorption and apposition of the alveolar
a
Associate professor, Department of Mechanical Engineering, Nagoya Institute
of Technology.
b
Professor, Department of Dental Materials Science, School of Dentistry,
Aichi-Gakuin University, Nagoya.
Reprint requests: Hisao Fukui, Aichi-Gakuin University, School of Dentistry,
Department of Dental Materials Science, 1-100 Kusumoto-cho, Chikusa-ku,
Nagoya, 464-8650, Japan; e-mail, fukui@dpc.aichi-gakuin.ac.jp.
Submitted, March 2003; revised and accepted, December 2004.
0889-5406/$30.00
Copyright © 2005 by the American Association of Orthodontists.
doi:10.1016/j.ajodo.2004.12.007
542
laws, the movements of a single tooth were simulated.
These methods necessitate the finite element subdivi-
sion in the tooth and the alveolar bone; thus, the
calculation when many teeth move simultaneously
becomes time-consuming. The general method to esti-
mate clinical tooth movements has not yet been estab-
lished.
We have proposed a 3-dimensional (3D) finite
element method for calculating orthodontic tooth
movement.8,9 By this method, the change in orthodon-
tic forces of the tooth movement can be calculated as a
coupled problem of teeth and appliance. In a previous
article,9 a simple uprighting of a mandibular second
molar was calculated, and the movements of the molar
and the anchorage teeth were simulated.
The main purpose of this article is to show that a
more complex case can be calculated by applying this
method. As a clinical orthodontic treatment, maxillary
canine retraction by sliding mechanics was selected.
Some mechanical problems remain to be clarified in
this treatment. For example, the frictional force be-
tween wire and bracket decreases the orthodontic force
and impedes tooth movement.10 Many experiments on
frictional force have been carried out,11-14 but the
magnitude of the decrease in orthodontic force was not
clarified. Also, other mechanical factors that affect the
American Journal of Orthodontics and Dentofacial Orthopedics
Volume 127, Number 5
Fig 1. Stress in PDL and spring element.
movement were not clarified. Another purpose of this
article is to calculate the movement of canine and
anchorage teeth as well as the frictional force and to
investigate the mechanism of sliding mechanics.
CALCULATION METHOD
Stress in the PDL
The calculation method has been explained in
previous articles.8,9 The important points of the method
and the assumptions used in the calculation were
shown. The tooth and the alveolar bone were assumed
to be rigid bodies. The validity of this assumption was
previously confirmed.8 The stress distribution in the
PDL is calculated by the 3D finite element method. A
remarkable point in this method is the use of a special
element that represents a tooth and the PDL.
According to an educational dental model, a model
of a tooth is made as shown in Figure 1. The surface of
the model is divided into small triangular elements. The
PDL, which is assumed to be a linear elastic film,
covers the elements. The nonlinear property and the
anisotropy of the PDL were not assumed to simplify the
calculation model. The thickness of the PDL was t ⫽
0.2 mm.15 The stress state in the PDL was assumed to
be a plane strain condition. With these assumptions, the
force and moment transmitted from the PDL could be
calculated when a translation or rotation was given at
Kojima and Fukui 543
Fig 2. Canine retraction by sliding mechanics.
the bracket position of the tooth. This relationship
between forces and movements is considered as a
stiffness equation of the finite element method. The
tooth can be replaced by a single element, called the
tooth element. This element has 6 degrees of freedom at
the node i. This degree is the same as those of a 3D
beam element. In the calculation of the tooth element,
only the element subdivision of the root surface is
necessary.
