progress in orthodontics 1 1 ( 2 0 1 0 ) 13–19

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/pio

Original article

Do the neoclassical canons still describe the beauty of faces?

An anthropometric study on 50 Caucasian models

Ferruccio Torsello ∗ , Lea Mirigliani, Raoul D’Alessio, Roberto Deli

Department of Orthodontics (Head: Prof. R. Deli) “Sacred Heart”, Catholic University of Roma, Italy

a r t i c l e i n f o a b s t r a c t

Article history: Objective: Since a long time has gone from their definitions, the authors’ aim was to deter-
Received 23 February 2009 mine whether or not the neoclassical canons are still valid to describe the facial beauty in
Accepted 9 December 2009 Italian women.
Materials and methods: The photographs of 50 italian models selected for an important beauty
Keywords: contest were obtained. All the pictures were taken by a professional photographer with a
Facial attractiveness professional digital reflex camera with a millimeter-graduated scale white background. All
Neoclassical facial canons the pictures were printed in 1:1 dimensions. Eleven neoclassical canons were measured and
Anthropometry the results statistically evaluated.
Orthodontics Results: The results of the present study showed that the medium face third is reduced when
Photogrammetry compared to the upper and the lower thirds and that the two upper quarters of the face are
decreased if compared to the two lower quarters (p < 0.01). The nose height was smaller than
ear height and the nose width smaller than 1/4 of facial width. In 72% of cases the ear axis
was steeper than the nose. The intercanthal width in this study resulted smaller than nose
width and smaller than eye width (p < 0.05). Mouth width was greater than 150% of nose
width.
Conclusions: It can be concluded that some of the neoclassical canons can be considered still
valid, while others seem to be changed over centuries. In particular it seems that relatives
reductions in facial medium third, in nose dimensions and in distance between eyes have
occurred as well as relative enlargement of eyes and mouth width.
© 2010 Società Italiana di Ortodonzia SIDO. Published by Elsevier Srl. All rights reserved.

or maxillofacial surgical techniques, our goal is to obtain aes-

1. Introduction
Measuring head and facial parameters allows us to obtain
valuable indications and diagnostic criteria as well as infor-
mation potentially usable in clinical practice and art. In
reconstructive and cosmetic surgery, proportions are assessed
using anthropometrics techniques and used as guidelines to
correct deformities and/or disproportions. Using orthodontic
thetically superior results for our patients. To judge the appeal
of a patient’s face it is compared with beauty norms that today
are well defined by canons or anthropometric proportions. The
availability of values for facial sizes enables us to reproduce
cosmetically attractive proportions for our patients.1
In the early years of orthodontics Angle paradigm stated
that a perfect occlusion should be achieved and then facial
beauty was thought to follow.2 In the following years the

