The purpose of this study is to provide basic palaeopathological procedures in diagnosing and interpreting the osseous syphilis found in excavated Edo cranial series, and finally to elucidate a prevalence of syphilis among the people of the Edo period.

The gross pathology or macroscopic morphology of the lesions in the cranium are emphasized in the diagnosis of osseous syphilis in this study. It is natural that this diagnostic approach may be limited in its strictness and thoroughness compared with modern clinical technological approaches. Further, in terms of the weighted classifi-cation of syphilitic crania used in this study, it should be noted that it is not possible to make an exact distinction between demonstrable cases and possible cases or between possible cases and questionable cases because of the continuity and variability of the morphology of syphilitic lesions, as shown in the previous chapter.

Nevertheless, the author believes with certainty that, in comparison with other methods such as microscopic examination of sectioned bone, the approach with mac-roscopic observation as practiced in the present study is one of the most reliable and valuable approaches to diagnosing the pathological changes that appear in ancient dried bone and that this approach will be used widely in the future in this field.

In this context, based on the close investigation of gross bone pathology of osseous syphilis, the discussion will stress epidemiological analysis in oreder to provide new knowledge of the prevalence of syphilis among the people of the Edo period.

1. Incidence of Cranial Syphilis among the Series or Excavated Sites

As previously mentioned, there were significant differences in incidence in two com-binations of the three cranial series, i.e., Unko-in and Muen-zaka, Joshin-ji and Muen-zaka.

There may be two possible reasons why such significant differences occurred among the incidences of the cranial series. First, one might assume that such differences occurred merely by chance and may not have existed originally in the materials themselves. Namely, syphilis should have been distributed equally among the population of Edo without regard to place of residence, location of family temple, or social class or position; and the results obtained were not able to support this assumption due to the haphazard nature of the material used.

Second, if one wished to emphasize the strictness of social class or position among the people of the Edo period, one might consider that such differences were not the result of mere chance. As is known, differences in the location of a person's residence and one's family temple during the Edo period reflected differences in social class. Thus, the differences in the incidence of cranial syphilis between the series may be considered the inevitable result of differences in the social class or position of the persons whose human skeletal remains were excavated.

There were four strict social classes during the Edo period, which were referred to as the so-called shi-no-ko-sho, i.e., shi meaning the military class, no the agricultural or farmer class, ko the artisan class, and sho the merchant class. The military class, shi or bushi, was the governing class, whereas the other three classes, no, ko, and sho, were denned as common people under the political control of the bushi class.

The people belonging to the military class, bushi or samurai, were generally well-educated and strongly self-disciplined people mainly under the influence of Buddhism and the Japanese code of noble ethics known as bushi-do. It thus seems natural that such people as those of the bushi class might control their sexual passions and might remain monogamous.

On the other hand, the majority of the less-educated or more humble people might not be so continent of their sexual desires.

As a result, one could say that the differences in social class might account for the significant differences in the incidence rates of syphilis between the cranial series.

The incidence of cranial syphilis in the Muen-zaka series (3.5%) was the lowest of all the cranial series, and it also showed a statistically significant difference from that of the Unko-in series (11.5%) and the Joshin-ji series (9.9%).

One of the reasons for the low incidence of cranial syphilis in the Muen-zaka series is suspected to be that the majority of those interred in Muen-zaka might have belonged to the bushi class. As pointed out by KAWAGOE (1975), it is probable that the skeletons excavated from Muen-zaka were those of parishioners of Muen-ji who were of the bushi or hatamoto-samwai class in the service of the Edo government. Excellent evi-dence to support this supposition was also reported by KAWAGOE. The skeletal remains of a young female from Muen-zaka were found wrapped in a traditional white Japanese double silk robe (a shiro-habutae-kiimono), clearly showing that she must have been a young woman of noble birth or princess of a bushi family.

On the contrary, the materials from the Unko-in and Joshin-ji series, which show high incidences of cranial syphilis and which were both excavated from Fukagawa area in Tokyo, are considered to be the skeletons of common people due to their burial conditions and byburials. The Fukagawa area was the large downtown area where the unlicensed red light district was located at that time. The majority of the common people or less-educated people who lived in the Fukagawa area might be buried in the temples located in that area, such as Unko-in and Joshin-ji.

