8. APPLICATIONS AND PREDICTIONS 143 



gather. Since the different cell types become competent in this respect at 

 diff'erent times, and since the stimulus occurs only when cells reach the outer 

 surface of an aggregate of randomly-migrating cells, a sorting into germ layers 

 results. Curtis has called this mechanism temporal specificity, to distinguish it 

 from the theories of area specificity that have been proposed to account for the 

 sorting-out process (Weiss 1947; Steinberg, 1958, 1962). By changing the time 

 at which different cell types commence reaggregation after dissociation from 

 gastrulae, Curtis was able to cause inversions in the normal layering of cells in 

 his reaggregates. For example, if endoderm begins the reaggregation process 

 6 h before ectoderm and mesoderm, then the endoderm ends up as the outer- 

 most layer with ectoderm and mesoderm being layered normally with respect 

 to each other inside. Intermediate states of mixing were also obtained by 

 altering the time intervals between commencement of reaggregation for the 

 diff'erent cell types. Thus Curtis seems to have extended the notion of compe- 

 tence to cover the phenomenon of cell sorting, at least in Xenopus laevis 

 embryos (Steinberg (1962) disputes the generality of the process), and his 

 experiments give strong evidence for the existence of some kind of timing 

 mechanism in the embryonic cell. 



It is of some interest to note that the time intervals which Curtis found to 

 throw the normal reaggregation process out of order, 4 to 6 h, are exactly the 

 same order of magnitude as those which we have estimated for primary oscil- 

 lations in biochemical control mechanisms in the cells of higher organisms. It 

 would not be unreasonable to suggest that the gradient of metabolic activity 

 which exists between embryonic cell types is reflected in diff'erent oscillatory 

 frequencies, ectodermal control systems oscillating faster than mesodermal, 

 which again are faster than the endodermal oscillators. We might suggest, then, 

 that the state of competence to respond to a change of shearing force in em- 

 bryonic cells occurs at a certain part of an endogenous cycle, when particular 

 protein populations are at critical levels. In a more general context, Flickinger 

 (1962) has proposed the theory that the timing mechanism involved in embryo- 

 logical competence operates by turning genes on and off'in a particular temporal 

 order, independently of the environment of the cell. Induction is then simply 

 the provision of an adequate supply of energy and precursors to a cell at a 

 particular time, so that the genes which are then in an active state can express 

 themselves, and in so doing they presumably become stabilized. Such a theory 

 throws much more emphasis on competence than on induction, for in it 

 inducers can be non-specific molecular species. This absence of specificity in 

 inductive stimuli has been a recurring observation by embryologists since the 

 demonstration that such totally dissimilar and unnatural agents as methylene 

 blue (Waddington, Needham, and Brachet, 1936)) and either high or low pH 

 (Holtfreter, 1945) could cause the induction of neural tissue in gastrula ecto- 

 derm. There is no doubt that competence and induction are complementary 

 phenomena in developing embryos, but the view that Flickinger develops is 

 that competence is a much more active and specific process than has been gener- 

 ally accepted heretofore. His suggestion that genes are turned on at certain 

 times in the diff'erentiation of a cell, independently of its environmental 