∗
Corresponding author. via Simone de Saint Bon 42 - 00195 Roma, Italy.
E-mail address: ferruccio.torsello@gmail.com (F. Torsello).
1723-7785/$ – see front matter © 2010 Società Italiana di Ortodonzia SIDO. Published by Elsevier Srl. All rights reserved.
doi:10.1016/j.pio.2010.04.003
14 progress in orthodontics 1 1 ( 2 0 1 0 ) 13–19
growth of cephalometrics focused orthodontists’ attention on
dentofacial structures.3–5 More recently it has been recognized
that soft tissue profile proportions and harmony should be the
goals of orthodontic and surgical treatments.6,7 For this rea-
son today, facial analysis is a fundamental step in evaluation
of patients who require cosmetic or reconstructive procedures
on the face. Anthropometric and soft tissue analysis has been
used to define soft tissue relationships and standards for the
purposes of preoperative objective facial analysis based on
standardized photogrammetric methods.8–11 More recently,
the evolution of technology has provided us with several,
precise, three dimensional methods to measure anthropo-
metric parameters such as facial plaster scans, digitizing of
anatomical landmarks recoded directly on patients and facial
laser surface scanning.12–15 These promising techniques are
still not widespread because of the expensive technology
required. Thus in the daily orthodontic practice the photogra-
phy remains the most common way to deal with facial analysis
and the neoclassical canons are still considered important ref-
erences for treatment planning in orthodontic and surgical
treatments.7,16,17
Since a long time has gone from their definitions, the
authors’ hypothesis is that the flowing of time could have
modified the aesthetic goals. For this reason a study has been
designed in order to determine whether or not the neoclassi-
cal canons, defined 5 centuries ago, are still valid in a sample
of 50 female models participating to an international beauty
contest.
2. Materials and methods
The digital photographs of 50 female models (age 16-25 years
old) participating to the final selection of the Italian repre-
sentative for an important international beauty contest were
obtained. All the models were white caucasian of Italian
descent and were selected among thousands of girls that
applied for this contest through previous local and regional
selections. Two to three models for each Italian region were
allowed to participate to the final stage of the contest.
The frontal and the right profile pictures taken with relaxed
face and with lips in rest positions were collected for this
investigation.
All the pictures were taken by a professional photogra-
pher with a professional digital reflex camera with macro lens
Fig. 1 – Canons in the lateral view.
(105 mm) in the same environment with the same conditions
of light. The pictures have been taken without any make-up
and with a millimeter-graduated scale white background. All
the pictures were printed on photographic paper, calibrating
the dimension so as to have 1:1 measurements on frontal and
profile photographs (this was easily achieved due to the mil-
limetre graduated scale in the background).
Five neoclassical canons were measured on the profile pic-
ture as follows18 :
– Two section facial profile canon: special head height (vertex-
endocanthion or v-en) = special face height (endocanthion-
gnathion or en-gn) (Fig. 1).
– Three section facial profile canon: forehead height II
(trichion-nasion or tr-n) = nose lenght (nasion-subnasale
or n-sn) = lower face height (subnasale-gnathion or sn-gn)
(Fig. 1).
– Four section facial profile canon: height of calvaria (vertex-
trichion or v-tr) = forehead height I (trichion-glabella or
tr-g) = special upper face height (glabella-subnasale or g-
sn) = lower face height (subnasale-gnathion or sn-gn) (Fig. 1).
– Naso aural proportion canon: nose lenght (nasionsubnasale
or n-sn) = ear lenght (supraaurale-subuaurale or sa-sba)
(Fig. 1).
– The nasal bridge inclination is the same as the ear inclina-
tion (Fig. 1).
– Six canons were measured on the frontal picture as follows:
– The orbitonasal canon: intercanthal distance (endo-
canthion-endocanthion or en-en) = nose width (alarealare
or al-al) (Fig. 2).
– The orbital canon:intercanthal distance(endocanthion-
endocanthion or en-en) = left eye fissure lenght
(exocanthion-endocanthion or ex-en) (Fig. 2).
– The naso-oral canon:mouth width (cheilion-cheilion or ch-
ch) = 1 ½ nose width (alare-alare or al-al) (Fig. 2).
– The width of the nose is one-fourth the width of the face
(Fig. 2).
– The lower face is divided in three equal portions (Fig. 3).
– The lower face is divided in four equal portions (Fig. 3).
The aim of the study was to determine if the neoclassical
canons still describe the beauty of female faces.
Statistical analysis has been performed with the aid of a
proper statistical software (SPSS 13.0, SPSS Inc. Chicago, IL,
USA). One way Anova test was performed in order to com-
pare mean values of each measurement. When more than two
groups were considered a Tukey’s Post Hoc Test was performed
 progress in orthodontics 1 1 ( 2 0 1 0 ) 13–19 15

Fig. 2 – Canons in the frontal view.

