the dental and facial skeletal characteristics and growth of males and females with Class II, division 1 malocclusion between the ages of 10 and 14 (revisited) -- Part I: Characteristics of size, form and position

--Part II: Vertical and Anteroposterior  growth between the ages of 10 and 14  

 by Ted Rothstein DDS, PhD and Cecile Yoon-Tarlie DDS, MS, as it appeared in the American Journal of Orthodontics and Dentofacial Orthopedics, March 2000, Vol. 117, No. 3, pp. 320-332, and reproduced with permission from Mosby, Inc.,  Aug. 2nd 2000. This copyrighted material may be used for personal use only and may not be distributed further.

As it is with all practitioners in the healing arts, an orthodontist’s primary objective is to describe and diagnose the malocclusion he or she wishes to treat in conformity with the primary dictum of medicine: primum, non nocere ("first do no harm"). Fortunately, most of the orthodontic patients are initially seen with an Angle Class I malocclusion including a Class I skeletal relationship. As a result, the patient is described as having a balanced and harmonious skeletal and dental relationship. In such situations, the concern of detrimental effects resulting from mechanotherapy are less of an issue. However, the moderate to severe Class II and Class III cases—as defined by a greater than 4-mm anteroposterior dental discrepancy—often alert the clinician to a more severe underlying cause. Usually the problem is a result of the size, form, or position of the jaw. To a modest (but meaningful) extent, the clinician can restrict or promote maxillary growth, as well as enhance or retard mandibular growth.¹ The particular situation may warrant the use of (or call for the avoidance of) a particular treatment or perhaps mandate surgical intervention to produce an ideal result.

The Class II, Division I malocclusion is a frequently treated problem. Estimates of its occurrence, as seen in the clinician’s office, range from 5% to 29% of the patient population.^2,3 Almost two thirds of the patients with Class II, Division 1 malocclusions were estimated in one study to have an associated skeletal dysplasia of significant importance.² Children and ado-lescents who are seen with this problem may encounter social and psychological problems, in that they live in a culture that places an emphasis on pleasing physical attributes and facial conformity.⁴ In addition, one may sometimes find a higher incidence of caries, periodontal disease, speech problems, and chewing and swallowing difficulties in these children and adolescents. Rarely will a malocclusion correct itself with age.

Class II malocclusions were initially characterized, in 1899, by Edward Angle.⁷ Assuming that the maxillary first permanent molar was stable in the anteroposterior relationship with respect to the cranium, Angle based his classification on the mandibular position as defined by the lower first permanent molar with respect to the upper first permanent molar.⁸ In a Class II molar relationship, the lower first permanent molar is more posterior to the upper molar first permanent molar. Angle believed that this was a result of either a short, underdeveloped lower jaw or a posterior positioned lower jaw. Numerous researchers have investigated Class II malocclusions, attempting to identify the main culprit responsible for this dysplasia. Reviews, summaries, critiques. and their findings have been reported by several authors.^9-12 In brief, various characteristics have been attributed to the Class II malocclusion, ranging from lower jaw deficiency to upper jaw protrusion. One researcher concluded that Class II, Division 1 malocclusions consisted of a random association of upper and lower facial skeletal and dental configurations.¹³ Historically, the predominant findings have reported that the discrepancy is caused by mandibular retrusion— with some notable exceptions.^14-18 Pancherz¹² noted a finding of mandibular retrognathia based on the angular measurements SNA, SNB, SN-Pog; however, he recognized that the use of these measurements may have given erroneous results. Other authors have noted this possibility, 10,19-20 yet they justify using these measurements because, in the words of one, ‘Their usage facilitates the communication between clinicians as well as between research workers." Thus, investigators began to acknowledge various problems underlying their mixed findings.