Following the Meiji Restoration, medical science, including syphilology and epide-miology, developed gradually with the rapid introduction of various Western sciences into the country. In the late part of the Meiji period, some clinical and epidemiological statistics of syphilis were reported. AOSHIMA (1907), for instance, investigated 1843 outpatients who suffered from venereal diseases in his clinic. Including syphilis (666 out of 1843 (36%)), the patients were analyzed from the epidemiological point of view on the basis of the following items: occupation, route of infection, and age distribution. The results of AOSHIMA'S analysis are shown in Table 5.1.

As can be seen from Table 5.1, which shows the occupations of the patients, such common people as farmers, day laborers and jobless men were relatively highly infected with venereal disease. The results of the occupational analysis reported by Ao-SHIMA seem to support the assumption that the incidence of syphilis may differ with the patient's occupation and social position.

2. Sexual Differences in the Incidence of Cranial Syphilis

As shown in Tables 4.5 and 4.6, using the demonstrable and possible cases, the incidence rate of cranial syphilis was 8.8 percent for males and 5.9 percent for females. There is no statistically significant difference between the incidence of the disease with regard to sex.

Unfortunately, we have no old medical documents written during the Edo period which refer to the precise incidence rate of syphilis in males and females separately. During the late Meiji period, however, "Krankenstatistick der dermatologischen Klinik an der kaiserlich-japanischen Universitaet zu Tokio: 1903" (NISHIKAWA and SAITO, 1903), an excellent report dealing with the clinical incidences of the various dermato-urological diseases including venereal syphilis, was published.

According to this clinical report, the total number of outpatients numbered 2547, and the number of patients diagnosed as suffering from syphilis numbered 231 (190 males and 41 females), that is, the frequency of syphilis was 9.07 percent among all outpatients. The number of patients with tertiary syphilis exhibiting bone and joint lesions or gummatous lesions numbered 23 (19 males and 4 females). These patients comprised 9.9 percent of all syphilitic patients. Additionally, using these clinical data, the author found that there was no statistically significant difference between the sexes in regard to incidence rates for tertiary syphilis.

From the evidence available, it is obviously necessary to pay special attention to the relationship between the extreme prevalence of syphilis and prostitution during the Edo period.

At that time only one licensed area of prostitution, Yoshiwara, existed directly under the supervision of the Edo government. However, there were many unlicensed areas or red light districts, the so-called okabasho, in such places within Edo as Fukagawa, Shinagawa, Itabashi, and Senju (NISHIYAMA, 1979; HIGUCHI, 1980).

Japanese prostitutes (yiijo) working in such place during the Edo period were obliged to have sexual relations with various customers every day and night, and the majority of them were also forced to endure poor daily meals and hard labor without sufficient rest. Given such a lifestyle, it would have been unusual if they had not been affected by venereal disease, gonococcal infection and syphilis possibly being the most common diseases among them, as MIYAMOTO (1979) pointed out.

KAMIMURA (1929), using some old documents concerned with prostitution in the late Edo period, has drawn the interesting conclusion that the incidence of syphilis among prostitutes might have been as high as 80 percent or more in 1867. During the late Meiji period, in addition to improvements in the hygienic environment due to careful management and effective treatment using antisyphilitic drugs such as salvarsan for venereal diseases, NAKANO et al. (1910) made serological diagnoses of syphilis using the WASSERMANN reaction on prostitutes and concluded that, "The incidence of syphilis must comprise 70 percent or more of all licensed prostitutes in whom syphilis either exists or has existed in the past."

In this context, it is possible to say that the incidence of syphilis among theyujo, regardless of their being licensed or unlicensed, could have been considerably higher in the Edo period than anyone would have imagined.

Furthermore, at least during the early part of the Edo period, it is said that the ratio of women to men was very low, this being particularly true among the lower classes. According to NAITO (1980), the relative sex ratio of the population in Edo was 100 males to 57.6 females in 1733, and 100 to 73.9 in 1798. MORIMOTO (1976) reported that the number of the male skeletons in his anthropological study was about three times greater than that of the female skeletons excavated from the Hitotsu-bashi site. With such an imbalance between the sexes in the population, it is understandable why venereal diseases should have been so prevalent among prostitutes of the Edo period.

3. Age Distribution of Cranial Syphilis

The ages of the skulls used in this study were estimated roughly and were classified into three age groups, i.e., young, adult, and mature and senile. As shown in Table 4.7, the incidence of cranial syphilis was 2.4 percent in the young, 8.7 percent in the adult, and 6.9 percent in the mature and senile. The individuals belonging to the young age group numbered only 42 cases, and among them only one case was diagnosed as cranial syphilis, this being a young female skull excavated from the Joshin-ji site.

