270 



METAMORPHOSIS 



that exceptions are known, as, for example, in cells which divide longitudinally 

 when one would expect transverse division (e. g. cambium cells). 



The occurrence of cell-walls as surfaces of ' minimum area ' might be ex- 

 plained most simply by assuming that the freshly formed wall is in a liquid con- 

 dition. Errera (1886) is responsible for this view, though observation would 

 lead us to an opposite conclusion. More recently, Wildeman (1893) has suggested 

 that it is sufficient if the cell-wall be in a liquid condition for one moment only and 

 then becomes rigid. Such a hypothesis cannot be denied, at least in the case of 

 simultaneous cell-wall formation, but it does not lend itself to the explanation 

 of the siiccedaneous type. There, undoubtedly, one part of the wall already existed 



^ 



V\f,. 60. After Berthold (Protoplasma mechanik). 



in the solid state, while the rest is not even deposited. What is it, then, that makes 

 the cell-wall grow on in a definite direction ? Wildeman has shown that even a 

 liquid lamella may increase by succedaneous accretions. If one places a soap film 

 within a rectangle made of iron wire (Fig. 61, /) into which runs a silk thread 

 attached at a and b, the ends of which pass through a straw suspended in the 

 soap film, and if the lamella be pierced between a and b and the threads, one 

 obtains the condition represented at Fig. 61, 2. If now the ends of the thread 

 be pulled (as indicated by the arrows in the figure) the soap film gradually grows 

 from right to left till it fills the whole rectangle. There are 

 great difficulties, however, in applying this experiment to 

 dividing cells. What corresponds in the cell to the silk 

 threads and to the frame ? No answer can be given to this 

 question ; but without the threads and the framework no 

 succedaneous growth of the lamella is possible. 



Although at first sight one is tempted to refer cell 

 division to the laws governing the equilibrium of liquid 

 membranes, still it is not possible for us to accept this 

 explanation. We must content ourselves with pointing 

 out the great similarity between these two sets of phe- 

 nomena and confessing that the reasons for this comparison 

 MAN'S experiment. ^^^ ^^-jj f^^ ^^^^ apparent. 



After the sketch we have now given of cell division it will be obvious that 

 we must attribute very important functions in that process to the nucleus, and 

 especially to the phenomena of mitosis. We must not, however, overrate their 

 importance, for we must remember that quite normal cell division may take 

 place without mitosis, and indeed without segmentation of the nucleus at all. 

 Nathansohn's (1900) and Wasiliewski's (1902) researches show that by em- 

 ploying certain anaesthetics (ether, chloral hydrate) the normal nuclear division 

 (mitosis) taking place in the cells of Spirogyra and of the root apex of Vicia jaba 

 may be transformed into the so-called direct nuclear division, where the 

 nucleus divides into two parts without any formation of chromosomes or nuclear 

 spindle and between these two nuclei a partition wall is formed. [Wisselingh 

 (1903, Bot. Ztg., 61, 201) has raised objections to this interpretation of Nathan- 

 sohn's discoveries. The controversy between these authors (Bot. Ztg., 1904, 62, 

 II, 17) suggests the necessity for a re-investigation of the whole question. Doubt 



Filj. 61. De Wii.de- 