Fig. 3 – Canons regarding the lower part of the face.
in order to determine the statistical significance between cou-
ples of groups.
The author(s) declare that the work has been realized in
agreement with the Helsinki Declaration principles and that
the Informed Consent has been achieved from all the partici-
pants involved in the study.
3. Results
The mean values, as well as standard deviations and per-
centual proportions are reported in Tables 1, 2 and 3.
(p < 0.01) and that the upper third is slightly increased when
compared to the lower one (p < 0.05). The percentages calcu-
lated are: 35,5% for the upper third, 30,3% for the medium third
and 34,2% for the lower third. The ideal proportion according
to this canon’s description was 33.3% for each third. In the
present sample only one model showed this proportion. In 48
models (96%) the middle third was the smallest of the three
portions.
The comparison of the facial quarters showed that v-tr<tr-
g<g-sn<sn-gn with an overall significance of p < 0.01 (Table 1).
The post-hoc comparison showed that the mean differences
between v-tr and tr-g, v-tr and g-sn, v-tr and sn-gn, tr-g and
g-sn, tr-g and sn-gn were statistically significant (p < 0.01). The
mean difference between g-sn and sn-gn was not statistically
significant (p > 0.05). The same analysis express in percentage
showed that v-tr accounts for 17,4% of total facial height, tr-
g for 26%, g-sn for 28,1% and sn-gn for 28,5% instead of the
expected ratio of 25% for each quarter. The equality of the four
portions was never found in the 50 models.
The comparison of nose and ear length showed a statisti-
cally significant difference (p < 0.01) with the ear length 11,4%
greater than the nose (6,7 mm), instead of an expected equiv-
alence of these two measurements (Table 2). Only in 3 cases
the nose height was greater than the ear (6%). In 2 cases the
lengths were perfectly equal and in 45 (90%) cases the ear
exceed the nose.

3.1. Vertical canons Table 1 – Mean, Standard deviations and percentual

proportions for the canons used to evaluate the face
In our sample, the special head height was significantly divided in half, thirds or quarters. All measurements are
greater than the special face height (p < 0.05). The mean differ- espressed in mm.
ence between v-en and en-go was 2,8 mm (Table 1). When the Canon Measurements
proportions were considered the results showed that the mean Equal halfs V-En En-Gn Difference
difference between head and face height is limited to 1,4%. In Mean 104.6 101.9 2.8
the original description of this canon a 50% to 50% ratio should sd 6.6 4.3 6.6
be considered as the optimal proportion.19 In our sample in 12 % 50.7 49.3 1.4
cases (24%) the two sections were equal or their difference was Equal thirds Tr-N N-Sn Sn-Gn
less than 3 mm, in 12 cases (24%) the face height was greater Mean 60.9 52.0 58.8
than the head height and in 26 cases (52%) the head portion sd 4.0 2.6 3.2
was greater than the face. % 35.5 30.3 34.2

When the facial thirds were compared (Table 1), the statisti- Equal quarters V-Tr Tr-G G-Sn Sn-Gn
cal comparison of mean values showed significant differences Mean 35.9 53.8 58.0 58.8
(p < 0.01). The post-hoc analysis showed that the medium third sd 7.3 4.0 3.2 3.2
% 17.4 26.0 28.1 28.5
is reduced when compared to the upper and the lower third
16 progress in orthodontics 1 1 ( 2 0 1 0 ) 13–19

Table 2 – Mean, Standard deviations. All measurements are espressed in mm.

Canons Measurements
Naso-Aural Nose Height Ear Height Diff. Sig.
Mean 52.0 58.7 6.7 **
Sd 2.6 6.3 11.4

Orbito-nasal Intercanthal Width Nose Width Diff. Sig.

Mean 32.0 33.6 −1.6 *
Sd 2.8 2.5 3.2

Orbital Intercanthal Width Eye Width Diff. Sig.

Mean 32.0 33.4 −1.4 *
Sd 2.8 2.6 3.3

Naso-Oral 150% Nose Width Mouth Width Diff. Sig.

Mean 50.4 52.0 1.6 *
Sd 3.7 4.4 4.3

Nose / 1⁄4 face widhts Nose Width 1

⁄4 Face Width Diff. Sig.
Mean 33.6 34.8 −1.2 *
Sd 2.5 2.2 2.1

*p < 0.05; **: p < 0.01; NS: Not significant.

