238 PHYSIOLOGIC GENETICS 



after having been tested in a wide variety of species and strains, the existence of marked 

 species differences in response to teratogens prohibits the conclusion that it is non- 

 teratogenic in human beings. 



2. What are the pathogenetic mechanisms underlying malformations? — It is often impossible 

 to infer correctly, by observing a malformation at birth, how it got that way. How- 

 ever, if a malformation can be produced with a high frequency by a teratogenic agent, 

 treated embryos at various stages can be compared with untreated controls, and the 

 sequence of events from the first deviation from normality to the full-blown malforma- 

 tion observed at birth can be observed. This approach has been used to elucidate 

 the pathogenesis of a number of malformations. For instance, Monie et al. 889 have 

 demonstrated how a variety of urinary tract anomalies produced in rats by a maternal 

 deficiency in pteroylglutamic acid could be explained by retarded development of the 

 urinary tract and vertebral column. Giroud and Martinet 436 have traced the patho- 

 genesis of anencephaly produced in rats, by maternal treatment with large doses of 

 vitamin A, from failure of the encephalic tube to close, through formation of a brain 

 that is in effect turned inside out, and its subsequent degeneration. Warkany et al. 1358 

 have analyzed, by this approach, the origin of myelomeningocoele produced by 

 maternal treatment with trypan blue in rats. Failure of the neural tube to close is 

 followed by overgrowth and eversion of nervous tissue and later by degeneration of the 

 neural plate and formation of a fluid-filled space between the pia of the neural plate 

 and the dura covering the vertebrae. Thus, the myelomeningocoele is essentially a 

 cyst in the subarachnoid space in this case. Many other types of malformations that 

 can be produced at will by teratogenic procedures deserve to have their origins worked 

 out by this approach. 



It must not be concluded, however, that the pathogenesis demonstrated for a 

 particular type of malformation produced by one teratogen at one stage of develop- 

 ment is the same for all malformations of this type. It has been demonstrated, for 

 instance, that cortisone-induced cleft palate in mice results from a delay in movement 

 of the palatine shelves from their original position on either side of the tongue to their 

 final position above the tongue, and that this delay seems to be due to interference with 

 the mechanism within the shelves that provides the force necessary for this move- 

 ment. 1354 Cleft palate following amniotic puncture, on the other hand, is probably 

 due to increased resistance of the intervening tongue. 1330 Diminished shelf width, or 

 increased head width, are other possible causes for the failure of the shelves to meet in 

 the midline at the proper time. 405 It is often impossible to tell, from the appearance 

 at birth, which of these mechanisms caused the cleft in a particular case. Here again, 

 extrapolation of experimental findings to human beings should be supported by 

 observations on human embryos, but the experimental observations are useful in 

 illustrating possible pathogenetic mechanisms to be looked for in human malformations. 



3. What are the biochemical properties of an embryo at various stages of development? — The 

 teratologic approach to this question was first formulated by Warkany and his group, 

 who put female rats on a diet deficient in vitamin A. They showed that the offspring 



