Although not enough cases have been described to permit generalizations to be made, new varieties developed by traditional crossing appear somewhat more likely to show human toxicity than transgenic varieties. For example, potatoes contain toxic glycoalkaloids that, in most varieties, are at relatively harmless concentrations in the tuber (Friedman and McDonald, 1977). However, the Lenape variety, a Solanum tuberosum x S. chacoense cross developed by traditional methods (Sturckow and Low, 1961Go) for pest resistance, was not released for general planting due to illness caused by ingestion of tubers with high alkaloid content (Zitnack and Johnson, 1970Go). Another potato variety (Magnum Bonum) popular in Sweden was withdrawn from the market for similar reasons (Hellenas et al., 1995Go).
 
It is also possible to bring enzyme and substrate together in such a way as to produce new and possibly toxic secondary plant chemicals. Again in potato, and by traditional crossing of S. brevidens and S. tuberosum, the progeny were found to contain demissine, a toxic steroidal alkaloid. Apparently, a hydrogenase found in S. brevidens that produces tomatidine from teinamine, produced demissine from solanidine, a compound found in S. tuberosum but not in S. brevidens (Laurila et al., 1996Go).
 
A new celery variety, developed by traditional genetic crossing and selection for resistance to Fusarium was almost ready for commercial use when it became apparent that it caused severe contact dermatitis in field workers. The cause of the dermatitis, and probably the Fusarium resistance, was the high content of linear furanocoumarins (Diawara and Trumble, 1997Go; Trumble et al., 1990Go).