One of the neoclassical canons dictates that the nose incli- The orbital canon assumed that intercanthal distance
nation should be equal to the ear axis. In our sample it has should be equal to the eye fissure length. This outcome was
been found that the nose was parallel to the ear axis in 9 cases found in 16 cases (32%); in 9 cases the intercanthal distance
(18%). In 36 (72%) cases the ear axis was steeper than the nose exceeded the eye width and in 25 cases (50%) the eye width was
and in the remaining 5 (10%) cases the nose was steeper than greater than the intercanthal distance. Among the 34 models
the ear axis. that showed different lengths of these measurements, in 16
cases (32%) their differences were smaller than 3 mm. Thus
3.2. Horizontal canons it can be concluded that in 64% of the cases the intercanthal
distance and the eye fissure length were equal or slightly dif-
The orbito-nasal canon assumed that intercanthal distance ferent (<3 mm). On average our sample demonstrated that the
should be equal to the nose width. This canon was confirmed intercanthal distance is 1,4 mm reduced if compared to the
in 22% of our cases (11 models), while the intercanthal dis- eye length (p < 0.05) (Table 2).
tance was smaller than nose width in 58% of the cases (29 When the naso-oral canon was analysed, the mouth width
models) and was greater than nose width in 20%10 of the cases. as well as the 150% of the nose width were measured and
It should be noticed that among the 39 models that showed dif- compared. These measurements were equal in 10 cases (20%),
ferent lengths of these measurements, in 18 cases (36%) their in 10 cases 150% of nose width exceeded mouth width and
differences were smaller than 3 mm. Thus it can be concluded in 30 cases (60%) mouth width was greater. In 12 cases their
that in 58% of the cases the intercanthal distance and the nose differences were <3 mm, thus in 22 cases (44%) the two mea-
width were equal or slightly different (<3 mm). On average the surements were equal or slightly different (<3 mm). On average
measurements performed on our samples showed a slight, but a tendency for the mouth width to be greater than 150% of nose
statistically significant difference (p < 0.05) between these two width was found: mouth width was on average 1,6 mm wider
parameters, being the nose width on average 1,7 mm (4,6%) than 150% of nose width and this difference was statistically
wider than the intercanthal distance (Table 2). significant (p < 0.05) (Table 2).
The nasofacial proportion canon demonstrated a tendency
for the facial width to be more than 4 times the width of the
nose. In 12 cases the two measurements were equal (24%), in
26 cases the nasal width was smaller than 1⁄4 of nose width
Table 3 – Mean. Standard deviations and percentual
and in 12 cases nose width exceeded 1⁄4 of facial width. In 18
proportions for the canons used to evaluate the lower
cases their differences were <3 mm, thus in 30 cases (60%) the
part of the face divided in thirds or quarters. All
measurements. two measurements were equal or slightly different (<3 mm).
On average nose width was 1.2 mm smaller than 1⁄4 of face
Canon Measurements
width and this difference was statistically significant (p < 0.05)
Equal thirds Tr-N N-Sn Sn-Gn (Table 2).
Mean 18.9 17.1 22.8
Sd 2.2 1.7 2.5
% 32.1 29.1 38.8 3.3. Lips and mouth
Equal quarters V-Tr Tr-G G-Sn Sn-Gn
Mean 18.0 16.5 11.9 10.9 When the division of the lower face in three equal portions
Sd 2.2 1.7 2.4 2.3 has been verified it was found that the three thirds were equal
% 32.1 29.1 20.2 18.6
in 4 cases (8%) and in 8 cases (16%) the differences between
 progress in orthodontics 1 1 ( 2 0 1 0 ) 13–19
the three portions did not exceed 3 mm. In 46 cases (92%) the
lower part exceeded the middle portion. On average the upper
lip accounted for 32,1% of the lower face while lower lip and
chin represented 29,1% and 38,8% (Table 3). The differences
between the three measurements were statistically significant
(p < 0,01).
The analysis of the lower face divided in four equal parts
showed that this proportion was never found in our sam-
