To some extent, parts of basalomas found remaining in situ following tumour excision tend to spontaneous regression. This is a well-known phenomenon and has significance for the recurrence of incompletely excised tumors. The present study involved a quantitative investigation of the relationship between recurrence and spontaneous regression. Following precisely defined excision of basalomas, the entire exterior of the excised material was examined by contrast microscopy in HE-stained paraffin sections (3-dimensional histology). Whenever tumour outgrowths were found, it was possible to document exactly their type, localization, extent, and depth of invasion. In 66 such cases no follow-up operation was performed, but only a follow-up examination after a minimum of 31 and a maximum of 113 months (average: 60 months). Only 50% of these undisturbed tumour outgrowths resulted in a recurrence during the follow-up period. A very high rate of spontaneous regression (71%) was found among the solid tumour outgrowths, but a significantly lower rate (19%) among the fibrosing tumours. Moreover, regression was dependent on the tumour remnant's mass and the clinical diameter of the tumour removed. It was independent of the depth of infiltration. Although the rate of spontaneous regression of tumour outgrowths persisting after therapy is relatively high, it cannot be predicted in individual cases. It is not possible to be certain that tumour removal has been achieved unless micrographic surgery has been continued until complete absence of tumour is proved. In all procedures that are not subsequently monitored, an unacceptably high rate of recurrence must be expected, especially in the case of fibrosing basaloma. This is commented on at length.