This study compared the widths of 20 commercially available preformed archwires to the widths of natural dental arches in 30 subjects with ideal occlusions. The study found that the preformed archwires were significantly narrower than the natural dental arches at both the canine and molar levels. Specifically, 14 archwires fell within 1 standard deviation of the mean canine width, but only 7 fell within 1 standard deviation of the mean molar width. The variations in current preformed archwire forms do not entirely correspond to the diversity of normal arch forms.

1. ORIGINAL ARTICLE
Commercially available archwire forms
compared with normal dental arch forms in
a Japanese population
Souichiro Oda,a
Kazuhito Arai,b
and Rizako Nakaharac
Tokyo, Japan
Introduction: The objective of this research was to evaluate commercially available preformed archwire
forms compared with normal dental arch forms. Methods: Thirty mandibular dental casts were made of
orthodontically untreated subjects with the most ideal occlusions, selected from among approximately
3500 students at Nippon Dental University. Canine and first-molar widths were measured on the dental casts
with a 3-dimensional laser scanning system. Twenty preformed mandibular archwires were scanned by using
a flatbed scanner. The archwire widths were measured at the level of the mean canine and first-molar depths,
and then compared with the natural dental arch widths, considering bracket thickness, measured with
a modified caliper. The results were statistically analyzed with the Mann-Whitney test. Results: The preformed
archwires that most closely matched the normal dental arch forms were the Orthos (Ormco, Glendora, Calif)
and Vari-Simplex large (Ormco) types. The preformed archwires were significantly narrower than the normal
dental arches at both the canine (P0.05) and molar (P0.01) levels. Conclusions: The variations in current
preformed archwires do not correspond entirely with various arch forms in a group with ideal occlusion. The
most commonly used archwires are similar and can be considered imitations of the Roth arch form, which was
designed primarily for extraction patients in the 1970s. Therefore, preformed archwires that are approximately
1 to 3 mm wider at the canine level and 2 to 5 mm wider at the first-molar level might be required for today’s
orthodontic needs. (Am J Orthod Dentofacial Orthop 2010;137:520-7)
T
he invention of the edgewise appliance1
in the
late 1920s brought about the consideration of
the relationship between the dental arch form
of a normal occlusion as a goal of orthodontic treat-
ment on the one hand and the archwire form of ortho-
dontic appliances on the other.2
In 1934, Chuck3
stated, ‘‘With edgewise arch appliance, arch form
becomes the basis of our treatment planning, because
in the use of the appliance we predetermine the arch
form and create an ideal archwire to which we move
the teeth.’’ The classic Bonwill-Hawley type of pre-
formed stainless steel archwire blanks4
were intro-
duced to improve the efficacy of orthodontic
treatment.5
When the preadjusted edgewise appliance
was introduced in the 1970s,6
preformed archwire
blanks for this new system also became available.7
This new preformed archwire, designed by Andrews
based on his extensive research on nonorthodontic nor-
mal occlusion and the clinical experience of Roth,8
met with great success. With the spread of this appli-
ance, a variety of archwire forms have been developed
for various reasons.9-12
To date, little research has been conducted on the ra-
tionale behind the prefabricated archwire form. Most
studies in this area sought stable orthodontic results,
and their findings have not varied.13,14
For example,
Felton et al13
compared 10 kinds of archwires with the
dental arch forms of 30 subjects with untreated normal
occlusions selected from the 120-patient sample of An-
drews.6
They successfully classified all 30 dental arch
forms into 10 archwire shapes and reported that 27%
of the subjects fit the Vari-Simplex form from Ormco,
and 20% were close to the Tru-Arch form from
Ormco/‘‘A’’ Company. On the other hand, Braun
et al14
analyzed 16 popular nickel-titanium (NiTi) arch-
wires and compared them with the mean shape of the
dental arch. They reported that all sampled strong-sell-
ing, preformed archwires were wider than the normal
dental arch form.
From the Department of Orthodontics, Nippon Dental University School of
Life Dentistry, Tokyo, Japan.
a
Assistant professor.
b
Professor and chair.
c
Former professor and chair [deceased].
Partly supported by a grant-aid for Scientific Research from the Ministry of
Education, Culture, Sports, Science and Technology-Japan, No. 07407061
and 11470457.
The authors report no commercial, proprietary, or financial interest in the
products or companies described in this article.
