II. THE PERIOD OF ORGAN DEVELOPMENT 



143 



right limbs are mutually symmetrical. Here, again, we face the 

 problem of how the development of a primordium into either 

 a right or a left limb is determined. In order to solve this, 

 Harrison transplanted limb rudiments of Ambly stoma punctatum 

 in various stages, rotating them 180° on any one of their three 

 axes (viz. rostro-caudal, dorso-ventral, and medio-lateral), or 

 on two or three axes simulta- 

 neously. If these experiments 

 were made at an early stage, 

 a short time after neurula- 

 tion, Harrison found that the 

 dorso-ventral and the medio- 

 lateral organisation of the 

 growing limb were always 

 in harmony with that of its 

 new environment. If, how- 

 ever, the limb had been 

 rotated on its rostro-caudal 

 axis, it grew out obliquely 

 forwards, instead of back- 

 wards, so that the symmetry 

 was reversed (Fig. 52). This 



Fig. 52. Amblystoma larva, into 

 which a right Umb bud, rotated 

 on its rostrocaudal and dorso- 

 ventral axes, was grafted in the 

 flank. The graft (tr) has produc- 

 ed a limb with normal dorso- 

 ventral structure, in which, how- 

 ever, the rostrocaudal organisation 

 is inverted so that the graft is 

 the mirror image of the normal 

 right fore-limb. After Harrison. 



shows that, at the time of the operation, the rostro-caudal 

 axis was already irrevocably fixed in the material, but that 

 the other axes had not yet been determined. Further exper- 

 iments showed that the determination of the rostro-caudal axis 

 takes place at the gastrula stage already, whereas that of the 

 other axes does not occur till later stages, irreversible determ- 

 ination taking place first in the dorso-ventral, and then in the 

 medio-lateral axis. Moreover, it was found that the different 

 species of urodeles did not behave identically in this respect. 

 In Triton taeniatus, for example, the direction of the dorso- 

 ventral axis is determined much earlier than in Amblystoma 

 (Brandt, 1922-28). The sequence, however, in which the three 

 axes become determined, is probably the same everywhere. 

 Presumably, the limb mesoderm is solely responsible for this 

 polarity, and the ectoderm does not play a role on this point 

 (Balinsky, 1931). 