There were other problems associated with these studies. First, they were statistically unreliable as a result of insufficient sample size. Second, the samples themselves were also mixed, with no distinction made regarding genders and ages. Third, the various samples

Table I. Sample description

 Control Samples

     N=47 Females   age=10 ± 6mo   and      48 Males 10 ± 6 mo

    N=42 Females    age=12 ± 6 mo  and      43 Males 12 ± 6 mo

    N=48 Females    age=14 ± 6 mo  and     45 Males 14 ± 6 mo

Class II, Division 1 Samples

    55 Females    age=10 ± 6 mo         and     51 Males    age=10 ± 6 mo

    58 Females    age=12 ± 6 mo       and      71 Males    age=12 ± 6 mo

    51 Females    age=14 ± 6 mo       and      50 Males    age=14 ± 6 mo

Both groups (control and Class II Division I ) had 6 subgroups 3 male and 3 female.

were grouped according to chronological age, rather than skeletal age. Fourth, the common use of certain angular measurements considered to be reliable indicators of jaw position were later proved less reliable.^13,20 Lastly, there were problems in communicating the results of the studies. For example, descriptive words used to describe the mandible included "retruded," "retrusive," "retropositioned," "distally positioned," "underdeveloped," and "posteriorly positioned."

The present study was undertaken as an attempt to correct for these discrepancies. A computerized descriptive method to measure the skull and its components was used,^21,22 and a sufficiently large sample, grouped according to skeletal age and gender, was gathered for measurements. The purpose of this investigation was to determine whether an Angle Class II malocclusion is characterized by either an underdeveloped or a posteriorly positioned mandible,

The radiographs for this study were collected from the W. M. Krogman Research Center for Child Growth and Development in Philadelphia. All radiographs were assessed for skeletal age by trained staff at the Growth Center by using the hand-wrist plates and methodology of Greulich and Pyle,²³ who based their atlas on the work of Tanner and Whitehouse.²⁴

The control group comprised children with "normal" occlusion. The ethnic distribution of these children was Northern European (German, Scandinavian). Southern European (Italian), Scottish, Irish. English, Middle European, and Eastern European. As a group. they were "reasonably representative of the white children in the city of Philadelphia"²⁵ during the 1950s and early 1960s.

The Class II, Division 1 sample represented children referred to the center by the Orthodontic Department School of Dental Medicine of the University of Pennsylvania. The inclusion criteria for this sample were as follows: (1) Class II. Division 1 malocclusion, as assessed from dental casts; (2) no prior orthodontic treatment; (3) no history of severe medical illness; (4) all first permanent molars present; (5) radiographs of high quality; and (6) a skeletal age appropriate to the limits of the study, regardless of chronological age. Their ethnic distribution was considered comparable to that of the normal control group. Table I lists the sample characteristics of the control group and the Class II, Division 1 comparison group. Each group (control and Class II, Division 1) had 6 subgroups: 3 male and 3 female.

The method for converting the lateral cephalograms into computer-readable x- and y-coordinate data was first described, in 1967, by Walker.²¹ When applied to a defined ethnic population, this method produces a schematic composite representing a single view of the average configuration for the group. The x- and y-coordinate data used to describe each lateral cephalogram and the computer-drawn composite figure representing the group constituted the "mathematical" model of the skull. Fig I shows the composite norm for the 10-year-old female sample and the 14-year-old male sample. Growth achieved in the x and y directions at nasion, point A, point B, and menton, as well as the standard deviation from the mean, are indicated. Krogman stated

that this composite type norm or type standard.. will be typical or representative of a face for each sex and for all ages, from birth to maturity. In a sense, each composite tracing will be "average" but geometrically or configurationally so rather than arithmetically. Such a type norm need not (probably cannot) be an absolute category but will be, rather, a relative framework. Faces need not be molded to it: they need only approximate it thus granting a full measure of individuality.²⁶

