148 SCIENCE PROGRESS. 



the centrosome to be regarded, generally speaking, as a 

 permanent if often concealed "organ of the cell". I mean 

 the wide-spread and apparently well-founded belief that it 

 is not formed ti^e novo, but only arises as the result of divi- 

 sion of a pre-existing centrosome. I shall revert to this 

 question later ; but in this place will mention three sets of 

 facts which are difficult to reconcile with the absolute con- 

 tinuity-of-existence theory. The first I take from the 

 animal kinQ^dom. Brauer^ observed, whilst investioatincr 

 the development of the parthenogenetic eggs of Arteinia 

 salina, that there was no centrosome associated with the 

 first polar division, but that afterwards w^hen the Qg^ was 

 ready to segment to form the embryo a centrosome became 

 visible. Brauer, it is true, remarks that the centrosome 

 might have existed in a modified or distributed state during 

 the first karyokinesis ; but this is merely a pious belief, and 

 all more recent work, with the exception of Hacker's state- 

 ments as regards Ascaris, seems to tell against it. Brauer 

 himself only puts it forward apparently as a means of saving 

 the reputation of his &gg centrosome. The second and third 

 sets of facts I shall draw from instances of which I have 

 more special personal knowledge. In the liverworts, the 

 two main subdivisions of Marchantiaceai and Jungerman- 

 niacese are characterised by differences in the formation of 

 their spores. The former, the Marchantiaceae, do not for 

 the most part call for special notice. With the exception 

 of Fegatella and Tai'gionia the divisions in their spore 

 mother cells so far as is known coincide in their main fea- 

 tures with those seen during the formation of dicotyle- 

 donous pollen grains. That is to say, the nucleus undergoes 

 two perfectly normal bipartitions. The shape of the mother 

 cells is commonly (with the exception of the two genera 

 above mentioned) spherical. It is widely different in the 

 Jungermanniaceae.^ Here the mother cell, as it grows 

 prior to its final division, becomes a four-lobed body, and 



^ Brauer, " Zur Kenntnis d. parthenogenetisch sich entwickelnden Eies 

 von Artemia salina," Archiv filr Mikr. Anat., Bd. xliii. 

 - Farmer, loc. cit. 