Reprint requests to: Kazuhito Arai, 1-9-20 Fujimi, Chiyoda-ku, Tokyo 102-
8159, Japan; e-mail, drarai@ndu.ac.jp.
Submitted, December 2007; revised and accepted, April 2008.
0889-5406/$36.00
Copyright Ó 2010 by the American Association of Orthodontists.
doi:10.1016/j.ajodo.2008.04.025
520

2. Therefore, it is necessary to study preformed arch-
wire forms to increase the efficacy of tooth movement
and to achieve stability of orthodontic treatment. The
purpose of this study was to evaluate whether the vari-
ous commercially available orthodontic archwire forms
are comparable with the diversity of natural normal den-
tal arch forms.
MATERIAL AND METHODS
From a total population of approximately 3500 stu-
dents at Nippon Dental University, 30 orthodontically
untreated subjects (15 men, 15 women; mean age, 23
years 2 months) were clinically examined according
to the following inclusion criteria: no history of ortho-
dontic treatment; Angle Class I molar relationship
with arch form symmetry and minimal arch-length dis-
crepancy; complete dentition, excluding third molars;
ideal overjet and overbite; no prosthetic crowns and
minimal restorations; minimal signs of occlusal attri-
tion; and balanced facial esthetics (no significant dis-
crepancies noted). About 100 sets of dental casts were
then made, and the most ideal Class I occlusions were
selected for analysis.15
The mandibular arch was ana-
lyzed for this study, because therapeutic possibilities
in the mandible are more limited than in the maxilla,
the maxillary arch form is strongly associated with the
mandible, and also maintaining mandibular canine
width is an essential element in achieving stable ortho-
dontic treatment results.10,13,16,17
Eight kinds of brackets for the central incisor, ca-
nine, and first molar with 0.022-in slots from 7 manu-
facturers were selected to obtain the mean bracket
base thickness (Table I).
Twenty kinds of preformed NiTi archwires from 8
manufacturers were selected for analysis (Table II). Al-
though we evaluated NiTi archwires, each manufacturer
also produces preformed stainless steel archwires iden-
tical in design to the NiTi archwires.
Mean bracket thicknesses were used for the distance
between the facial axis point (FA point) and the bracket
slot point (BS point). A ceramic sphere (diameter, 4.75
mm) was attached to 1 tip of a digital caliper (Digimatic
caliper, NTD12-15C, Mitutoyo, Kawasaki, Japan) to fit
the curvature of the bracket base. Brackets were then
tied with elastic ligatures to a 0.0215 3 0.028-in stain-
less steel wire. The distances between the center of the
bracket mesh base and the base of the bracket slots for
the central incisor, canine, and molar were measured
by the modified caliper in 0.01 mm resolution (Fig 1).
The means and standard deviations of bracket thick-
nesses at the mandibular central incisor, canine, and first
molar were 1.34 6 0.16, 0.75 6 0.11, and 0.73 6 0.08
mm, respectively.
Thirty mandibular dental casts were scanned and an-
alyzed by using a noncontact 3-dimensional orthodontic
cast measuring system (Surflacer VMS-250R, UNISN,
Osaka, Japan).18
The FA point for 14 teeth (from both
sides of the central incisors to the second molars)
were plotted on the surface of a cloud data set from
a 3-dimensionally digitized mandibular dental cast ac-
cording to the original anatomic definition of the
points.6
In addition, the 30 sets of 14 x- and y-coordi-
nates of the FA points for each dental arch were ana-
lyzed by using Excel 2000 (Microsoft, Redmond,
Wash).