Young’s modulus and Poisson’s ratio of the PDL
were assumed to be E ⫽ 0.2 MPa and ␯ ⫽ 0.47,
respectively. With these values, the labiolingual mobil-
ity and the axial mobility of the maxillary first premolar
became about 20 and 10 ␮m/N respectively. These
values of mobility are roughly the same as the mea-
sured values.16 Also, the PDL was considered to be
nearly incompressible material, and Poisson’s ratio was
used: as ␯ ⫽ 0.45 and ␯ ⫽ 0.49.2,7
A simple spring element as shown in Figure 1 was
used to connect the bracket and the wire. Special
contact elements were not used. If the spring constant is
a very large value in some direction, the relative
displacement between wire and bracket is completely
restricted in its direction. If the spring constant equals
0, then the wire can move freely in its direction. This
corresponds to the bracket at the canine. The wire is
assumed to be an elastic body, and it is divided into the
3D beam elements. After the tooth elements, the spring
elements (brackets) and the beam elements (wire) are
connected, the unknown variables are only the nodal
displacements of the wire and bracket, and so the
calculation time becomes very short. When many teeth
are connected with a wire and move simultaneously,
this can be calculated simply.
Absorption and apposition of the alveolar bone
Bone remodeling is a well-known phenomenon in
orthopedics.17 However, the biological mechanism of
544 Kojima and Fukui
Fig 3. Frictional forces produced by force transmitted from bracket slot to wire.
bone remodeling has not been clarified. We considered
only orthodontic tooth movement.
The reassessment18 of many experiments shows
that, with orthodontic force up to about 2 N, the rate of
tooth movement is increased by an increase in the
force. The orthodontic force corresponds to the stress in
the PDL, and tooth movement corresponds to the
absorption of alveolar bone. Then, the bone absorption
depends on the stress. By using the simplest relation-
ship, the rate of bone absorption was assumed to be
proportional to the mean stress in the PDL. The mean
stress ␴mi is defined as an average of 3 normal stresses,
␴1i, ␴2i, and ␴3i. The ␴mi becomes a negative value for
compressive stress. Because the stress-strain relation-
ship was assumed to be linear, the stress is equivalent to
the strain as the bone remodeling stimulus.
The absorption of the alveolar bone for a time
increment ⌬T is assumed to be -C␴mi ⌬T. C is the
amount of absorption (␮m) per unit of time (day) and
unit of stress (kPa). The speed of tooth movement is
proportional to the coefficient C, because the relation-
ship between bone absorption and stress is assumed to
be linear. The movement is controlled by the parameter
CT. An increase in absorption rate C is equal to the
decrease in time T. Because the value of C is unknown
at the present time, the progress of tooth movement is
indicated by the parameter CT.
Incremental procedure for tooth movement
A mathematical method is proposed to move the
tooth corresponding to the bone remodeling.9 In the
bone absorption site, the surface of the alveolar bone
shifts to the outside from that of the tooth root. This
American Journal of Orthodontics and Dentofacial Orthopedics
May 2005
action is seen as the PDL is elongated and pulls the
tooth to the moving direction. On the other hand, in the
bone apposition site, the PDL is compressed and
presses the tooth. The strain produced by the elongation
or the compression in the PDL can be written as
⌬␧1i ⫽ –C␴mi⌬ T ⁄ t
and the corresponding normal elastic stress
E(1 ⫺ v)
⌬␴1i ⫽ ⌬␧1i
(1 ⫹ v)(1 ⫺ 2v)
is generated. It is assumed that the moving forces
produced by this stress made the tooth move. The
moving forces are calculated by integrating the stress
⌬␴1i over the tooth root.
By using the moving forces, the long-time tooth
movement is calculated as following incremental pro-
cedure. (1) At time T, the stresses in the PDL are
calculated, and the moving forces for the small-time
increment ⌬T are calculated. (2) Applying the moving
forces to the teeth, the tooth positions at time T ⫹ ⌬T
are calculated. (3) The change in the force transmitted
to the wire is calculated, and the stresses in the PDL are
modified. These calculations from (1) to (3) are re-
peated.
Analysis model
Figure 2 shows the calculation model when a
maxillary canine is retracted by sliding mechanics. As
anchorage, incisors, a premolar, and molars were con-
nected with a wire. Wire bending such as a gable bend
was not used. The wire and the bracket of these teeth
American Journal of Orthodontics and Dentofacial Orthopedics
Volume 127, Number 5
Fig 4. Translation, tipping angle, and rotational angle
with time CT.
were tightened so that they would not slide into each
other. The width of the bracket was Wb ⫽ 3 mm. It was
assumed that the wire was not deformed in the bracket
slot. A 0.016-in square wire was selected as a reference.