The age distribution of the syphilitic patients observed by FUNAKOSHI (quoted from TACHIKAWA, 1976 and 1979) can be seen in Table 5.2. The syphilitic patients in this table can be classified into three age groups in the same manner as the author did. The results are shown in Table 5.3.

As is clearly evident from Table 5.3, among all of the syphilitic patients there was only one case belonging to the mature and senile age group. A comparatively high frequency of syphilis is seen in the young age group, particularly in the young females, who comprise about 28 percent of all female cases. The youngest case of syphilis reg-istered by FUNAKOSHI was a girl of only ten years of age who was the daughter of a poor carpenter and, as pointed out by OYA in 1972, seems to have had a severe papilloma or condyloma latum caused by syphilis in her urogenital region.

During the Meiji period, as shown in Table 5.1, AOSHIMA (1907) studied the clinical statistics, including the age distribution, of the venereal diseases of 310 patients. A very high frequency of disease is seen in those under thirty years of age particularly in those in their early and middle twenties.

The age distribution of 231 syphilitic patients reported by NISHIKAWA and SAITO (1903) is shown in Table 5.4. As is evident from this table, the clinical incidence of syphilis shows clearly that the most frequently involved were those of both sexes in the adult age group. Table 5.5, shows the age distribution of syphilis among 421 prosti-tutes in the special hospital for venereal diseases as reported by SAITO (1910). Because of their special occupation, most of the prostitutes infected by syphilis occurred between the ages of twenty and twenty-five and comprised about 74 percent of all cases.

According to the above-mentioned clinical statistics of the Meiji period, the highest peak of the incidence of syphilis was generally seen in the third decade of life. However, in. females, a relatively high peak occurred before the age of twenty, in the young age group.

On the other hand, syphilis after the age of fifty was rare in clinical reports, a finding which seems to contrast greatly with the present palaeopathological investigation. The reason why such a difference in the age distribution of syphilis has arisen between the clinical reports and the present study must clearly be due to the fact that the materials used in this investigation were inevitably limited to tertiary syphilis alone. As is gen-erally known, the cranial syphilis dealt with in this study is one of the most character-istic and common lesions of tertiary syphilis and one which usually occurs in older persons, i.e., the mature and senile age group. Recent epidemiological data on the age distribution of syphilis, for instance, show that the peak in frequency of tertiary syphilis occurs in those over forty-five years of age, as shown in Table 5.6.

4. Incidence of the Location of Syphilitic Lesions in the Cranium

As mentioned previously, the bones most frequently involved are the skull, tibia, femur, sternum, ulna, and radius. The results of the skeletal distribution of syphilitic lesions among the 239 cases reported by STOKES (1927) are already shown in Table 3.1.

It is evident that the skull is one of the most frequently involved regions in osseous (tertiary) syphilis. In the skull, as shown in Table 4.8, the frontal bone is by far the most frequently affected, being involved in 86.3 percent of all locations in syphilitic crania in the present investigation. Parietal bones on both sides are also often affected, and about 65 percent showed such involvement in this study. There were no significant differences in the incidence of syphilitic lesions between the right and left parietal bones.

STEINBOCK (1976), who examined thirty-seven syphilitic crania, indicated the incidences of the location of syphilitic lesions as shown in Table 5.7. As is evident from this table, the incidence rates are in approximate agreement with those of the author.

With regard to syphilitic lesions in the facial cranium, particularly in the nasal and palatine regions, the involvement reached about 11 percent (8 cases) of all syphilitic crania. Furthermore, only one case of a matured female cranium, labeled as "Keishicho No. 4," showed severe destruction around the nasal cavity and, consequently, seems to have had the typical "saddle-nose" appearance while she was alive.

Hence, the incidence of the typical syphilitic destruction of the nasal bone (which might have given the saddle-nose appearance during the lifetime) may be considerably lower in palaeopathological studies, based on the results of STEINBOCK (1976) and the author's present investigation. Such results contrast quite clearly with those of clinical studies in which such a saddle-nose formation is certainly one of the characteristic features or symptoms of tertiary syphilis. Concerning this discrepancy, STEINBOCK wrote as follows: "Perhaps this difference is due to an inclusion by other workers of cases exhibiting cartilagenous destruction of the nasal area and destruction of the soft palate which are not present in dried bone material."

