Amphibians 



233 



ficance. This problem has a long tradition; it 

 is the modern version of the "preformation- 

 versus-epigenesis" controversy. 



The classic constriction experiments on 

 2-cell stages of urodeles (Endres, 1895; Her- 

 litzka, 1897; Spemann, '01, '03; and others) 

 and on blastulae of anurans (Schmidt, '33), 

 and the isolation experiments of early blasto- 

 meres (Ruud, '25), which resulted in the 

 formation of twin embryos or duplications, 

 demonstrate convincingly that in amphibians 

 no rigid pattern of organ precursors exists in 

 the 2-cell stage. This notion is supported by 

 the fusion experiments of Mangold ('20) 

 and Mangold and Seidel ('27), in which two 

 2-cell stages were united crosswise, resulting 

 in giant normal embryos or two to four axial 

 systems. 



The constriction experiments gave at the 

 same time a clear indication of an incipient 

 difference between the future dorsal and 

 ventral sides. Spemann found that, in numer- 

 ous cases, only one blastomere formed a 

 normal embryo, whereas the other one formed 

 a viable spherical structure without axial 

 organs, in which only pronephric tubules, 

 mesenchyme and blood cells were differenti- 

 ated (Bauchstuck, belly piece). Two com- 

 plete individuals were formed whenever 

 both separated blastomeres contained part 

 of the upper blastoporal lip. One embryo and 

 a belly-piece resulted when the entire dorsal 

 lip was contained in one blastomere. Ap- 

 parently, factors or materials which are 

 necessary for invagination and for the forma- 

 tion of axial organs are localized early on 

 one side. Spemann ('14, '28) found by con- 

 striction of tmcleaved eprgs that this prelocali- 

 zation exists before cleavage begins, and 

 Fankhauser ('30) and Streett ('40) traced 

 it back to stages shortly after fertilization. 

 Centrifugation experiments of Pasteel's ('41) 

 indicate that this precursor of axial organs 

 becomes stabilized at the time of the forma- 

 tion of the gray crescent. 



The constriction and fusion experiment<5 

 do not rule out the possibilitv that a more 

 detailed pattern of regional differences exists 

 in very earlv staeres; but if so, it is verv 

 labile, and reversible. The results are in full 

 agreement with those obtained in the ex- 

 periments on axial determination; both dem- 

 onstrate that the prelocalization of a 

 dorsalizing factor is the first step in organ de- 

 termination. Whereas there is general agree- 

 ment that the area corresponding to the 

 chorda-mesoderm field of the earlv gastrula 

 is blocked out first, the theoretical interpreta- 

 tion of its origin has followed different lines. 



INTERPRETATION OF DORSALIZATION 



IN TERMS OF LOCALIZED 



PRECURSORS 



Spemann, Mangold, and others have con- 

 sidered the dorsalizing agent to be the 

 actual precursor material of the chorda- 

 mesoderm, and Lehmann ('42a, '45) has 

 suggested that the pigmented marginal 

 plasma which Banki ('27a) and he himself 

 have found in the equatorial zone of some 

 urodeles corresponds to the marginal zone 

 of the gastrula. This would not necessarily 

 imply that this precursor is already en- 

 dowed with inductive capacity from earliest 

 stages on. Holtfreter ('34b), Mayer ('35) 

 and Waddington ('38b) have shown that the 

 uncleaved egg contains agents which are 

 capable of inducing neural tissue, but there 

 is no evidence for their prelocalization at the 

 dorsal region. 



Certain experimental findings do not seem 

 to be in agreement with the concept of local- 

 ized precursors. Penners and Schleip ('28a, b) 

 and Penners ('36) found in their inversion 

 experiments that, in exceptional cases, the 

 hemisphere not containing the gray crescent 

 formed a blastopore and subsequently an 

 embryo. Pasteels ('38, '40, '41) observed 

 in rotation and centrifugation experiments 

 that occasionally a blastopore originated at 

 a distance from the site of the gray crescent. 

 Spemann ('38) and Lehmann ('45) have 

 interpreted the results of Penners and Schleip 

 in terms of a translocation or splitting of 

 mesoderm-precursor material dvie to the ro- 

 tation, and the results of Pasteels could be 

 explained in the same way. 



Hollander and Darby ('49) and Dollander 

 ('50) found in constriction experiments on 

 Triturus palmatus, in which the gray cres- 

 cent had been marked, that the blastomere 

 opposite to the gray crescent gave rise to 

 normal or microcephalic embryos in a con- 

 siderable nvimber of cases. The indistinctness 

 of the gray crescent in this species introduces 

 a serious source of error. Furthermore, one 

 should realize that the gray crescent is not 

 in all species a reliable indicator of the 

 dorsal side. For instance, Schechtman ('36) 

 found in Triturus torosus, in not less than 

 27 per cent of the eggs, a divergence between 

 the plane of symmetry of the gray crescent 

 and that of the gastrula, ranging between 25 

 and 90 degrees. 



Altogether, the assumption of an early 

 prelocalization of a precursor material for 

 chorda-mesoderm has not been invalidated 

 by experimental data. 