However, this wire might not be adequate in clinical
situations. The wire was made of stainless steel
(Young’s modulus, 200 GPa; Poisson’s ratio, 0.3).
In an actual patient, there would be some clearance
between the wire and the bracket slot. This clearance
was not considered in the present calculation, and the
canine bracket moved closely along the wire. Initial
friction produced by the ligation was assumed not to
exist.
The frictional force Fr was produced by the forces
applied to the bracket and divided into 5 forces—Fr1,
Fr2, Fr3, Fr4, Fr5—as shown in Figure 3. Fr3 and Fr4
decreased as the bracket width Wb increased. Fr5
became smaller as the wire size B increased. The
frictional coefficient ␮ between the bracket and the
wire was assumed to be 0.2 by referring to measured
values.11-14
Distal and mesial forces of 2N were applied to the
brackets of the canine and the first molar. The magni-
tude of the force was maintained constantly for the
tooth movement.
RESULTS AND DISCUSSION
Movement process in sliding mechanics
Figure 4 shows the change of translation u, tipping
angle ␣, and rotational angle ␥ of the canine with time
CT. The u was defined as the movement at the bracket,
the ␣ as the inclination angle of a vertical axis fixed in
the bracket, and the ␥ as the rotational angle about the
vertical axis. The abrupt change of ␣,␥ in Figure 4 was
caused when the bracket node of the canine passed a
wire node. This change was a calculation error. The
Kojima and Fukui 545
Fig 5. Variation of frictional force with time CT.
change in translation u and rotational angle ␥ had
almost the same tendency. The u and ␥ rapidly in-
creased immediately after the orthodontic force was
applied, and the increase rate approached a constant
value as time elapsed. The tipping angle ␣ rapidly
increased and approached a constant value. Most of the
rapid change in ␥ was caused by the change in
curvature of the archwire. The change in the tangential
direction of the wire from the canine to the first
premolar was 10°, so the canine was rotated along the
wire.
Figure 5 shows the change in the frictional forces
Fr1 to Fr5 and the total force Fr with time CT. The net
force P transmitted to the canine was obtained by
subtracting Fr from P0. The spike-like change of the
frictional forces in Figure 5 was a calculation error
caused for the same reason as in Figure 4. After the
orthodontic force was applied, the frictional forces
fluctuated and approached constant values at CT ⬎ 200.
Most of the total frictional force Fr was Fr4 and Fr3.
These forces were produced by the uprighting moment
Mn and the antirotating moment Mz. The Fr5 was about
17% of the total frictional force Fr. The Fr5 was
produced by the torque Mt that prevented the axial
rotation of the wire. The frictional forces Fr1 and Fr2
were not plotted in Figure 5 because they were very
small.
Figure 6, A, shows the teeth at CT ⫽ 1, immediately
after the orthodontic force was applied. The movement
of the teeth was very small, so it was magnified 20
times. The stress distribution in the PDL is indicated by
color contours. The teeth drawn with blue lines show
the initial positions. The movement pattern of the
canine was mainly tipping and rotation. On the distal
surface of the canine, compressive stress was produced
546 Kojima and Fukui American Journal of Orthodontics and Dentofacial Orthopedics
May 2005

Fig 6. Tooth movement at 3 different states. A, Initial orthodontic movement (CT ⫽ 1 ␮m/kPa); B,
steady state orthodontic movement (CT ⫽ 1000 ␮m/kPa); C, recovery movement after unloading
(CT ⫽ 2000 ␮m/kPa).
American Journal of Orthodontics and Dentofacial Orthopedics
Volume 127, Number 5
Fig 7. Forces transmitted to canine and anchor teeth.
