Genie Control of Development 415 



field plus the following anterior wing field from the rest, again a 

 mirror-image regulation occurs which now produces a trisegmented 

 structure, the tip segment of which is inflated. If this structure has 

 time (depending upon the time of incidence of the split) to con- 

 solidate, the resulting structure greatly resembles a haltere. There is 

 reason for the belief that the normal haltere is homologous only to 

 the anterior half of the wing. The stages of transformation of a haltere 

 into a wing in the mutant tetraptera (Astauroff, 1929) bear this out. 



The next grade of transformation of the wing anlage in pod and 

 especially tet is produced when one cleft separates the cubital part 

 ( with or without a part of rhe anterior wing blade ) at an early stage, 

 and another cleft separates the posterior wing and alula anlage. If 

 this happens, the parts behind the cubital cleft cease to be determined 

 as wing, but become redetermined as thorax and assume the structure 

 of one-half scutum in mirror image, from the wing blade, and one-half 

 mirror-imaged scutellum, from the alula anlage. It is impossible that 

 material of the thoracic disc takes part in this transformation: the 

 thorax remains always perfect and unaffected. Either the thoracic disc 

 acts as an inductor for thoracic structure in the wing anlage or the 

 wing anlage contains — in strict comparison with aristopedia — both 

 wing and thorax inductors, the latter of which gains ascendancy in the 

 mutant. However, the cubital part never is affected by the thoracic 

 inductor, but always forms its wing leg or a rudimentary segmented 

 palpus or the haltere-like structure. It seems that the anterior part of 

 the wing bud contains cubitus inductor, leg inductor, and haltere 

 inductor; the rest of the disc, wing inductor and thorax inductor. Thus 

 the discussion of aristopedia applies in its general form also to this 

 case. A new factor is that the adjacent thoracic disc influences the 

 redetermination of wing anlage into thorax at least to the extent of 

 determining its laterality (mirror image). But this may be something 

 different from induction, belonging, rather, to the phenomenon of 

 regulation, as mirror imaging is typical for superregeneration (Bate- 

 son's law). 



The series of tetraltera transformations includes a still more 

 extreme case of embryonic regulation and integration which takes 

 place after the redetermination of wing anlage into thorax. If this 

 redetermination takes place very early, the development of the pos- 

 terior wing part into an appendix (even one with the structure of a 

 half thorax) ceases. Instead, we find all stages of the incorporation of 

 the wing anlage into the general structure of the thorax. In the end, 

 the wing anlage is pushed behind the genuine thorax and incorporated 