The conversion of a lateral cephalogram into a mathematical model, and then into a composite drawing, consisted of tracing and digitizing 177 derived anatomical landmarks. The tracing was oriented on a Benson-Lehner optical scanner and recorder (OSCAR) so that sella represented the origin of a Cartesian coordinate system. The x-axis was defined as a line parallel to the Krogman-Walker plane (occipitale-maxillon) through sella. The y-axis was perpendicular to the x-axis through sella. The Krogman-Walker horizontal plane, as seen in Fig 2, is defined by two highly reproducible endpoints, maxillion and occipitale, which are located in the mid-sagittal plane. Each landmark was given an x- and a y-coordinate and sequentially located in a predefined sequence (Fig 3), which was stored in a database. Each cephalogram from both the control and the Class II, Division 1 sample was digitized, registering all 177 landmarks. A computer software program was able to calculate the mean x- and y-coordinate values for each of the 177 landmarks for each subgroup (3 female and 3 male subjects) within the control group and Class II, Division 1 group. Therefore, 12 composite drawings and 12 sets of landmark means (control, 3 females and 3 males; Class Il, Division 1, 3 females and 3 males) were constructed so that statistical differences between the control and Class II, Division I groups matched for gender and age (6 tests) could be demonstrated if present.

Fig 4 illustrates the skeletal and dental measurements that were taken. Fig 5 shows those measurements that were significantly different in at least 5 of the 6 inter-subgroup comparisons between the control and Class II, Division 1 malocclusion groups.

In the construction of the mathematical model of cephalograms, several types of errors may be introduced. To test the error of scanning, one of us (T.R.) and 2 other operators scanned the same tracing 10 times. The average of standard deviations for all 3 operators for locating the coordinate distance from sella to each of the 24 common endpoints was ± 0.15 mm along the x-axis and ± 0.28 along the y-axis. To test the error of tracing, one of us (T.R.) and 2 other operators each traced 10 different radiographs. The average of standard deviations for 24 endpoints for all 3 operators was ± 0.48 mm along the x-axis and ± 0.47 mm along the y-axis. To test the intra-rater reliability for the author, we retraced, redigitized, and rescanned 10 radiographs from each of the 6 subgroups within the Class II, Division I group.

The results indicated that the error was within tolerable limits; however, certain landmarks, such as ANS, PNS, and basion, were subject to higher levels of measurement error. To test the inter-rater reliability test between the observers, we retraced, redigitzed, and rescanned 10 cephalograms from each subgroup of the control group. The results indicated that there was a predominance of agreement between measurements of different observers; however, they were less precise and more variable than the author’s intra-rater reliability test.

For each of the variables listed, descriptive statistics, including mean and standard deviation, were determined for each subgroup within the control and Class II, Division I groups. (Additional statistics were also determined and presented in the main author’s 1971 dissertation.) Student t-tests were conducted, comparing the control sample to the Class II Division 1 sample, matching the subgroups for gender and age. Thus, 6 t-tests were carried out. Table II lists all the means and standard deviations for each of the subgroups within the control sample. Table III lists all the means and standard deviations for each of the subgroups within the Class II, Division 1 sample, including levels of significant differences between subgroups. Each table lists all the variables, including their definitions as well as the diagram from which each variable can be visualized. Fig 6 represents a diagrammatic view of the 6 intergroup differences between the control samples and the Class II, Division 1 samples.

Click below to see enlarged
Fig 4. Measurements taken to describe size, form, and position of skeletal and dental anatomy in children with Class II, Division 1 malocclusion as compared with normal control group.

The following results reflect those measurements that were significantly different in either all 6 inter-subgroup comparisons or in at least 5 of the 6 comparisons. In a few of the cases the results were presented if they were significant for either all 3 female or male subgroups. Specific measurements are identified by the measurement number and diagram letter as seen in Fig 4.

· Cranium: All 3 female and male samples showed significantly larger frontal bone thickness at the level of the frontal sinus (A2).

· Cranial Base: The anterior cranial base and posterior cranial base (A4 and A5) were significantly larger in all 3 female and male samples.