ple and that in all cases the third and the fourth portions
were smaller than the first and the second ones. On average
the four components represented respectively 32,1%, 29,1%,
20,2%, 18,6% (Table 3). These differences were statistically sig-
nificant (p < 0,01).
4. Discussion
Some considerations on the photogrammetric tech-
niques should be addressed. Although several studies
showed that direct anthropometry is more precise than
photogrammetry,20,21 photogrammetric techniques are very
useful to study aesthetic facial proportions as photography is a
widely used, quick, and economical clinical record. It has been
shown that although less precise than anthropometry, the
photography can be used for clinical analyses and comparison
if it is performed with adequate standardization22–24
The aim of the present paper was to verify if the neo-
classical canons are still actual. It should be noticed that the
description of these ideals of facial beauty was done by sev-
eral artists such as Leonardo da Vinci (1452-1519),19 Albrecht
Durer (1471-1528),25 Piero della Francesca (1412-1492),26 Girard
Audran (1640-1703)27 and Johann George Bergmuller (1688-
1762)28 four to six centuries ago.
The authors’ hypothesis was that the cultural and social
modifications occurred through the centuries could have
determined some changes in the perception of beauty. Thus
it has been decided to obtain a sample of present-day good
looking girls. For this reason the photographs of 50 models
participating to the final stage of an important beauty contest
in Italy were collected. It should be noticed that the beauty
contest that selected the models used as study sample was
not focused on facial beauty only. However the strict selec-
tion process from such a huge sample of thousands of italian
girls allowed only models with attractive bodies and faces
to be part the final stage. We remind that only models from
the final stage of the contest were studied in this investiga-
tion.
This study found that special head height accounted for
50,7% of total face height and that special facial height
accounted for 49,3% of total face height. Even though this dif-
ference was shown to be statistically significant, this result
seem to be close to the two-section facial profile canon
originated by Leonardo da Vinci. The data from this study
differ from data reported by another study that found a
ratio of 46% between special head height and total cranial
height and of 54% between special facial height and total
cranial height.28 The discrepancy between these two studies
could be originated by the differences in the studied sam-
ples: this paper deals only with Italian women, while the
previous paper involved north American Caucasian of many
17
different origin (Anglo-Saxons, Latin, German, Slavic and
others).
The sample analysed in the present paper did not corre-
spond to the three section facial profile canon. The upper,
medium and lower thirds accounted respectively for 35,5%,
30,3% and 34,2% instead of 33,3% as expected. The main find-
ing is the smaller medium third when compared to the other
two thirds. Similar results demonstrating a reduced medium
facial third has been found in North American Caucasians and
in Turkish Young adults.29–30
The division of the cranium in quarters originated by
Audran did not correspond to our samples, since each quar-
ter accounted for 17.4% 26%, 28% and 28.6% instead of 25%.
Farkas found similar results since he demonstrated 22.5% 21%,
28.4% and 28.1% for each quarter. It is evident that in both
studies the first and second quarter are smaller than the third
and the fourth and that the upper part (first and second quar-
ter) accounts for ∼43% of total height when the lower part of
the cranium (third and fourth quarter) accounts for ∼57% of
total height. It should be noticed that there is no coincidence
between of upper and lower parts proportions determined
with the Audran’s method and the two section facial profile
canon described by Leonardo because the latter uses point
N instead of point G for the division of the head in two
parts.
The naso-aural proportion canon did not show to be cor-
respondent to our sample: the ear height appeared to be
