The retromolar space is a space at the rear of the mandible, between the back of the last molar and the anterior edge of the ascending ramus where it crosses the alveolar margin, and this space is also frequently seen in the Neanderthal mandible and is considered one of the characteristic features of Neanderthal faces [2]. However, this space can hardly see in modern man.

Howells [3] argues that Neanderthal facial features (prognathism) were caused by an anterior shift of the dental arch, resulting in a large retromolar space. Subsequently, Rak [11] attributed the large retromolar space to a degenerated dental arcade length and backward migration of the mandibular ramus, while Trinkaus [2] argued the backward migration of the masticatory muscle region relative to the dentition. Nara [4], based on a survey of material from Neanderthals of both young adult without heavy tooth attrition and advanced adult with progress tooth attrition, stated that severe adjacent surface wear of the molars led to the uniquely wide posterior molar space of the Neanderthals. In other words, he considers the wide retromolar space to be one of the secondary acquired factors associated with the anterior shift of the molar dentition due to interproximal attrition wear.

Nguyen [5] confirm large increases in retromolar space through growth. Kahl [6] have demonstrated that 97.4% of impacted teeth had insufficient space for eruption. On the other hand, Al-Gunaid [7] notes that a smaller retromolar space is associated with impaction.

From an anatomic perspective, it is difficult to compare the distal direction of the molar space from the third molar, but it is relatively possible to compare the space created posteriorly from the second molar, which is strongly related to the eruption of the third molar. In this study, the space from the distal margin of the mandibular second molar to the alveolar bone edge was considered as the retromolar space (M₂-retromolar space), and the relationship between the length of this trait and the eruption status of the third molar on radiographs was examined.

The materials examined were dental intraoral radiographs of human skulls excavated from archaeological sites in the Jomon, Yayoi, Kofun, Kamakura, Muromachi, and Edo eras [8], and the modern human materials were panoramic radiographs of patients visiting a dental clinic.

The intersection of the internal oblique line and the straight line connecting the proximal most projection of the first molar and the distal most projection of the second molar on the radiograph was defined as point Z. The distance from point Z to the distal surface of the second molar was defined as the size of the M₂-retromolar space (“D") (Figure 1).

The M₂-retromolar space and distance from the Z point to Gonion (DZGo) were measured as the size of the mandibular angular region. A caliper (CD-S15C 1/50 mm: Mitutoyo Co. Ltd) was used for the measurements. Measurements were taken for the left and right sides of both sexes, and statistics were calculated using the number of jaw sides. Statistics were processed by era for the archaic human skulls, and by decade for the modern humans from birth in the 1930s to the 1980s. Because radiographic measurements are more magnified than absolute measurements, the corresponding side measurements were standardized by the width (mesiodistal crown diameter) of the second molar in the right or left jaw, and then summed for the right and left sides to obtain the average value.

The molars were excluded from the study if they suffered from extremely adjacent interdental wear, if the internal oblique line was obscured on radiographs, if the most projections of the first and second molars were difficult to measure with prosthetic materials, or if the eruption status of the first or second molar was inappropriate for measurement due to torsion or other reasons.

Figure 1: Distance of M₂-retromolar space (D) from point Z.

Table 1 shows the M₂-retoromolar space and DZGo, standardized by the width of the mandibular second molar, from the 1930s to the 1980s. A paired t-test for left-right difference for M₂-retoromolar values in the 1970s, which represents a relatively large sample, revealed no significant differences (t=1.3725: df =53). A t-test for sex difference also showed no significant difference (t=1.4172: df =233). Similarly, no significant difference was found for DZGo between left and right (t=0.5068: df =53), but males were significantly larger than females (t=4.6196: P<0.01) for the DZGo distance.

In the 1930s and 1940s, the M₂-retoromolar, standardized by the width of the second molar, occupied more than 90% of the width of the second molar for both sexes, but it has gradually narrowed over time, decreasing to less than 60% in the 1980s.

Table 1: Basic statistics of DZGo and M₂-retromolar standardized by width of second molar.

DZGo: Distance from Z point to Gonion.

Mean (sd, n), n= number of jaw sides.