· Mandible: Symphyseal height (B11 was significantly larger for all 3 male subgroups. Point B relative to the cranial base (SNB, C17) was significantly smaller for all 6 inter-subgroup comparisons; however, mandibular position relative to sella was not significant relative to the controls (D12 and D13). The decreased value in SNB may have been due to a more protrusive nasion (A5), while the position of point B was similar to that of the controls (D13).

· Maxilla: All height and depth measurements of the maxilla were significantly larger (G19-23), except for the palato-alveolar height (G23) in females. The anterior moiety of the maxilla (630) was significantly larger in all 6 inter-subgroup comparisons. The angle SNA (C26) in the 10-year-old male Class II, Division 1 subgroup was sig-nificantly different, with a more retrusive maxilla than

what was seen in the controls. However, Fig 5, measurement C25, shows that the maxilla is actually more protrusive when measured from the Cartesian coordinate system. The anteroposterior position for the Class II, Division I group is significantly more protrusive than that of the controls. Making use of the angle N-A-Pog (E27) to evaluate anteroposterior position of the maxilla also proved to he misleading. This indicator would have led to the conclusion that only the 12-year-old female Class II, Division 1 subgroup possessed a significantly more protrusive maxilla relative to the controls, while the 10-year-old male Class II, Division 1 subgroup possessed a significantly more retrusive maxilla. However, the retrusion in the male subgroup is only relative with respect to an increased protrusion of nasion (A5) and pogonion (D13). In the female subgroup, point A (C25) was protrusive; nasion was also protrusive but less so (A5); and pogonion was the same as in the controls (D12). An interesting finding was a significantly larger palato-alveolar height in Class II, Division 1 male subgroups relative to the controls; despite the fact that the distal cusp tip of the molar (F42), which was more occlusally positioned, was not significantly different from the controls except for the 10-year-olds.

· Maxilla/Key Ridge: The anteroposterior position of the key ridge as measured from the Cartesian coordinate system (C28) was not significantly different relative to the controls.

· Maxilla/Palate: The slope of the palate relative to the cranial base (E31) was significantly smaller in all 6 inter-subgroup comparisons. The slope of the palate relative to the mandibular plane (E32) was larger in all 6 inter-subgroup comparisons (except the 12- and 14-year-olds), which may indicate that the palate is tipped up anteriorly.

Maxilla-Mandible: The ANB angle (C33) and distance from point A to point B as measured along the x-axis of the Cartesian coordinate system were significantly different relative to the controls except the 10-year-old male subgroup.

· Mandibular Dentition: No significant differences were detected relative to the Cartesian coordinate system of angular measurements (F35, B36-38, F39). However, the symphyseal height (B11) including the lower incisor was larger in all male samples. Furthermore the molar-incisor distance (B38) tended to increase in males, reaching a maximum in the 14-year-olds. The lower incisor inclination (F39) was not significantly larger any subgroup, but it was more labially inclined in the 14-year-old males.

· Maxillary Dentition: Significant differences were detected anteroposteriorly in all 6 inter-subgroup comparisons. The incisors were more anterior and more labially inclined (F40-41, F43-45), while the molars were mesially positioned (Fig 6). No vertical differences were detected.

Facial Depth and Height: The upper and middle facial depth (G46-47) was significantly larger, but the lower facial depth showed no significant differences. No significant differences were noted in upper/lower anterior or total anterior/posterior facial heights (G49-52), except for the 12-year-old male inter-subgroup comparison. The significantly increased palato-alveolar height (G23) and inferiorly positioned upper molar, although not significantly different, may explain why gonion was more inferiorly positioned (G51) with an increased lower facial height (G52).

 

Fig. 5 Measurements that were significantly different in Class II, Division 1 samples (see Table I).