The BS point represents the position of the base of
the bracket slot in relation to the tooth. The positions of
the BS points on the labial and buccal sides of the teeth
were defined by first connecting a line between the FA
Table I. The 8 kinds of brackets selected for this study
Bracket Manufacturer
MiniTwin Ormco/‘‘A’’ Company, Orange, Calif
Mini Diamond Ormco, Glendora, Calif
Mini Master American Orthodontics, Sheboygan, Wis
Elite Opti-MIM Ortho Organizers, San Marcos, Calif
MicroArch GAC International, Bohemia, NY
Victory 3M Unitek, Monrovia, Calif
Metal Bracket Dentsply, Sankin, Japan
NU Edge-LN TP Orthodontics, LaPorte, Ind
Table II. The 20 preformed archwires selected for this
study
Archwire form Manufacturer
Accu Form GAC International, Bohemia, NY
Broad Arch (large) Ormco, Glendora, Calif
Broad Arch (small) Ormco, Glendora, Calif
Natural Arch Form I American Orthodontics, Sheboygan, Wis
Natural Arch Form II American Orthodontics, Sheboygan, Wis
Natural Arch Form III American Orthodontics, Sheboygan, Wis
Ortho Form I 3M Unitek, Monrovia, Calif
Ortho Form II 3M Unitek, Monrovia, Calif
Ortho Form III 3M Unitek, Monrovia, Calif
Orthos (large) Ormco, Glendora, Calif
Orthos (small) Ormco, Glendora, Calif
Pro Form Ortho Organizers, San Marcos, Calif
Smooth Arch Form Oral Care, Tokyo, Japan
Standard Form GAC International, Bohemia, NY
Tru-Arch (medium) Ormco/‘‘A’’ Company, Orange, Calif
Tru-Arch II (medium) Ormco/‘‘A’’ Company, Orange, Calif
Tynilloy (large) Dentsply-Sankin, Tokyo, Japan
Tynilloy (small) Dentsply-Sankin, Tokyo, Japan
Vari-Simplex (large) Ormco, Glendora, Calif
Vari-Simplex (small) Ormco, Glendora, Calif
American Journal of Orthodontics and Dentofacial Orthopedics Oda, Arai, and Nakahara 521
Volume 137, Number 4

3. points on the 2 adjacent teeth on both sides of the tooth
of interest. A line perpendicular to this axis was then
extended so that it passed through the FA point on
the tooth of interest. The BS points were then estab-
lished at the outward distance of the mean bracket
thickness from the corresponding FA point (Fig 2).
An axis was extended between the BS points on the
central incisors, and a reference point was established
on the axis halfway between the BS points. This mid-
point served as the point of reference from which the
depths of the canine and the first molar could be mea-
sured (Fig 2). The means and standard deviations of
the canine and the first-molar depths of the 30 normal
subjects were 5.09 6 0.70 and 26.95 6 1.56 mm,
respectively.
The 20 kinds of archwires were scanned with milli-
meter gauges by using a flatbed scanner (ES-2000,
Epson, Suwa, Japan) at a resolution of 600 dpi (23.6
dots per mm) (Fig 3) by using image-editing software
(version 7.0, Photoshop, Adobe, San Jose, Calif). The
images were then analyzed by using image processing
and analysis software (Scion Image, National Institutes
of Health, Bethesda, Md). The points on the lingual sur-
face at which the archwire and the line parallel to the
line connecting the distal ends of the wire intersected
with the level of the mean canine and first-molar depths
were digitized. The distance between the bilateral inter-
section points at the canine and first-molar levels was
calculated as the width of the archwire. The means
and standard deviations of the archwire widths at the ca-
nine and first-molar levels of 20 preformed archwires
were calculated and statistically analyzed.
The absolute error of the FA point identification on
dental casts with the 3D scanning system we used was
less than 0.08 mm.15
All sample archwires were measured twice in
randomized order more than 2 months later. The mea-
surement error was evaluated according to Dahlberg’s
formula.19
The errors of the method of measurement on
the scanned images of archwires were 0.04 mm for canine
width and 0.02 mm for molar width. The means of the 2
repeated measurements were used for the analysis.
Statistical analysis
Means and standard deviations; median, lower
(first) quartile, 25th percentile (Q1), and upper (third)
quartile,75th percentile (Q3); and the differences in
the median values of the arch widths were calculated.
The nonparametric Mann-Whitney test was used to an-
alyze the average differences in canine and first-molar
widths at the 5% and 1% levels. A nonparametric test
was used for this study because the distributions of
Fig 1. Method for measuring bracket thickness.
Fig 2. Definition of the bracket slot (BS) points, and the
canine and first-molar width and depth of the normal
dental arch.
Fig 3. Calculations of archwire width at the canine and
first-molar levels.
522 Oda, Arai, and Nakahara American Journal of Orthodontics and Dentofacial Orthopedics
April 2010

4. archwire widths among the 20 commercial products did
not fit normal distribution. Additionally, the distribu-
tions of preformed archwire widths at the canine and
first-molar levels were graphically compared with
means and standard deviations of normal dental arch
widths, because the distribution of natural dental arch
widths is usually considered normal distribution.