Table 2 shows the M₂-retromolar space of archaic humans and modern humans. The values for modern humans include both sexes. The distance of the M₂-retromolar during the Jomon era was 1.23, the longest in Japanese history as far as the research could be determine. Subsequently, this value was decreasing as time went by, and then once increased in the Muromachi era, but in the Edo era, the value was almost the same as that in the Kamakura era. The modern humans (1930s-1980s) show a rapid decreasing trend from the 1940s to today.

Table 2: Historical changes of M₂-retromolar space standardized by mesiodistal diameter of second molar in combined sexes.

Mean (sd, n), n= number of jaw sides.

Figure 2 shows two line graphs of the M₂-retromolar space of archaic human skulls from the Jomon era to the Edo era and that of modern Japanese from the 1930s to the 1980s. The solid line is interrupted in the middle because of dental radiographs of the oral cavity in archaic humans and panoramic radiographs in modern humans.

Figure 2: Line chart width of M₂-retromolar space standardized by mandibular second molar.

Table 3 shows the basic statistics of the standardized DZGo for males and females. The male DZGo was around 2.9 times the second molar width from the 1930s to 1940s, decreasing to 2.6 times by the 1950s and subsequently fluctuating in the 2.6 times range. Females expressed this value lower throughout the generations than males. In the 1930s and 1940s, the value was around 2.7 times, but after the 1950s it fluctuated in the 2.5 times range. The difference in size of the mandibular angle region between males and females since the 1960s was almost constant.

Table 3: Basic statistics of DZGo standarized by mesiodistal diameter of second molar.

Mean (sd, n), n= number of jaw sides.

A Neanderthal skull excavated from the Amud Cave in Israel about 40,000 years ago [1], Plate 33 shows that all three mandibular molars erupted completely upright, and there is a wide retromolar space behind the third molar. The high frequency of mandibular retromolar spaces among the Neanderthals is often cited as a derived character relative to Early and Middle Pleistocene members of the genus Homo. When this skull is viewed laterally, the M₂-retromolar space is estimated as a distance of more than 1.5 times the second molar crown widths from the distal surface of the second molar to the anterior margin of the mandibular ramus.

Four explanations have been put forward relating variation in Neanderthal dentofacial variables to the high frequency of retromolar spaces: (1) an anterior migration of the dental arcade [9,3,10]. (2) a posterior "retreat" of the zygomatic and anterior ramal regions relative to a fixed molar position [2]; (3) a shortening of the dental arcade due to mesiodistal molar diminution [11]; and (4) a shortening of the dental arcade due to a combination of anterior migration of the postcanine dentition and posterior migration of the anterior dentition [12,13]. States that all of these have been attributed to the creation of the retromolar space. According to Wolpoff [14], extensive interdental wear is seen in all hominids, living and extinct.

Despite the fact that the retromolar space in the mandible has played a major role in human evolution, it is not easy to measure the size of the retromolar space. Nara [4] used the distance of from the distal direction of the third molar to the anterior margin of the mandibular ramus as a measurement of the length of the retromolar space. However, the alveolar region, where the teeth emerge in the mandible, and the mandibular ramus are not appropriate to measure a retromolar space due to the difference in mandibular bone configuration [15]. It is almost impossible to measure them in vivo, moreover. In the present study, a Z point was devised as a reference for the retromolar space area on the radiographs. This point is not affected by occlusal and adjacent wear and is easy to draw and relatively stable.

The study showed that the length of the M₂-retromolar space in the prewar 1930s and 1940s was about the same as the width of second molar, but after the 1950s this amount was gradually decreasing, and by the 1980s it had shrunk to about 50% of the width of second molar in both sexes. The eruption status of the third molars documented generally the fact that the majority of the third molars before the 1960s were in the upright position, but the majority since the 1970s, were in eruption difficulty (inclined and horizontal positions) [16]. In other words the status of the M₂-retromolar space likely coincides with the eruption of the third molar.

The study of archaic humans also showed a decreasing trend in M₂-retromolar space over time from the Jomon era to the Edo era. When the M₂-retromolar space was wide, especially during the Jomon and Kofun eras, it was common for the third molar to erupt in an upright position [16]. A strong relationship also exists between M₂-retromolar space and the eruption status of the third molars. The Jomon era showed the maximum M₂-retromolar space in Japanese history, but this value decreased over time as the diet became softer and softer, and especially after World War II, the M₂-retromolar space was dramatically decreasing.

