In 1993, Chun and coworkers,⁵¹ in an article given to the subject of "Malignant melanoma in children," presented the results of a retrospective study of melanomas that occurred in Puerto Ricans up to 16 years of age. During the period between 1973 and 1990, seven children out of 742 patients with melanoma in the Puerto Rico Cancer Registry were identified as having melanoma. No photomicrographs were shown and discussion of the histopathologic findings was limited to these two sentences:

Chun et al. observed correctly that "although rare, malignant melanoma in children can be as aggressive as in adults."

In 1994, Bartoli and associates⁵³ published their findings in 17 patients 14 years of age or younger whose cutaneous melanoma was diagnosed histopathologically at the Instituto Nazionale Tumori of Milan during the period of 1975 to 1991. Nine children were 10 years of age or younger. No histopathologic findings were mentioned other than the thickness of the lesion and the Clark's level when it was available, and no photomicrographs were pictured. Nine patients were alive and said to be free of disease; the remaining eight were dead. Of the children who died, three were 10 years of age or younger, namely, a 2-year-old male with a Stage III melanoma, the primary neoplasm having originated on a leg, a 9-year-old male with Stage III disease, the primary neoplasm having arisen on a foot, and a 10-year-old girl with a Stage I melanoma situated on the chest wall. No details regarding the thickness of the melanoma, the clinical course, or the details of her death were provided. According to Bartoli et al. three of the melanomas in their series began in a "congenital nevus" other than the giant type, one developed in conjunction with an acquired nevus, and six arose de novo. No other information was available in seven patients. On the basis of information given in this article, it was not possible to determine whether or not the melanoma in the children who died arose de novo. Bertoli and colleagues concluded rightly as follows:

In 1995, Nitta and associates, in an article about "Malignant Spitz's nevus in a 2-year-old Japanese child,"⁵⁴ told of a large nodule over the Achilles tendon that they diagnosed histopathologically as a "malignant Spitz's nevus" (Fig. 12). This is how they described the histopathologic findings in that neoplasm:

The conclusion of the authors is illogical; "prominent lymphatic vessel invasion by melanocytes" is virtually synonymous with metastasis of a malignant neoplasm.

By their own admission, Nitta and coworkers had reason to doubt the legitimacy of their thesis that the neoplasm under consideration was a malignant Spitz's nevus. This is how they couched those reservations:

"Our case showed many features of spindle cell and epithelioid cell nevus (Spitz's nevus), but several findings were out of the ordinary: the large size (27 x 17 mm), invasion of the subcutis, mitotic figures deep in the tumor, no maturation of melanocytes, and prominent lymphatic vessel invasion by melanocytes. These findings, obviously beyond the spectrum of an ordinary Spitz's nevus, could lead one to consider the case under study as malignant melanoma . . . Since our case presented features quite similar to those described by Smith et al., we considered the child to be another typical case of MSN [malignant Spitz's nevus]."

Instead of concluding that Smith et al. at the AFIP had opted wrongly in 1989 when they fostered the concept of malignant Spitz's nevus, Nitta et al. followed them down the same cul de sac and with the same unfortunate results.