Rothstein and Yoon-Tarlie 331
Volume 117, Number 3

Class 11. Division 1 malocclusions can occur as a result of dental and/or skeletal configuration such as small mandible, a retrognathic mandible, or a steep mandibular plane angle. Moyers et a1²⁷ go even further by subdividing this malocclusion into 11 categories. That there is such a variety of opinions among researchers may be due to several factors, such as the use of mixed samples of age and gender and the use of samples based on chronological age as opposed to skeletal age. Other factors that may have contributed to those mixed findings include the unreliability of the source of the radiographic plates, the small sample size, and the number of inter-group comparisons. The use of certain commonly employed horizontal planes or orientation located far too superior to the maxilla and mandible also may contribute to the mixed findings in the literature. Commonly used planes such as Sella-Nasion (anterior cranial base) and the Frankfort horizontal have been shown to introduce error. First, they are superiorly located, and consequently they exaggerate the variability of points in the lower face. Second, the endpoints are less reliably identified and not mid-sagitally located. Third, the Frankfort horizontal plane has been shown to deviate sufficiently when the head is held in its natural head position, so that if this plane were used in a coordinate system strictly limited to the measurement of projected linear dimension it might result in foreshortening of the structures of the lower face (ie, a smaller and retrusive mandible). (The Krogman Walker plane was used in this study because this plane passes through the midface, where there would be a reduction of variability of endpoints in the middle and lower face. Moorrees observed that the degree of variation increases with the distance of the landmarks from the basic orientation axis in both the vertical and horizontal distance.^28,29 The Krogman Walker plane also closely approximates the natural position of the head, making it more reliable in the assessment of anteroposterior deviations.²⁸) Last, the use of certain angular measurements¹⁰ and the lack of a visual, pictorial method of presenting the results have contributed to the variety of inconclusive findings. Nonetheless, derived from these flawed findings are the numerical reference standards clinicians use daily in order to assess how patients vary from the norm and as a guide to planning treatments.

Rubin³⁰ noted that some cephalometric measurements have their value in the making of the initial diagnosis and others are useful in monitoring changes. The SNA, SNB, and ANB angles are examples of angles not useful in studying assessments. Chang³¹ states that "variation in the spatial position of Nasion and/or vertically and the point A or B vertically is a normal anatomic occurrence." The anteroposterior position of the jaws is not accurately reflected in the ANB angle. For example, if SNA, SNB, and ANB were used as the only measures of mandibular and maxillary position without pictorial representation, we would have been forced to conclude that the central tendency of children with Class II, Division I malocclusions was a retrognathic mandible (Table III, measurement C17) with a normal maxilla (C26). Angle N-A-Pog (E27) indicates that 12-year-old Class II, Division 1 females have a significant midfacial convexity and that 10-year-old males with Class II, Division 1 malocclusions have a retrusive maxilla (C26, E26). However, on further examination of the findings within 10-year-old males, the frontal bone at the level of the sinus (A2) is significantly increased in thickness, contributing to an increased length in the anterior cranial base (A5). This finding was present for all 6 inter-subgroup comparisons between normal and Class II, Division 1 malocclusions. This helps to explain why the SNB angle can be quite misleading.

We propose that "excessively developed" sinuses may contribute to a more protrusively positioned maxilla. That maxillary protrusion was not a finding in the 10-year-old Class II, Division 1 males may be due to remaining growth unexpressed at the age of 10. Oktay³² has noted a few studies relating sinus size and malocclusions. From his investigation, he concluded that "female subjects with Class II, Division 1 have larger sinuses...no explanation exists... as to why this is so for the female subjects." Could the size of the maxillary sinus contribute or be associated with the development of Class II or Class III malocclusions? Moss³³ considers the form and spatial position of the bony elements to be a direct response to the primary growth of the functioning soft tissues and spaces that they protect and support. The maxilla, for instance, is viewed as a conglomerate of relatively independent bony components, the form of which is related to the teeth, orbital contents, respiratory function, and muscle attachments.