significantly greater than the nose height. For what concerns
the nasoaural inclination canon it was seen that in the vast
majority of cases the ear axis was steeper than the nose. These
results are similar to those obtained by previous reports.29–31
The horizontal measurements regarding the nose showed
the tendency of mouth width and of facial width to be, respec-
tively, greater than 150% or 400% of nose width.
The decreased medium facial third, the nose height smaller
than ear height and the nose width smaller than 1⁄4 of facial
width indicated that the models had a smaller nose rather
than canons’ description. These findings were also expressed
in a review paper by Cellerino.32
The intercanthal width in this study resulted smaller than
nose width (p < 0.05) and smaller than eye width (p > 0.05).
These results indicated that the models had a shorter dis-
tance between eyes and that the eyes were relatively larger
compared with their distance.
For what concerns lips and mouth two different canons
were described. Piero della Francesca used to divide the lower
face in three equal portions: from subnasale to stomion, from
stomion to mental sulcus and from the mental sulcus to the
lower point of the chin. Durer in 1528 divided the lower face
in four equal parts: the first and the second portions were
the same as Della Francesca’s description, while the third and
fourth portions corresponded to the third of the italian artist.
It is clear that these two canons describe different ideals, since
Durer lower face showed a markedly high chin.31
Some differences were found between the recorded values
and the expected ratio of 33% for each portion. Thus it may be
concluded that, in this sample of italian models, there was a
statistically significant tendency to have a higher chin and a
smaller lower lip. The upper lip roughly corresponded to 1/3
of the lower face as expected.
18 progress in orthodontics 1 1 ( 2 0 1 0 ) 13–19
The Durer’s canon appeared to be more distant from this
models’ sample. According to this artist it should be expected
that each of the four portions accounted for 25% of lower face
height. In our study we found that the upper and lower lips
were significantly greater than the expected 25% and that the
two sections that together represented the chin were much
smaller than 50%. These data are in accordance with a pre-
vious report31 on an american sample of women. This paper
reported data for american caucasian women in general, for
the women perceived as attractive and for that faces that
resulted the best faces. Since our sample was composed by
models our material should be probably compared to the best
american faces that demonstrated 30,3% of lower face height
for upper lip, 26% for lower lip and 43,7% for the chin. Both
materials showed a slightly decreased upper lip, a decreased
lower lip and an increased chin. These finding were more evi-
dent in the american sample.
5. Conclusions
It can be concluded that some of the neoclassical canons can
be considered still valid, while others seem to be changed over
centuries.
In particular it seems that relatives reductions in facial
medium third, in nose dimensions and in distance between
eyes have occurred as well as relative enlargement of eyes and
mouth width.
Conflict of interest
The authors have reported no conflicts of interest.
Riassunto
Obiettivo: Determinare se, a distanza di alcuni secoli dalla loro
definizione, i canoni neoclassici siano ancora validi nel descrivere la
bellezza del viso nelle giovani donne italiane.
Materiali e metodi: Sono state reperite le fotografie di 50 model-
le italiane, selezionate per la fase finale di un importante con-
corso di bellezza nazionale. Tutte le fotografie sono state scattate
da un fotografo professionista con una macchina fotografica digi-
tale “reflex”, con uno sfondo bianco sul quale era presente una scala
millimetrata. Tutte le immagini sono state stampate in dimensioni
reali (1:1). Undici canoni neoclassici sono stati misurati sulle foto e i
risultati analizzati statisticamente.
Risultati: I risultati hanno mostrato che l’altezza del terzo medio