A, Initial orthodontic movement (CT ⫽ 1 ␮m/kPa); B,
steady state orthodontic movement (CT ⫽ 1000 ␮m/
kPa)
near the neck, and tensile stress was produced near the
apex. The rotational center, at which the stress becomes
0, was about 7 mm from the apex. The maximum stress
in the PDL of the canine was 20 to 30 kPa. The
anchorage teeth—the incisors, the second premolar,
and the molars—also tipped. On the mesial or labial
surface of these teeth, compressive and tensile stresses
were produced.
Figure 6, B, shows the teeth at CT ⫽ 1000, when
the canine moved 4.5 mm. The tipping angle and the
rotational angle of the canine were ␣ ⫽ 3.4° and ␥ ⫽
8.7°. Most of the tipping and rotation were produced in
a short time (CT ⬍ 200). At CT ⫽ 1000, a compressive
stress was produced over the distal surface of the
Kojima and Fukui 547
Fig 8. Change in net force transmitted to canine with
frictional coefficient ␮.
canine, and a tensile stress was produced over the
mesial one. This stress distribution indicates that the
movement pattern of the canine was almost all bodily
movement. The maximum stress was about 3 kPa. This
value was smaller than that for the tipping movement.
At CT ⫽ 1000, the frictional force Fr was 1.47N,
and the net force transmitted to the canine was 0.53N.
This was 27% of the force applied to the bracket (2N).
The same frictional force 1.47N was applied to the wire
in the distal direction, so that the total force transmitted
to the wire in the mesial direction became 0.53N. This
magnitude is the same as the force on the canine. By
this force, all anchor teeth moved about 1 mm in the
mesial direction. The incisors, the premolar, and the
first molar showed bodily movement; the second molar
showed tipping movement. The incisors extruded about
1 mm. This extrusion was caused by the moment Mn of
the canine transmitting to the wire, and an extrusion
force was transmitted to the incisors as the reaction
force. The premolar was intruded by the opposite force.
Figure 6, C, shows the teeth at CT ⫽ 2000, when
sufficient time had elapsed after the orthodontic force
was removed at CT ⫽ 1000. In this state, the canine
became upright, and the tipping angle ␣ decreased to
1.6°. The rotational angle ␥ also decreased to 7.8°. The
anchor teeth moved slightly in the distal direction.
These phenomena were produced by the elastic recov-
ery of the wire. At this time, the frictional force
impeded the slide of the bracket on the canine along the
wire, so that the canine could hardly move at the
bracket. Without friction between wire and bracket, the
canine goes back in the mesial direction with upright-
ing. These movements occurred under a small stress
level.
The process of tooth movement is considered by
548 Kojima and Fukui American Journal of Orthodontics and Dentofacial Orthopedics
May 2005

Fig 9. Effect of magnitude of orthodontic force (P0 ⫽ 4N).

Fig 10. Effect of wire size (B ⫽ 0.02 in).

Fig 11. Effect of round wire.

American Journal of Orthodontics and Dentofacial Orthopedics
Volume 127, Number 5
using the results in Figures 4, 5, and 6. When orthodon-
tic force is applied, tooth movement consists of the
initial unsteady state and then the steady state. In the
unsteady state—Figure 6, A—the M/F ratio of the
canine was 1.4. This value is not sufficient to maintain
the bodily movement,2 and the canine indicates tipping
movement. However, this movement is prevented by
the moments Mn produced by the elastic deformation of
the wire. The tipping of the canine is gradually re-
stricted. In Figure 6, B, the M/F ratio increased to 9.8.
The movement of the canine is changed to a bodily
movement.2 The frictional force Fr4 increased with the
increase in the moment. The net force transmitted to the
canine decreased and approached a constant value. The
canine kept the bodily movement at constant speed.
This is the steady-state movement. All anchor teeth
move in the mesial direction, the incisors elongate, and
the premolar depresses by the moment transmitted to
the wire. The forces applied to the teeth are changed
with the movement. This change is the general phe-
nomenon that occurs when the teeth are connected with
a wire. For such a case, it is difficult to estimate
long-time tooth movement by using the initial move-
ment.
