The overall aim of the studies reviewed in this communication was to investigate the factors influencing the induction of malformations in diabetic pregnancy. Some of the main conclusions from a large body of clinical and theoretical studies were summarized by Norbert Freinkel into the concept of "fuel-mediated teratogenesis"--that alterations in the fuel mixture offered to the conceptus due to the metabolic derangement of the mother are instrumental in the induction of dysmorphogenesis in the offspring. This concept also involves the assumption that an alteration in the fuel mixture given to the conceptus would have different effects at different time periods of the pregnancy, i.e., it would be teratogenic in early pregnancy, leading to CNS disturbances in middle gestation (potentially altering behavior) and causing subtle and complex disturbances of an anthropometric-metabolic nature in the offspring when the changed fuel mixture is present in the later stages of gestation. The notion of a disturbed metabolic signal from mother to conceptus in early pregnancy as a teratological messenger is an attractive idea. The exact nature of the signal, however, as well as its gestational timing, are still obscure. Likewise, the enigmatic mechanism by which the postulated metabolic signal is translated into an alteration of the genetic machinery is unknown. There has recently been some progress in the understanding of the pathogenesis of the congenital malformations in diabetic pregnancy. The suggestion by Norbert Freinkel that alterations in the net transfer of cellular fuels from the diabetic mother to her offspring may cause embryonic dysmorphogenesis has obtained experimental support in the sense that metabolism of the fuel(s) may produce compounds that may alter a variety of cellular processes, thus impairing the embryonic development.