appare ridotta rispetto alle altezze dei terzi superiore e inferiore del
viso; le altezze dei due quarti superiori appaiono ridotte rispetto ai
due quarti inferiori (p < 0,01). L’altezza del naso è stata inferiore
a quella dell’orecchio e la larghezza del naso inferiore a 1/4 della
larghezza del viso intero. Nel 72% dei casi l’asse dell’orecchio è stato
più ripido dell’inclinazione del naso. La distanza tra i due occhi è
risultata minore della larghezza del naso e della larghezza degli occhi
(p < 0,05). La larghezza della bocca è risultata maggiore del 150%
della larghezza del naso.
Conclusioni: Alcuni canoni neoclassici possono essere considerati
ancora validi, mentre per altri il nostro campione sembra indi-
care alcune differenze rispetto alla loro originale definizione. In
particolare i risultati del presente studio sembrano suggerire delle
diminuzioni relative dell’altezza del terzo medio del viso, delle
dimensioni del naso e della distanza tra gli occhi, così come un
relativo aumento delle dimensioni degli occhi e della larghezza della
bocca.
Résumé
Objectif: Puisqu’un long temps a passé de leurs définitions, le but
des auteurs était de déterminer si les canons néoclassiques sont
encore valides pour décrire la beauté faciale dans les femmes ital-
iennes.
Materiaux et méthodes: On a obtenue les photographies de 50
modèles italiens choisis pour un concours de beauté important.
Toutes les photos ont été prises par une photographie professionnelle
avec un appareil-photo réflexe numérique professionnel avec un fond
millimètre-gradué. Toutes les images ont été imprimées dans des
dimensions de 1:1. Onze canons néoclassiques ont été mesurés et les
résultats statistiquement ont été évalués.
Résultats: Les résultats de la présente étude ont prouvé que le tiers
moyen de visage est réduit une fois comparé aux tiers supérieurs et
inférieurs et que les deux quarts supérieurs du visage sont diminués
si comparé aux deux quarts inférieurs (p < 0.01). La largeur de bouche
était plus grande que 150% de la largeur de nez.
Conclusions: On peut conclure que certains canons néoclassiques
peuvent être considérés encore valides, alors que d’autres semblent
être changés au-dessus des siècles. En particulier il semble que il y ait
eu une reduction troisième faciale moyen, des dimensions de nez et de
la distance entre les yeux se sont produites comme l’agrandissement
relatif des yeux et de la largeur de bouche.
Resumen
Objetivo: Desde hace mucho tiempo ha pasado por algunas defini-
ciones, el objetivo de los autores fue determinar si los cánones
neoclásicos siguen siendo válidos para describir la belleza facial en
las mujeres italianas.
Materiales y métodos: Se obtuvieron 50 fotografías de mode-
los italianas seleccionados para un importante concurso de belleza.
Todas las fotos fueron tomadas por un profesional con una cámara
réflex digital profesional, graduada con un milímetro de escala y con
fondo blanco. Todas las imágenes fueron impresas en dimensiones
1:1. Once cánones neoclásicos fueron medidos y los resultados se
evaluaron estadísticamente.
Resultados: Los resultados del presente estudio demostraron que
la cara frente a tercio medio se reduce en comparación con los tercios
superior / inferior y se reducen los dos tercios superiores de la cara
si se compara con las dos partes inferiores (p < 0,01). La altura de
la nariz era más pequeña que la altura del oído y el ancho de la
nariz más pequeña que 1/4 de la anchura facial. En el 72% de los
casos el eje de la oreja era más pronunciado que la nariz. El ancho
intercantal en este estudio resultó menor que el ancho de la nariz y
la más pequeña que el ancho de los ojos (p < 0,05). Ancho de boca fue
mayor del 150% de la anchura de la nariz.
Conclusiones: Se puede concluir que algunos de los cánones neo-
clásicos pueden considerarse todavía en vigor, mientras que otros
parecen ser cambiado durante siglos. En concreto, parece que se
hayan producido reducciones en el tercio medio facial, las dimen-
siones de la nariz y la distancia entre los ojos, así como la ampliación
relativa de los ojos y ancho de la boca.
 progress in orthodontics 1 1 ( 2 0 1 0 ) 13–19 19

