1. With the aseptic operation method of Woerdeman (1930) it is possible to rear duplicitas larvae, formed after grafting of the dorsal blastoporal lip in the place of the ventral one at the early gastrula stage (Spemann, 1918), till an age of about 4 weeks, a time at which they have completely consumed their yolk material.
2. Various forces acting in the normal development and in the development of the duplicitas larvae are discussed in connexion with the external form and the internal organization of the latter.
3. The sources of errors in the method of graphic reconstruction of the brain used in this investigation are checked and their elimination is discussed.
4. The structure of the duplicitas shows that there exists a mediolateral and probably also a cranio-caudal regulation in grafts of the dorsal blastoporal lip of the early gastrula.
5. In the brain one may distinguish between ‘primary’ structures which are always present, and ‘secondary’ structures which are more variable. Therefore, one may probably divide the causal analysis of the development of the central nervous system into the determination of the ‘ground plan’ of the primary structures and the more dependent development of the ‘secondary’ ones (specific formations).
6. The brains of the secondary embryonic rudiments of these duplicitas are formed in all cases up to a certain level; caudally to this level the brain-parts are normally proportioned, but the most anterior part may be reduced in size. This material can be arranged into a complete regression series, beginning with the reduction of the telencephalon and ending with a completely acephalic embryo. From this development one may conclude that the determination of the central nervous system occurs in a large number of successive zones, which are qualitatively ‘equivalent’, and which are determined by (qualitatively or quantitatively) different values of the organizing ‘agent’.
7. There are indications that the infundibulum and hypophysis are determined by a triple contact between presumptive neural plate, prechordal plate and cephalic ectoderm after the appearance of the neural plate.
8. After special staining these incomplete nervous systems may form a very important source of data for the causal analysis of the development of tracts and nuclei in the central nervous system.
9. With the conclusions drawn from these experiments and the data given in the extensive literature on this subject, a new working hypothesis on the determination of the central nervous system is put forward. This theory involves the assumption (among others) of an equilibrium reaction between a strong mesodermal and a weak ectodermal gradient. The character of the determination process probably changes from being at first purely quantitative to become qualitative with increasing age.
10. Finally, the regional determination of the secondary rudiment in duplicitas is probably determined by a triple interaction between the induction field of the graft, the primary regional structure of the ectoderm and the regional influences of the primary rudiment.