The nature of covering epithelium of the ovary is intriguing. Although only a part of peritoneal mesothelium, it proliferates to repair the minor trauma due to the ovulation, and thus has occasionally tumorigenic potential. The concept that the epithelium simulates the müllerian form in tumor formation has developed by degrees and now becomes a firm policy in the classification of ovarian cancer [16]. Although mucinous tumors are also placed in the common epithelial category, it is still disputed whether the origin of a certain group of mucinous neoplasms composed of intestinal type cells arose from germ cells, namely monophyletic teratomata.
In our previous experimental study, in which a chemical carcinogen, 7,12-dimethylbenz [a]anthracene (DMBA), was directly applied to the rat ovarian surface, an ovarian cancer was observed in about half of the DMBA-treated rats. The histology of the induced tumors also simulated the epithelia of rat genital tracts. In the experiment, however, an unexplainable cancer composed of heterologous osteoid tissue was observed [17]. At that time, if the covering tissue of the ovary were not called "surface epithelium", but named "germinal epithelium" instead, the tumor might have been classified as a teratomatous osteosarcoma.