1. The skull is the region most commonly affected by venereal syphilis. Using 923 skulls of the Edo period, palaeopathological and palaeoepidemiological reasearch on syphilis was carried out in the present study.

2. Palaeopathological diagnosis of cranial syphilis is based upon gross morpho pathology of the lesion. Neither roentgenological examinations nor microscopic observation of the syphilitic lesion provide adequate information or value for the diagnosis of syphilis.

3. There are three important and characteristic signs (triad) in diagnosing cranial syphilis by means of gross bone pathology:

1. Syphilitic lesions are usually multifocal and diffused or widely spread throughout the entire cranium.
2. They show various pathological changes such as stellate scars with bone necrosis, sclerotic clumps, crater-like depression, perforation of the diploe, and coalesced gummatous destruction.
3. They display the healing processes or new bone production caused by inflammatory reaction of the bone. Syphilitic crania, in typical cases, show a "wrinkled" or "worm-eaten" appearance.

4. In terms of the differential diagnosis of cranial syphilis, the diseases which often produce features resembling those of osseous syphilis in dried bone were discussed from the palaeopathological point of view.

5. Eighty-nine of 923 (9.64%) skulls studied in the present study were classified into one of the following three categories based on the certainty of cranial syphilis:

1. Demonstrable cases showed the triad of cranial syphilis.
2. Possible cases showed at least two signs of the triad, and could be differentiated from other bone pathology such as primary pyogenic periost-osteomyelitis in the cranium and some bone tumors.
3. Questionable cases showed atypical or faint pathological changes on the skull which may have been caused by syphilis, but which could not be differentiated from other bone pathology with certainty.

6. The highest incidence of demonstrable puls possible cases among the caranial series or excavated sites was in the Unko-in series (11.5%) and the lowest was in the Muen-zaka series (3.5%). Significant differences in incidence were recognized in some combinations among the series. The reason for these differences seems to have been the social position or class of the interred, i.e., the incidence of cranial syphilis might have been considerably higher among the common people or the humble classes than. among the higher classes such as samurai or bushi.

7. In terms of sexual differences, 8.8 percent of the males and 5.9 percent of the females showed syphilitic lesions on the skull. This was not considered to be a statistically significant difference.

The important subjects discussed with regard to the higher incidence of syphilis in females were (a) the smaller number of females than males in Edo, particularly during the early part of the Edo period, and (b) the great prevalence of prostitution in Edo.

8. The incidence of cranial syphilis among the three age group was 2.38 percent in the young (-ca. 20 years old), 8.74 percent in the adult (ca. 20-45 years old), and 6.91 percent in the mature and senile (ca. 45+years old). There were no statistically significant differences in incidence between the three age groups.

9. The most frequently involved region was the frontal bone, the incidence of involvement being 86 percent in this study. Next, lesions of both parietal bones were involved in about 70 percent, whereas those of the occipital bone were involved in about 36 percent. The results obtained agree approximately with those of previous authors. Special attention should be paid to the remarkably low incidence of the typical syphilitic "saddle nose" found in this study.

10. In order to ensure strict accuracy in the palaeoepidemiological estimation of the prevalence of syphilis during the Edo period (a) the observed incidence of only demonstrable cases (p₁= 0.054) for the proportion of tertiary syphilis among the population of Edo period, and (b) the percentage of tertiary syphilis among syphilitic patients as a whole (p₂=0.099), based upon reliable clinical statistics of the preantibiotic era, were used in this study.

Consequently, using these two basic proportions, it was possible to estimate the incidence rate of syphilis among the adult population during the Edo period, the statistical estimated value being from 39.4 percent to 69.7 percent at the level of 95 percent confidence limits, and 54.5 percent on the average.