On a time scale, the M₂-retromolar space has been decreasing 40% of the width of second molar for about 2,400 years, from the beginning of the Yayoi era, to 1868, the end of the Edo era. A 40% reduction was also observed in the 60-year period from the 1930s to the 1980s. The graph shows that the reduction is almost the same in the ancient skulls, but the period spent in reduction is 1/40 shorter in the post-World War II. This shows how dramatically the latter period was decreasing. The fact that the M₂-retromolar space is still decreasing even in today when adjacent surface wear is almost non-existent suggests that adjacent surface wear itself is not responsible for the formation of the M₂-retromolar.

In the study of the facial skeleton and food in England, Moore [17] found that over the centuries the mandible has become smaller and the mandibular angle more obtuse in relation to dietary changes in modern society that have resulted in softening of food. Inoue [8] investigated the size of the mandible in modern Japanese from X-ray cephalometric and Kaifu [18] from anthropological surveys, suggesting that the shortening of the mandibular ramus and the expansion of the mandibular angle over time may indicate that the reduced development of the masticatory muscles (masseter muscles) attached to the mandible in modern Japanese may be due to the softening of food.

Experimental animal studies have also reported that the mandibular angle and mandibular ramus are affected by a soft or hard diet [19,21]. Corruccini [22] found significant differences in occlusion between squirrel monkeys (Saimiri sciureus) raised on natural and artificial soft diets, and Kikuta [23] and Inoue [8] reported similar results, including a reduction in mandibular ramus height and a more open mandibular angle on a kneaded diet.

In the postwar period, Japanese people's food intake has shifted to soft and easily digestible ramen and udon noodles, sweet and easily chewable cakes and puddings, bananas and other fruits, rather than hard and tough foods such as dried goods like surume (dried squid) and scallops, nuts, and dried fruits that they had often eaten before.

As a result, the masticatory muscles, especially the masseter muscles, were weakened by reducing their function, resulting in obtundation of the mandibular angle.

The trends of both males and females in DZGo increased once after the 1930s, but subsequently decreasing drastically, and the current level showed almost the same as in the 1950s or even higher. Although males were well closed to female in the values of DZGo in the 1950s, subsequently the strong sex difference indicates that the DZGo is becoming smaller in females.

Enlow [24,25], in his study of the mandible, noted that the mandibular ramus increases in bone addition during growth, especially at the lower posterior margin of the mandible, with a corresponding resorption of bone at the anterior margin of the mandible, especially at the lower part, and a gradual backward growth of the mandible. Decreased masticatory muscle activity has been shown to lead to reduced bone resorption during mandibular ramus growth, resulting in less posterior growth and early termination of posterior growth of the M₂-retromolar. As a result, the posterior molars remain narrow even after mandibular growth, and the most likely conclusion is that the mandibular third molars are forced to erupt in the inclined or horizontal position at the time of eruption. The degree of wear between adjacent teeth is not so different between the 1930s and 1980s. In the postwar period, there were no Japanese individuals with extreme adjacent interdental wear. Nevertheless, the M₂-retromolar space is still decreasing with time.

The study showed that the size of the M₂-retromolar space in the prewar 1930s and 1940s was about the same as the second molar width, but after the 1950s this amount was gradually decreasing, and by the 1980s it had shrunk to about 50% of the second molar width in both sexes. Archaic humans also showed a decreasing trend in M₂-retromolar space over time from the Jomon era to the Edo era. The softening of food weakens the masticatory muscles and causes degeneration and enlargement of the mandibular angle. As a result, growth in the distal direction from the Z point stagnates, and the Z point at the time of third molar eruption no longer moves in a more distal direction, and the third molars are likely to erupt in the inclined or horizontal position. Looking at the time axis, the M₂-retromolar space for about 2,400 years from the end of the Jomon era to the end of the Edo era and for 60 years from the 1930s to the 1980s each shows a 40% decrease in the width of the second molar, and although the decreasing trend looks almost the same on the graph, the time required for the reduction is 1/40 of the difference. The latter was dramatically more decreasing than the former. The amount of M₂-retromolar space is not related to the anterior shift of the dental arch due to interdental wear of the postcanine molars.

The materials of human bones excavated from archaeological sites in various eras (housed in National Museum of Nature and Science) were provided by Mr. Naohiko Inoue and his research collaborators who provided X-rays (intraoral radiographs). The materials on modern humans was provided by the dental clinics. The author would like to express gratitude to all of them.