5. Palaeoepidemiological Approach to the Prevalence of Syphilis during the Edo Period

It appears that syphilis was especially rampant among the common people during the Edo period when uncontrolled prostitution existed. Due to the lack of epidemio-logical statistics for venereal disease, it is almost impossible to learn directly how fre-quent the incidence rate of syphilis may have been during the Edo period, or even during the Meiji period. However, in reporting on the early prevalence of syphilis in the U.S.A., STOKES (1927), for example, stated as follows: "Here again the WASSERMANN survey has been the chief made of approach, and has yielded very high results in the colored race, ranging between 25 and 30 percent in healthy negro adults as compared with 15 percent in corresponding types of whites. Among the negro sick the percentage has been estimated as high as 50."

As mentioned previously, there was only one description concerning the prevalence of syphilis among the earlier populations in this country based on the palaeopathological investigation. SUZUKI (H, 1963) indicated briefly that three out of twenty-three skulls, which accounted for 13.0 percent of all skeletal remains of the Muromachi period ex-cavated from Kajibashi, Tokyo, had the typical features of osseous syphilis. Special attention should be paid to his brief description in which 1) the typical syphilitic skulls were those of Muromachi period and 2) the incidence of cranial syphilis is clearly higher than that in this research.

In this palaeoepidemiological study, a large enough number of skulls were used for statistical analysis to obtain directly the tentative incidence rate of syphilis among the population of the Edo period.

The confidence interval at the 95 percent level for the frequency of cranial syphilis among the people is given by

Since some different proportions of cranial syphilis were obtained in the present investigation, the estimation of tertiary syphilis among the people will vary according to the proportion used. Below, some confidence intervals at the 95 percent level for the proportion of cranial syphilis are calculated using three representative proportions observed in this study:

1. with the maximum proportion derived from the Unko-in series;
   p=0.115 0.0710<0.159
2. with the minimum proportion derived from the Muen-zaka series;
   p=0.035 0.0100<0.060
3. with the observed proportion derived from the demonstrable cases alone;
   p=0.054 0.0390<0.067

As a result, the estimated proportion of tertiary syphilis varies from 1 to 16 percent.

The incidence of cranial syphilis which should be employed when using a strictly palaeoepidemiological approach in the present study is that derived from the demon-strable cases alone (p₁=0.054). This incidence certainly should be the most reliable value for estimation from the practical point of view. Consequently, the prevalence rate of tertiary syphilis among adults during the Edo period can be estimated to have been from approximately 3.9 to 6.9 percent at the level of the 95 percent confidence limit.

Another important problem to be discussed concerns what relation the rate of tertiary syphilis bears to syphilis as a whole. From the clinical statistics reported by NAKANO (1911), out of 894 syphilitic patients there were 40 cases of the tertiary stage, com-prising 4.4 percent of all cases. According to other syphilological statistics reported by NISHIKAWA and SAITO (1903), out of 231 syphilitic patients there were 23 cases of the tertiary stage, comprising 9.9 percent of all the cases.

Recently STEINBOCK (1976) has collected a number of reports concerned with the frequency of bone lesion in cases of acquired syphilis in Western countries, as shown in Table 5.8. Concerning this problem, he stated as follows:
"... the frequency of bone lesions in cases of aquired syphilis ranges between 10 and 20 percent. Perhaps the most reliable figures are those given by BRUUSGAARD and VONDERLEHR et al, who followed a series of untreated patients over many years."

There seems to be a variety not only of frequencies of bone lesions in syphilis but also of research techniques, and these should be the subject of further investigation. At this point, it is satisfactory to consider that the proportion of the tertiary stage of syphi lis within syphilis as a whole is about ten percent. Therefore, the author will use a temporary standard proportion of 9.9 percent (p₂=0.099), which is derived from reliable clinical statistics reported by NISHIKAWA and SAITO (1903).

Thus, the tentative prevalence rate of syphilis as a whole among the adult population of the Edo period can be estimated with the two representative proportions p₁=0.054 and p₂=0.099.

As a result, we can finally obtain a value of 54.5 percent (from 39.4 to 69.7 percent at the level of 95 percent confidence limits) as the proportion of total syphilitic patients among the adult population of the Edo period.

Surprisingly, we must conclude that about the half of the adult population of the Edo period might have had venereal syphilis. On this point, many medical descriptions and essays, as previously mentioned, relating to the severe prevalence of syphilis among the people written by Nankei TACHIBANA, Genpaku SUGITA, Keisuke FUNAKOSHI, Ryojun MATSUMOTO, etc., during the Edo period seem to support the author's conclusion more than anything. Both the author's study and the above-mentioned reports point to a general tendency towards a high prevalence of venereal syphilis among the adult population during the Edo period.