The tooth movement pattern in the steady state is
reasonable compared with clinical orthodontic tooth
movements. Also, the tooth movement process shown
by the calculated results is reasonable in mechanical
considerations. However, at the present time, it is
difficult to obtain definite experimental evidence for the
unsteady-state tooth movement. For example, in the
measurements by Huffman and Way,19 who measured
tooth movements along archwires of 2 difference sizes,
the rapid increase of tooth movement in the unsteady
state was not observed. The incubation time for tooth
movement might be included in these measurements.
This incubation time, which is not considered in our
calculation, misreads the unsteady-state tooth move-
ment. In the 14 measurements of Andreasen and Zwan-
ziger,20 the increase patterns of the tooth movement
were different with each subject. In several cases, a
rapid increase of tooth movement was observed. On the
other hand, in frictionless mechanics, the initial in-
crease of tooth movement was observed.21 There are no
studies on the uprighting of canines after orthodontic
forces are removed. If this recovery movement is
realized in clinical situations, the tipped canine in the
sliding mechanics will be upright by removing the
force. The precise comparison between these uncertain
phenomena and the clinical results is left for a future
study.
The magnitude of tooth movement is proportional
to the coefficient of bone absorption rate C. The value
Kojima and Fukui 549
of C is estimated by referring to an experimental result.
Ziegler and Ingervall22 reported that the canine moved
an average of 4.67 mm for 100 days when a 2N force
was applied. This time is assumed to correspond to the
case in Figure 6, B, substituting T ⫽ 100 days into CT
⫽ 1000; then the coefficient C is estimated to be
10␮m/kPa·day. This value is roughly the same as the
value estimated from another experimental result.23
Force transmitted to anchor teeth
Figure 7 shows the movement of the anchor teeth
and the forces transmitted to them. In the case of CT ⫽
1, tooth movement is magnified 20 times. In the initial
unsteady state (CT ⫽ 1), the moment Mz in the xy plane
was applied to the wire by the rotation of the canine.
The reaction forces were transmitted to the incisors in
the labial direction and to the premolar in the lingual
direction. In the steady-state movement (CT ⫽ 1000),
the direction of all forces transmitted to the anchor teeth
was approximately opposite to that of the canine. In
both cases of CT ⫽ 1 and 1000, the largest force was
transmitted to the first molar; it was about twice that
transmitted to the second molar.
Effect of frictional coefficient
Figure 8 shows the net force transmitted to the
canine in the steady-state movement, when the fric-
tional coefficient ␮ changed. The net force P (P0-Fr)
decreased with an increase in ␮. At the same value of
␮, the ratio of net to transmitted force, P/P0, was about
equal, regardless of applied force P0 and wire size B.
The results show that the frictional coefficient ␮ de-
pends on the material of the wire and the bracket. In the
stainless steel wire and bracket, it is between ␮ ⫽ 0.1
and 0.3.11,13 By using the calculated result in Figure 8,
the net force in a clinical situation can be estimated to
be between 20% and 40% of the applied force. In other
words, the decrease in orthodontic force by friction is
between 60% and 80%.
Effect of magnitude of orthodontic force
Figure 9 shows the applied force P0 increased from
2N to 4N. The net force transmitted to the canine was
1.15N. This value was about twice that for P0 ⫽ 2N.
The stress in the PDL increased, and the bone absorp-
tion per unit of time increased. Therefore, the time
required to move the canine 4.5 mm was decreased to
CT ⫽ 368. This is a natural result derived from the
linear relationship between stress and bone absorption.
In the same range of applied force, the results measured
by Hixon et al24 showed that the movement rate of the
canine increased with an increase in applied force.
However, a more precise comparison on the linear
550 Kojima and Fukui
relationship between stress and bone absorption was
impossible to examine.
The tipping angle of the canine increased to ␣ ⫽
5.5°, and it was about 1.7 times greater than that in
Figure 6, B (P0 ⫽ 2N, ␣ ⫽ 3.4°). This is because the
deformation of the wire was increased in proportion to
the force transmitted to the wire. The movement of the
anchor teeth increased slightly.