RESULTS
Table III shows the means and standard deviations,
medians, maximums, and Q1 and Q3 values of the ca-
nine and first-molar widths of the 30 normal subjects
and the 20 preformed archwires. Statistically significant
differences between the subjects and the archwires were
observed for the widths of both canine and first molar
(canine width: U 5 188.5, P 5 0.0272; molar width:
U 5 92.0, P 0.0001 [U is the calculation of a statistic
in a non-parametric Mann-Whitney test, given by the
formula: U 5 R-n(n11)/2, where n is the total number
of observations, and R is the sum of the ranks in the
sample]).
Figure 4 shows the distribution of widths at the ca-
nine level of the 20 archwires compared with the means
and standard deviations of the normal dental arch width,
and Figure 5 shows the distribution of the widths at the
first-molar level. Fourteen of the 20 archwires had
widths at the canine level within 1 SD of the mean for
the normal sample. In contrast, only 7 archwires were
within 1 SD of the mean for normal first-molar width.
DISCUSSION
Previous studies showed that dental arch width
changes with age. Arch width rapidly increases espe-
cially during the mixed dentition.20
However, most cur-
rent orthodontic treatments include the second molars.
Reported results of average arch-width changes after
the eruption of the second molars were varied, eg, arch
width was constant,20
increased for male subjects and
Table III. Comparison of the arch widths at the canines and first molars between the normal dental arch and preformed
archwire, with the results of nonparametric statistical analyses (mm)
Normal dental
arch (mm) Preformed archwire (mm)
Mean (SD) Median Q1 Q3 Mean (SD) Median Q1 Q3 Difference between medians
Canine width 30.13 (1.32) 30.13 29.43 30.93 29.05 (1.56) 29.79 28.38 30.02 0.34*
First-molar width 55.28 (2.66) 54.90 53.24 57.06 51.96 (2.00) 52.27 51.00 52.88 2.62†
*P 0.05; †
P 0.01.
Fig 4. Distribution of the canine widths for each archwire compared with the means and standard
deviations of the normal dental arch, including mean bracket thickness.
American Journal of Orthodontics and Dentofacial Orthopedics Oda, Arai, and Nakahara 523
Volume 137, Number 4

5. decreased for female subjects,21
decreased for all sub-
jects,22
and did not change or decreased slightly.23
Ad-
ditionally, these changes in arch width during
adolescence were smaller than 1 mm. Therefore, the
adult dental arches used in this study were adequately se-
lected.
It is necessary to consider the variation of bracket
thicknesses to analyze the horizontal relationship be-
tween tooth surfaces and archwires.14,24
Our results
showed that the maximum difference in the bracket
thicknesses between brands was 0.80 mm for the man-
dibular incisors. Anteroposterior position of the arch-
wire at the incisors moves because of the variation in
bracket thicknesses, and consequently archwire widths
at the canine and first-molar levels changed. Therefore,
in this study, the mean values of bracket thicknesses of
popular products were used to evaluate the general rela-
tionship between variation of archwire forms and natu-
ral diversity of normal occlusion. This method can be
used to analyze the influence of bracket thickness for
archwire width evaluation without damaging the fragile
dental casts by replacement of many bracket sets.
Today, a wide variety of preformed archwires are
available for orthodontics. An orthodontist theoretically
selects the size and shape of preformed archwires to
match the patient’s original dental arch forms as closely
as possible to achieve a stable treatment result. Using
preformed archwires that fit the most commonly ob-
served natural dental arch forms minimizes the need
for adjustments to the archwires’ curvatures and reduces
the patient’s chair time. Therefore, our primary objec-
tive was to determine which commercially available or-
thodontic archwire form most accurately fits the most
common natural dental arch forms. Figures 4 and 5
show the distribution of the canine and first-molar
widths of the 20 archwires compared with the normal
dental arch forms. The width of most of the sampled
archwires was narrower than the mean of normal dental
arch widths. At the canine level, only 2 of the 20 arch-
wires were wider than the mean of the normal dental
arch width, whereas, at the first-molar level, only 1 arch-
wire was wider than the mean of the normal dental arch
width. Archwires with overall width measurements
closest to the mean of the normal dental arch widths
were the Orthos large and small (Ormco, Glendora, Ca-
lif), and Vari-Simplex large (Ormco). Although the de-
tails are still unclear, the Orthos archwire forms appear
to have been designed solely according to the results of
research conducted on 100 orthodontic patients in the
United States and might not have included adjustments
based on clinical findings.25
This result supports previ-
ously published findings, and, therefore, these arch
forms are considered suitable for orthodontic patients
in both Japan and the United States.13,25
In contrast, it
is obvious that orthodontists who need expanded arch-
wires for some reason, or patients who have wider
arches than the average might require wider preformed
archwires than the current selection in the market.