For decades the composite method as described by Krogman²⁶ represented the most effective standard in craniometry; it was the forerunner of roentgenographic cephalometry. This method was used in conjunction with a coordinate system, providing a descriptive and analytic tool that precluded most of the major problems arising from other methods. In 1994, Kerr et a1¹⁸ also employed a computerized composite technique to pictorially demonstrate that there was no difference in size, form, and position of the mandible in the samples they compared. Other researchers have also recommended that we divest ourselves of numerical standards or that we at least support our findings more graphically. We have used composite forms in which visualization of size, form, and position of the mandible, as well as vertical dysplasias of the maxilla, are well addressed.

Perhaps it is time, as Rubin¹⁹ suggests, for the world’s anthropologists and orthodontists to meet, as they did in 1929, in order to adopt a method of improved communication within the orthodontic and anthropologic literature regarding the assessment of craniofacial morphology as well as the changes due to growth and treatment.

332 Rothsten and Yoon-Tarlie
American Journal of Orthodontics and Dentofacial Orthopedics March 2000

CONCLUSIONS

The data of this study support the following conclusions:

1. The results of this study do not support Angle’s hypothesis that Class II, Division 1 malocclusions are character-

ized by an underdeveloped and or posteriorly positioned mandible. The mandible and dentition of Class II, Division I malocclusions were found to be identical to those of the control subjects in size, form, and position for both genders at the ages of 10, 12, and 14 years, except for symphyseal height, which was larger in all 3 males.

2. The maxillary first permanent molar in Class II, Division 1 malocclusions is more mesially positioned.

3. The anterior segment of the maxilla is more protrusive and superiorly positioned in Class II, Division 1 malocclusions.

4. Vertical dysplasias are not typical findings in Class II, Division I malocclusions.

5. SNA, SNB. ANB, and NAB must be evaluated with caution unless nasion is situated within its "normal" locus for age and gender, both superiorly-inferiorly and anteriorly-posteriorly.

6. Excessive anterior cranial base length, characterized by enlarged frontal and maxillary sinuses, may be a contributing factor in the development of Class II, Division 1 malocclusions.

7. This study does not support the hypothesis that an increase in cranial base flexure (Ba-S-N, 3A) contributes to a retruded mandibular position,³⁴ for this angle was found to be larger in all 6 inter-subgroup comparisons, significantly in 3 of the 6 comparisons, while the mandible is not posteriorly positioned in any of the samples.

8. The use of composite norms not only provided a pictorial representation of the 6 inter-subgroup comparisons but also was essential to the understanding and interpretation of the angular and linear measurements that revealed differences between the control sample and Class II, Division 1 sample.

We thank Esther Lafair for her assistance with this article.

1. Tulloch JFC. Phillips C, Koch G, Proffit W. The effect of early intervention of skeletal

pattern in Class LI malocclusion: a randomized clinical trial. Am J Orthod Dentofac Orthop 1997;l 11:391-400.

2. Woodside DG. The present role of the general practitioner in orthodontics (Figure 1)

based on Burlington Orthodontic Research Center report Number 3, 1958. Dent Clin North Am July 1968; p. 483-508.

3. Massler M, Frankel JM. Prevalence of malocclusion in children aged 14-18 years. Am J Orthod 195l;37:751-68.

4. Goffman E. Human Relations 1957;X:52.

5. Frohlich FJ. Changes in untreated Class II type malocclusion. Angle Orthod

1962;32: 167-79.

6. Baccetti T, Franchi L. McNamara JA Jr. Tollaro I. Early dentofacial features of Class II malocclusion: A longitudinal study from the deciduous through the mixed dentition. Am J Orthod Dentofac Orthop 1997;l 11:502-9.

7. AngIe EH. Classification of malocclusions. Dental Cosmos 1899;41:248-64, 350-57.

8. Angle EH. The upper first molar as a basis of diagnosis in orthodontia. Dental items of Interest 1 907;28:42 1-39.

9. Rothstein T. Facial morphology and growth from 10-14 years of age in children presenting Class 11 division 1 malocclusion: a comparative roentgenographic cephalomet-ric study [dissertation]. Philadelphia: University of Pennsylvania; 1971.