1. Suzuki H, Takai F. The Amud man and his cave site. Suzuki H, Takai F. Therapeia, Tokyo. 1999.
2. Trinkaus E. The Neandertals face: Evolutionary and functional perspectives on arecent hominid face. J Human Evolution. 1987; 16: 429-443.
3. Howells WW. Neanderthal man: Facts and figures. In “Paleoanthropology: Morphology and Paleoecology”, RH Tuttle (Edition). Mouton, Paris. 1975; 389-407.
4. Nara T, Hanihara T, Dodo Y, Vandermeersch B. Influence of the interproximal attrition of teeth on the formation of Neanderthal retromolar space. Anthropological Science. 1998; 106: 297-309.
5. Nguyen A, Caplin J, Avenetti D, Durfee S, Kusnoto B, Sciote JJ, et al. A longitudinal assessment of sex differences in the growth of the mandibular retromolar space. 2022; 143: 105547.
6. Kahl B, Gerlach KL, Hilgers RD. A long-term, follow-up, radiographic evaluation of asymptomatic impacted third molars in orthodontically treated patients. Int J oral and Maxillofacial Surgery. 1994; 23: 279-285.
7. Al-Gunaid TH, Bukhari AK, El-Khateeb SM, Yamaki M. Relationship of mandibular ramus dimensions to lower third molar impaction. European J Dentistry. 2019; 13: 213-221.
8. Inoue N, Ito G, Kamegai T. Microevolution of Occlusion and Dental Disease: A Study of Tooth to?Denture Base Discrepancy. Tokyo. Ishiyaku-shuppan. 1986.
9. Coon CS. The Origin of Races. Alfred A. Knopf, Publisher, New York. 1962.
10. Wolpoff MH. Neanderthals and their relatives. In: Knopf AA, Paleoanthropology. New York. 1980; 252: 1996-1997. Edition. McGraw Hill: New York.
11. Rak Y. The Neanderthal: A new look at an old face. J Human Evolution. 1986; 15: 151-164.
12. Spencer MA, Demes B. Biomechanical analysis of masticatory system configuration in Neandertals and Inuits. American J Physical Anthropology. 1993; 91: 1-20.
13. De la Cova C. The retromolar space: a morphological curiosity observed amongst the protohistoric Arikara and Mandan. Int J Osteoarchaeology. 2015; 26: 610-620.
14. Wolpoff MH. Interstitial wear. American J Physical Anthropology. 1970; 34: 205-228.
15. Kamijyo Y. Primary Oral Anatomy. Anatome Inc. Tokyo. 1969.
16. Yamada H. Tooth size in Japanese islands. In: Japanese Dentition. World Scientific Publishing Co. Pvt. Ltd., Singapore. 2020; 85-112.
17. Moore WJ, Lavelle CLB, Spence TF. Changes in the size and shape of the human mandible in Britain. Br Dent J. 1968; 125: 163-169.
18. Kaifu Y. Changes in mandibular morphology from the Jomon to modern periods in eastern Japan. American J Physical Anthropology. 1997; 104: 227-243.
19. Watt DG, Williams CHM. The effects of the physical consistency of food on the growth and development of the mandible and the maxilla of the rat. Am J Orthod. 1951; 37: 895-928.
20. Ito G, Kuroe K, Yasuda H, Inoue N, Kamegai T. Experimental study on the degeneration of the jawbone, J Orthodontic J. 1982; 41: 708-715.
21. Ito G, Mitani S, Kim H. Effect of soft diets on craniofacial growth in mice. Anat Anz Jena. 1988; 165: 151-166.
22. Corruccini RS, Beecher RM. Occlusal variation related to soft diet in a nonhuman primate. Science. 1982; 218: 74-76.
23. Kikuta T. Effects of dietary consistency on growth and development of neurocranium and facialcranium in the rat. Tsurumi Shigaku. 1985; 11: 141-170.
24. Enlow DH. Handbook of Facial Growth, Philadelphia, W.B. Saunders Company, London. 1975; 234-
25. Franciscus RG, Trinkaus E. Determinants of retromolar space presence in Pleistocene Homo mandibles. J Human Evolution. 1995; 28: 577-595.