The average widths of the 20 archwires at the canine
and first-molar levels were narrower than normal dental
Fig 5. Distribution of the first-molar widths of the archwires compared with the means and standard
deviations of the normal dental arch, including mean bracket thickness.
524 Oda, Arai, and Nakahara American Journal of Orthodontics and Dentofacial Orthopedics
April 2010

6. arch widths measured in this study. The differences in
median values between the dental and archwire widths
at the canines and first molars were 0.34 and 2.62
mm, respectively, and were statistically significant
(P0.05 and P0.01). In contrast to our results, Braun
et al14
reported that all sampled preformed archwires
were wider than the normal dental arch form. Sixteen
popular NiTi archwires were indirectly compared with
the mean shape of the dental arch as represented by
a mathematical curve with a beta function. The reported
mean differences between the dental and archwire
widths at the mandibular canine and first molar were
5.95 and 0.84 mm, respectively, wider than human den-
tal arch.14
Beta function is defined by the width and
depth of a dental arch at the molar area. But as Noroozi
et al26,27
stated, we can have 3 dental arches of ortho-
dontic patients with different canine widths, but their
molar depths and widths are the same. So we will
have the same beta function model for the 3 arches, al-
though they have completely different shapes. There-
fore, it might be possible that a beta function is not
flexible enough to evaluate the natural dental arch
form, especially in the canine area.26-28
One genetic factor contributing to the dental arch
form is the patient’s ethnic background, which is often
considered when selecting preformed archwire blanks.29
Therefore, ethnic differences should also be considered
when evaluating the results of this study. We compared
the widths of commercially available orthodontic arch-
wires with normal dental arch forms in Japanese dental
students. However, the original form for the straight-
wire appliancewas determined based on the mean dental
arch form of an orthodontically untreated normal occlu-
sion sample in the United States.6
Furthermore, most or-
thodontic archwires were designed in the United States
and have been distributed all over the world without
much research. Recently, it was reported that, in compar-
ison with age- and sex-matched white subjects, the mean
dental arch widths of Japanese Class I subjects with mal-
occlusions were approximately 0.9 and 1.5 mm wider at
the canine and molar levels, respectively.30
It is impossi-
ble to directly compare the results of these studies be-
cause of differences in sample selection, different
definitions of reference points, and the consideration
of bracket base thickness. Thus, when ethnic differences
between subjects from the United States and Japan are
considered, the averagewidth of the preformed archwire
forms seems to still be 1.5 to 2 mm narrower than the nat-
ural dental arch form at the first molars. Future studies
should evaluate additional ethnic groups by using this
new method of analysis.
While conducting this research, we collected sam-
ple wires and asked Japanese corporate distributors of
the 4 brands of 3 major orthodontic manufacturers about
trends in the sale of preformed archwires. Figure 6
shows the archwires and widths of the most popular
NiTi archwires. The differences in width between the
main archwire products from the major companies
were all within about 1 mm at the canine level and 2
mm at the first-molar level. These wires are so similar
because the Roth prescription remains the most com-
mon,30
and these wires can be all imitations of the orig-
inal Roth arch form. In 1934, Chuck3
wrote that his goal
in orthodontic treatment was to use the edgewise appli-
ance to align the teeth according to the shape of the
archwire. Because he had treated few premolar extrac-
tion patients, he did not need to consider the space clo-
sures of premolar extractions. In contrast, Roth’s
treatment goal was not directly equal to his archwire
form. It is widely accepted that, when the original
Tru-Arch form was designed about 1970, the frequency
of extractions was much higher than in the 1930s or to-
day.16,31,32
In general, during extraction-space closure,
the posterior teeth tend to rotate into the extraction
site. Therefore, the Roth arch form includes overcorrec-
tion on the brackets and a narrower posterior arch width
to counter the archwire’s tendency to curve outward at
the molar level in extraction patients.8
It remains to be
seen whether the archwire form developed in the
1970s, when extraction orthodontic treatment was
most popular, can be used successfully in the future,
Fig 6. Main products for each brand. Note the close re-
semblance between the 4 archwires. The archwire
widths of these products are all within a narrow range
and were close to the dimensions of the traditional
Roth arch form.