10. Rosenblum R. Class II malocclusion: mandibular retrusion or maxillary protrusion. Angle Orthod l995;65:49-62.

11. McNamara JA Jr. Components of Class II malocclusion in children 8-10 years of age. Angle Orthod 1981;51:177-202.

12. Pancherez H. Cephalometric characteristics of Class II division 1 and Class II division malocclusions: a comparative study in children. Angle Orthod 1997;67:1 11-20.

13. Wylie WL. The assessment of anteropostenor dysplasia. Angle Orthod 1947;17:87-109.

14. Blair ES. A cephalometric roentgenographic appraisal of the skeletal morphology of Class I Class II division I and Class 11 division II (Angle) malocclusions. Angle Orthod 1954;24: 106-19.

15. Altemus LA. Horizontal and vertical dentofacial relationships in normal and Class II division I malocclusion in girls 11-15 years. Angle Orthod 1955;25:120-37.

16. Cohen SG. Craniofacial analysis of skeletal patterns exhibiting Class II division I(Angle) [unpublished thesis]. Philadelphia: Temple University; 1960.

17. Amoric M. Etude critique des evaluations cephalometrique composant les classs II d’Angle selon McNamara. Rev Orthop Fac 1985;19:563-8.

18. Kerr WJS, Miller S, Ayme B, Wilhelm N. Mandibular form and position in 10 year old boys. Am J Orthod Dentofac Orthop 1994;106:115-30.

19. Drelich RC. A cephalometric study of untreated Class II Division I malocclusion. Angle Orthod 1948;17:70-5.

20. James GA. A clinical analysis of 100 Class II, division I malocclusions with special reference to the cranial base. Dent Pract 1963;14:35-46.

21. Walker GF. Analysis of skull shapes by electronic methods [1965, 1966, and 1967 annual report submitted to National Institute of Health, Bethesda, Md, in support of Grant HD 00181-02 and 00181-05].

22. Walker GF. Summary of a research report on the analysis of craniofacial growth. N Z Dent J 1967;63:22-9.

23. Greulich WW, Pyle SI. Radiographic atlas of skeletal development of the hand and wrist, 2nd ed. Stanford (Calif): Stanford University Press; 1959.

24. Tanner JM, Whitehouse RH. Standards for skeletal maturity based on a study of 3000 British children. London: University of London, 1959.

25. Krogman WM. Physical growth of philadelphia white children, age 7-17 years. Philadelphia: Center for Research in Child Growth, 1965.

26. Krogman WM. Validation of the roentgenographic cephalometric technique. Am J Orthod 1953;44:933-9.

27. Moyers R, Riolo ML, Guire KE, Wainright RL, Bookstein FL. Differential diagnosis of Class II malocclusion. Am J Orthod 1980;78:477-94.

28. Moorrees CFA, Kean MR. Natural head position, a basic consideration in the interpretation of cephalometric radiographs. Am J Physic Anthrop 1958;16:213-34.

29. Moorreess CFA. Normal variation and its bearing on the use of cephalometric radiographs in orthodontic diagnosis. Am J Orthod 1953;3:942-50.

30. Robin R. Making sense of cephalometrics [editorial]. Angle Orthod 1997;67:83-5.

31. Chang H. Assessement of anteroposterior jaw relationship. Am J Orthod Dentofac Orthop 1987;92:1 17-22.

32. Oktay H. The study of the maxillary sinus areas in different malocclusions Am J Orthod Dentofac Orthop 1992; 102:143-6.

33. Moss ML. Vertical growth of the human face. Am J Orthod 1964;50:359-76.

34. Dibbets JM. Morphological associations between the Angle classes. Eur J Orthod

1996;18:111-8.