Effect of wire
Figure 10 shows the calculated result when the wire
size is increased from 0.016 to 0.020 in. The net force
transmitted to the canine was hardly changed. The
tipping angle of the canine decreased to ␣ ⫽ 2.0°
compared with the 0.016-in wire. Also, the rotational
angle ␥ decreased slightly. This effect is because the
flexural rigidity of the wire became almost double, and
the deformation of the wire decreased. Oppositely,
when the wire size was decreased to 0.012 in, the
tipping angle of the canine increased to ␣ ⫽ 7.4°.
In the experiment by Huffman and Way19 with
0.016-in wire, the canine moved 3.37 ⫾ 1.07 mm and
tipped 5.30° ⫾ 2.37° over 10 weeks, and, with 0.020-in
wire, the canine moved 2.90 ⫾ 1.04 mm and tipped
1.70° ⫾ 1.35°. Assuming that the moving rate is fixed,
these values are converted when the canine moved 4.5
mm, and then the tipping angles become 7.08° ⫾ 3.16°
and 2.56° ⫾ 2.03°, respectively. This measured ten-
dency, when the tipping angle was decreased with an
increase in wire size, was the same as the calculated
one.
Figure 11 shows the square wire changed to a
0.016-in round wire. This wire can be rotated freely in
the bracket slot; ie, the torque Mt in Figure 3 is 0. The
frictional force produced by the Mt vanished, so that the
net force transmitted to the canine increased about 20%
compared with square wire. Therefore, the time re-
quired to move the canine 4.5 mm was decreased to CT
⫽ 829. The movement pattern of the canine was almost
equal to that of square wire. However, the movement of
the anchor teeth increased. The tipping of the incisors
increased, because the uprighting moment Mt was not
applied. Square wires are effective to prevent the
movement of the anchor teeth. This effect will be
consistent with the clinical experience that bodily
movement is effective to prevent movement of the
anchor teeth.10
CONCLUSIONS
Although the simplest assumptions were used in
this calculation, actual orthodontic tooth movement
such as sliding mechanics could be simulated. The
calculated results were reasonable in mechanical con-
American Journal of Orthodontics and Dentofacial Orthopedics
May 2005
siderations. This method may enable one to estimate
various tooth movements in a clinical environment.
However, the precise comparisons between the calcu-
lated and clinical results, and the improvement of the
calculation model, are left for a future study.
For sliding mechanics, this mechanical simulation
showed the following results.
1. The movement of the canine consisted of unsteady
state (tipping) and then steady state (bodily) move-
ment.
2. The decrease in orthodontic force by friction in
clinical situations was estimated to be between 60%
and 80%.
3. After the orthodontic force was removed, the ca-
nine did not return to the mesial direction and
became upright with time.
4. The tipping of the canine and the movement of the
anchor teeth (especially the elongation of the inci-
sors) decreased when the wire size increased.
5. The tipping of the canine increased when the
orthodontic force increased.
6. The square wire was more effective to decrease the
movement of the anchor teeth than the round wire.
We thank Professor Kuniaki Miyajima (Center for
Advanced Dental Education, Saint Louis University)
for his helpful advice.
REFERENCES
1. Tanne K, Sakuda M, Burstone CJ. Three– dimensional finite
element analysis for stress in the periodontal tissue by orthodon-
tic forces. Am J Orthod Dentofacial Orthop 1987;92:499-505.
2. Tanne K, Koenig HA, Burstone CJ. Moment to force ratios and
the center of rotation. Am J Orthod Dentofacial Orthop 1988;94:
426-31.
3. Bruce SH, William AS, Michael D. Auxiliary springs in contin-
uous arch treatment: Part 1. An analytical study employing the
finite-element method. Am J Orthod Dentofacial Orthop 1990;
98:387-97.
4. Raboud DW, Faulkner MF, Lipsett AW, Haberstock DL. Three-
dimensional effects in retraction appliance design. Am J Orthod
Dentofacial Orthop 1997;112:378-92.
5. Bourauel C, Freudenreich D, Kobe D, Jager A. Simulation of
orthodontic tooth movements. J Orfac Orthop 1999;60:136-51.
