II. i CELL-DIVISION 55 



has observed a similar reunion of the artificially isolated cells of 

 the Frog's egg". This Cytotropism, as Roux calls it (Cytotaxis 

 would be a preferable term), is noticed when the slide is kept 

 perfectly horizontal and streaming movements of the medium 

 (albumen) are rigidly excluded. The cells become rounded, and 

 then approach one another in, more or less, a straight line, oscil- 

 lating slightly backwards and forwards. The cells must not be 

 too far apart, not further than a radius of small, or less of large 

 cells. Groups of two or more cells behave in the same way. 



The movement may be simply a surface-tension phenomenon, 

 or, as Roux suggests, more complex, of the nature of a response 

 to a mutual chemotactic stimulus. 



These various kinds of cell-motion are also an important 

 feature in such processes of differentiation as the union of cells 

 to form muscles, tendon, epithelia, and so forth. 



A review of all the facts thus leads us to conclude that while 

 some of the phenomena of segmentation the flattening of 

 cells against one another, the pattern made by the cells in 

 cleavage, especially of the spiral type are largely referable to 

 the action of the purely physical laws of surface tension, there 

 are many cases, the radial and bilateral types, and the radial and 

 bilateral periods of spirally segmenting eggs, in which the opera- 

 tion of these laws is restricted and confined by other causes. 

 But in any case those laws can only co-operate with other factors, 

 which are to be looked for in the rate and direction of division, 

 and in the magnitude of the cells, factors which themselves are 

 dependent on the relation between the cell and its nucleus. 



Before concluding this section we have to call attention to 

 some experiments which may possibly throw some light on an 

 event of fairly frequent occurrence in ontogeny the division of 

 the nucleus without the division of the cell, l as in the formation 

 of coenocytia such as striated muscle fibres and the trophoblast of 

 the placenta ; or the fusion of distinct cells into a syncytium, as 

 in the trophoblast again ; or the secondary union of yolk-cells. 



1 In the Alcyonaria the nucleus may divide three, four, or five times 

 before the egg simultaneously breaks up into eight, sixteen, or thirty-two 

 cells. See especially E. B. Wilson, 'On the development of Renilla,' 

 Phil. Trans. Roy. Soc., clxxiv, 1883, 



