442 M. N. Choloclkowsky on the 



which are originally simple to subsequently split longitu- 

 dinally, whereby two pairs of appendages are produced, which 

 all taken together may be homologous with one pair of legs. 

 In the second place, if the hypothesis that the ventral styles 

 correspond to the coxal appendages is correct, the ventral 

 styles of the eighth and ninth segments in Machilis may 

 represent the coxal appendages detached from the trunk of 

 the extremities. Other objections against the homologization 

 of the gonapophyses with the legs are based upon the late 

 appearance of the former, which are therefore supposed to be 

 " secondary " structures. I have already had occasion to 

 point out that more or less late appearance of the organs in 

 development is of little importance for the setting up of homo- 

 logies ; I am convinced that even organs which first appear 

 in post-embryonic life may be equivalent to those which are 

 developed at a very early stage, since there really exists no 

 radical difference between embryonic and post-embryonic 

 development. If certain organs may be referred to purely 

 post-embryonic adaptation, we are nevertheless not bound to 

 consider as phylogenetically secondary all structures which 

 are post-embryonic in appearance. In the particular case of 

 those Insects in which the development of the gonapophyses 

 has been sufficiently investigated [e. g. the bee) the homology 

 of the latter with the legs appears to be precisely very pro- 

 bable. According to Uljanin*, in the bee the sting develops 

 from two pairs of appendages at the posterior end of the abdo- 

 men, and the hindermost appendages very quickly split 

 longitudinally. It is stated by Biitsclili t that in the embryo 

 of the bee the two posterior pairs of abdominal appendages 

 are especially developed. Grassi J, too, alludes to these 

 appendages, although (contrary to Biitschli) he denies the 

 presence of the rest of the abdominal extremities. It appears 

 that these hindermost abdominal appendages subsequently 

 greatly diminish in size, so that immediately before hatching 

 takes place they are represented by flat ectodermal disks 

 (Kowalevsky §). It is only after the second ecdysis of the 

 larva (according to Uljanin) that they commence to grow 

 bigger, and, what is especially interesting, they even exhibit 

 an indistinct segmentation. The development of these 

 abdominal appendages therefore retrogrades somewhat to- 

 wards the end of embryonic life, and it is only in post- 



* O2). cit. t Op. cit. 



X B. Grassi, " Intorno alio sviluppo delle Api nell' novo," Atti dell' 

 Acad. Gioenia di Scienze Natural, iu Catauia, S. 3°, vol. xviii., 1884, 

 78 pp., 10 plates. 



§ Op, cit. 