6. Bourauel C, Vollmer D, Jager A. Application of bone remodeling
theories in the simulation of orthodontic tooth movements. J
Orfac Orthop 2000;61:266-79.
7. Schneider J, Geiger M, Sander FG.. Numerical experiments on
long-time orthodontic tooth movement. Am J Orthod Dentofacial
Orthop 2002;121:257-65.
8. Kojima Y, Takano M, Fukui H, Mizutani M, Hasegawa J. A
simple method for calculating the initial tooth mobility and stress
distribution in the periodontal ligament. J Jpn Soc Dent Mat
Device 1999;18:210-6.
9. Kojima Y, Fukui H. A numerical analysis of tooth movement by
orthodontic forces. J Jpn Soc Dent Mat Device 2002;21:40-8.
American Journal of Orthodontics and Dentofacial Orthopedics Kojima and Fukui 551
Volume 127, Number 5

10. Proffit WR. Contemporary orthodontics. Saint Louis: Mosby;
1986.
11. Kusy RP, Whitley JQ. Effect of surface roughness on the
coefficients of friction in model orthodontic systems. J Biomech
1990;23:913-25.
12. Kusy RP, Whitley JQ, Prewitt MJ. Comparison of the frictional
coefficients for selected archwire-bracket slot combinations in
the dry and wet states. Angle Orthod 1991;61:293-302.
13. Noda T, Okamoto Y, Hamanaka H. Frictional property of
orthodontic wires: evaluation by static frictional coefficients. J
Jpn Orthod Soc 1993;52:154-60.
14. Kusy RP, Schafer DL. Effect of salivary viscosity on frictional
coefficients of orthodontic archwire/bracket couples. J Mat Soc
Mat Med 1995;6:390-5.
15. Coolidge ED. The thickness of the human periodontal mem-
brane. J Am Dent Assoc Dent Cosmos 1937;24:1260-70.
16. Goto T. An experimental study on the physiological mobility of
a tooth. Shikwa Gakuho 1971;71:1415-44.
17. Fung YC. Biomechanics, motion, flow, stress and growth. New
York: Springer; 1990.
18. Quinn RS, Yoshikawa DK. A reassessment of force magnitude in
orthodontics. Am J Orthod 1985;88:252-60.
19. Huffman DJ, Way DC. A clinical evaluation of tooth movement
along arch wires of two different sizes. Am J Orthod 1983;83:
453-9.
20. Andreasen GF, Zwanziger D. A clinical evaluation of the
differential force concept as applied to the edgewise bracket.
Am J Orthod 1980;78:25-40.
21. Boester CH, Johnston LE. A clinical investigation of the concept
of differential and optimal force in canine retraction. Angle
Orthod 1974;44:113-9.
22. Ziegler P, Ingervall B. A clinical study of maxillary canine
retraction with a retraction spring and with sliding mechanics.
Am J Orthod Dentofacial Orthop 1989;95:99-106.
23. Lee BW. Relationship between tooth-movement rate and esti-
mated pressure applied. J Dent Res 1965;44:1053.
24. Hixon EH, Aasen TO, Arango J, Clark RA, Klosterman R, Miller
SS, et al. On force and tooth movement. Am J Orthod 1970;57:
476-89.

Editors of the International Journal of Orthodontia (1915-1918),

International Journal of Orthodontia & Oral Surgery (1919-1921),
International Journal of Orthodontia, Oral Surgery and Radiography (1922-1932),
International Journal of Orthodontia and Dentistry of Children (1933-1935),
International Journal of Orthodontics and Oral Surgery (1936-1937), American
Journal of Orthodontics and Oral Surgery (1938-1947), American Journal of
Orthodontics (1948-1986), and American Journal of Orthodontics and Dentofa-
cial Orthopedics (1986-present)

1915 to 1931 Martin Dewey

1931 to 1968 H. C. Pollock
1968 to 1978 B. F. Dewel
1978 to 1985 Wayne G. Watson
1985 to 2000 Thomas M. Graber
2000 to present David L. Turpin