American Journal of Orthodontics and Dentofacial Orthopedics Oda, Arai, and Nakahara 525
Volume 137, Number 4

7. when nonextraction treatment might be more common.
To be successful, the form must maintain the same level
of efficiency of treatment by minimizing the treatment
duration, the patient’s chair time, and the stable occlu-
sion. Therefore, more research is needed to elucidate
this matter.
This study confirmed that the range of selection in
current commercially available preformed orthodontic
archwire forms does not appropriately cover the re-
quired natural human diversity. For example, a major
orthodontic technique recommends the 3 shape varia-
tions of tapered, square, and ovoid, shown in Figures
4 and 5 as Ortho Forms I, II, and III (3 M Unitek,
Monrovia, Calif), respectively.33
However, even for
the widest shape, the canine width of the square arch-
wire was narrower than the average of the normal var-
iation. Additionally, these 3 archwire shapes were all
narrower than the mean: 1 SD at the molar level. Al-
though we compared preformed archwires with normal
samples as a reference group, the diversity in malocclu-
sions is apparently greater.34-36
Hence, the variation in
prescribed categories of archwire templates might
require not only shape selection but also a wider size
variation for clinical applications. Also, further study
comparing the prefabricated archwire form with the
pretreatment dental arch forms of orthodontic patients
could be required.
These facts suggest that, even if an ethnically
matched preformed archwire is selected for a patient
from the currently available variations, archwire ad-
justments are often necessary. Therefore, even with
the latest orthodontic appliances, education in the bi-
ologic diversity of our patients and reasonable tech-
nical training for archwire fabrication and
adjustment are still essential in advanced orthodontic
programs.
CONCLUSIONS
1. The average width of preformed archwires is nar-
rower than the average width of the normal dental
arch form.
2. The variation in available preformed archwires
does not entirely correspond with the range of di-
versity of natural arch form widths.
3. The preformed archwires with measurements that
most closely matched the dental arch form of the
sampled normal subjects were the Orthos and
Vari-Simplex large types.
4. Although there are many commercially available
archwires with varying arch widths, the most
commonly used archwires are similar and can
be considered imitations of the Roth arch form,
which was designed with clinical modifications
primarily for extraction patients in the 1970s,
when extraction orthodontic treatment was at
a peak.
5. Therefore, preformed archwires that are approxi-
mately 1 to 3 mm wider at the canine level and 2
to 5 mm wider at the first-molar level might be
required for today’s orthodontic needs.
REFERENCES
1. Angle EH. The latest and best in orthodontic mechanism. Dent
Cosmos 1928;70:1143-58.
2. Angle EH. Treatment of malocclusion of the teeth. Philadelphia:
S.S. White Dental Manufacturing; 1907. p. 21-4.
3. Chuck GC. Ideal arch form. Angle Orthod 1934;4:312-27.
4. Hawley CA. Determination of the normal arch, and its application
to orthodontia. Dent Cosmos 1905;47:541-52.
5. Renfroe EW. Edgewise. Philadelphia: Lea & Febiger; 1975. p.
113.
6. Andrews LF. Straight-wire: the concept and appliance. San Diego:
L Wells; 1989. p. 159.
7. Roth RH. Treatment mechanics for the straight wire appliance. In:
Graber TM, Vanarsdall RL Jr, editors. Orthodontics, current prin-
ciples and techniques. 2nd ed. St Louis: Mosby; 1994. p. 685-711.
8. Roth RH. The straight-wire appliance 17 years later. J Clin Orthod
1987;21:632-42.
9. Brader AC. Dental arch form related with intraoral forces: PR 5
C. Am J Orthod 1972;61:541-61.
10. Ricketts RM. Design of arch form and details for bracket place-
ment. Denver: Rocky Mountain Orthodontics; 1979.
11. McLaughlin RP, Bennett JC. Arch form considerations for stabil-
ity and esthetics. Rev Esp Ortod 1999;29(Suppl 2):46-63.
12. McKelvain GD. An arch form designed for use with a specific
straight wire orthodontic appliance [thesis]. Dallas: Baylor Uni-
versity; 1982.