references 17. Tolleth H. Concepts for the plastic surgeon from art and
sculpture. Clin Plast Surg 1987;14:585–98.
18. Bashour M. History and Current Concepts in the Analysis of
facial attractiveness. J of Am Soc of Plast Surg 2006;118:741–56.
1. Vegter F, Hage JJ. Clinical Anthropometry and Canons of the
19. Mc Murrich JP. Leonardo da Vinci. The anatomist, 1452–1519. 1st
Face in Historical. Plast Reconstr Surg 2000;106:1090–6.
ed. Baltimore: Williams and Wilkins Co.; 1930. pp. 104–11.
2. Angle EH. Classification of malocclusions. Dental Cosmos
20. Farkas LG, Bryson W, Klotz J. Is photogrammetry of the face
1899;41:248–63.
reliable? Plast Reconstr Surg 1980;66:346–55.
3. Steiner CC. Cephalometrics for you and me. Am J Orthod
21. Ghoddousi H, Edler R, Haers P, Wertheim D, Greenhill D.
1953;39:729–55.
Comparison of three methods of facial measurament. Int J
4. Ricketts RM. A foundation for cephalometric
Oral Maxillofac Surg 2007;36:250–8.
communication. Am J Orthod 1960;46:330–57.
22. Bishara SE, Cummins DM, Jorgensen GJ, Jakobsen JR. A
5. Merrifield LL. The profile line as an aid in critically
computer assisted photogrammetric analysis of soft tissue
evaluating facial esthetics. Am J Orthod 1966;52:804–22.
changes after orthodontic treatment. Part I: Methodology
6. Sarver DM. Esthetic orthodontics and orthognatic surgery. St.
and reliability. Am J Orthod Dentofacial Orthop 1995;107:633–9.
Louis: Mosby; 1998.
23. Han K, Kwon HJ, Choi TH, Kim JH, Son D. Comparison of
7. Proffitt WR, White RP, Sarver DM. Contemporary Treatment of
anthropometry with photogrammetry based on a
dentofacial deformity. St. Louis: Mosby; 2003. pp. 109–20.
standardized clinical photographic technique using a
8. Arnett GW, Bergman RT. Facial keys to orthodontic diagnosis
cephalostat and chair. J Craniomaxillofac Surg 2010;38:96–107.
and treatment planning. Part I. Am J Orthod Dentofacial Orthop
24. Malkoç S, Demir A, Uysal T, Canbuldu N. Angular
1993;103:299–312.
photogrammetric analysis of the soft tissue facial profile of
9. Arnett GW, Bergman RT. Facial keys to orthodontic diagnosis
Turkish adults. Eur J Orthod 2009;31:174–9.
and treatment planning. Part II. Am J Orthod Dentofacial
25. Durer A. Les quatre livres d’Albert Durer, peinctre et
Orthop 1993;103:395–411.
geometrien tres excellent, de la proportion de parties
10. Peck S, Peck L. Selected aspects of the art and science of
ourtraicts de corps humains. Paris: C. Perrier, 1557.
facial esthetics. Seminar Orthod 1995;1:105–26.
26. Della Francesca P. Study of the proportions of the head. In:
11. Riveiro PF, Chamosa ES, Quintanilla DS, Cunqueiro MS.
Berry WA, editor. Drawing of the human form : Methods, sources,
Angular photogrammetric analysis of the soft tissue facial
Concepts. New York: Van Nostrand Reinhold; 1977. p. 116.
profi le. Eur J of Orthod 2003;25:393–9.
27. Audran G. Proportiond du corps humain, mesurées sur les
12. Ferrario VF, Sforza C, Poggio CE, Serrao G. Facial
plus belles figures de l’antiquitè. Paris. 1683.
three-dimensional morphometry. Am J Orthod Dentofacial
28. Bergmumuller JG. Anthropometria. Augsburg: JJ Lotter; 1723.
Orthop 1996;109:86–93.
29. Farkas LG, Hreczko TA, Kolar JC, Munro IR. Vertical and
13. Sforza C, Laino A, D’Alessio R, Grandi G, Dellavia C, Tartaglia
Horizontal Proportions of the Face in Young Adult North
GM, Ferrario VF. Three-dimensional facial morphometry of
American Caucasians: Revision of Neoclassical Canons. Plast
attractive Italian women. Prog Orthod 2007;8:282–93.
Reconstr Surg 1984;75:328–38.
14. Da Silveira AC, Daw Jr JL, Kusnoto B, Evans C, Cohen M.
30. Bozkir MG, Karakas P, Oguz O. Vertical and horizontal
Craniofacial applications of three-dimensional laser surface
neoclassical facial canons in Turkish young adults. Surg
scanning. J Craniofac Surg 2003;14:449–56.
Radiol Anat 2004;26:212–9.
15. Holberg C, Schwenzer K, Mahaini L, Rudzki-Janson I.
31. Farkas LG, Kolar JC. Anthropometrics and art in the
Accuracy of facial plaster casts. Angle Orthod 2006;76:605–11.
aesthetic of woman’s faces. Clin Plast Surg 1987;14:559–616.
16. Tolleth H. Observations on form and Proportions. In: Hetter
32. Cellerino A. Psychobiology of facial attractiveness. J
G, editor. Lipolysis: Theory and practice. Boston: Little Brown;
Endocrinol Invest 2003;26:45–8.
1984.