13. Felton JM, Sinclair PM, Jones DL, Alexander RG. A computer-
ized analysis of the shape and stability of mandibular arch form.
Am J Orthod Dentofacial Orthop 1987;92:478-83.
14. Braun S, Hnat WP, Leschinsky R, Legan HL. An evaluation of the
shape of some popular nickel titanium alloy preformed archwires.
Am J Orthod Dentofacial Orthop 1999;116:1-12.
15. Kobayashi M, Arai K. Ishikawa. Vertical difference between two
interpretations of the FA point. Orthod Waves 1999;58:21-26.
16. Proffit WR, Fields HW. Contemporary orthodontics. 3rd ed. St
Louis: Mosby; 2000. p. 240-93.
17. Little RM. Stability and relapse of dental arch alignment. Br
J Orthod 1990;17:235-41.
18. Arai K, Ishikawa H. Application of non-contact three-dimen-
sional shape measuring system to dental cast—reduction of blind
region. Orthod Waves 1999;58:148-53.
19. Dahlberg G. Statistical methods for medical and biological stu-
dents. New York: Interscience Publications; 1940. p. 90-3.
20. Moorrees CFA. The dentition of the growing child. Cambridge,
Mass: Harvard University Press; 1959.
21. Knott VB. Size and form of the dental arches in children with
good occlusion studied longitudinally from age 9 years to late ad-
olescence. Am J Phys Anthropol 1961;19:263-84.
22. Sinclair PM, Little RM. Maturation of untreated normal occlu-
sion. Am J Orthod 1983;83:114-23.
23. Bishara SE, Treder JE, Jakobsen JR. Facial and dental changes in
adulthood. Am J Orthod Dentofacial Orthop 1994;106:175-86.
526 Oda, Arai, and Nakahara American Journal of Orthodontics and Dentofacial Orthopedics
April 2010

8. 24. Knox J, Jones M, Durning P. An ideal preformed archwire?
J Orthod 1993;20:65-70.
25. Andreiko Craig. DDS, MS, on the Elan and Orthos Systems.
Interview by Dr. Larry W. White. J Clin Orthod 1994;28:459-68.
26. Noroozi H, Djavid GE, Moeinzad H, Teimouri AP. Prediction of
arch perimeter changes due to orthodontic treatment. Am J Orthod
Dentofacial Orthop 2002;122:601-7.
27. NorooziH,NikTH,SaeedaR.Thedentalarchformrevisited.Angle
Orthod 2001;71:386-9. Erratum in: Angle Orthod 2001;71:525.
28. Seba A, Eman AA, Abdulaziz A. Mathematical analyses of dental
arch curvature in normal occlusion. Angle Orthod 2008;78:281-7.
29. Nojima K, McLaughlin RP, Isshiki Y, Sinclair PM. A comparative
study of Caucasian and Japanese mandibular clinical arch form.
Angle Orthod 2001;71:195-200.
30. Keim RG, Gottlieb EL, Nelson AH, Vogels DS III. 2002 JCO study
of orthodontic diagnosis and treatment procedures, part 3: more
breakdowns of selected variables. J Clin Orthod 2002;36:690-9.
31. Peck S, Peck H. Frequency of tooth extraction in orthodontic
treatment. Am J Orthod 1979;76:491-6.
32. Proffit WR. Forty-year review of extraction frequencies at
a university orthodontic clinic. Angle Orthod 1994;64:
407-14.
33. McLaughlin RP, Bennett JC, Trevisi H. Systemized orthodon-
tic treatment mechanics. London, UK: Mosby International;
2001. p. 71-85.
34. Huth J, Staley RN, Jacobs R, Bigelow H, Jakobsen J. Arch widths
in class II-2 adults compared to adults with class II-1 and normal
occlusion. Angle Orthod 2007;77:837-44.
35. Uysal T, Memili B, Usumez S, Sari Z. Dental and alveolar arch
widths in normal occlusion, class II division 1 and class II division
2. Angle Orthod 2005;75:941-7.
36. Uysal T, Usumez S, Memili B, Sari Z. Dental and alveolar arch
widths in normal occlusion and Class III malocclusion. Angle
Orthod 2005;75:809-13.
American Journal of Orthodontics and Dentofacial Orthopedics Oda, Arai, and Nakahara 527
Volume 137, Number 4