S U R F A C E S r 1< U C T U R E S 37 



produce one large and one small daughter bacillus, i.e. containing more and 

 less cells. 



Filaments of a different type, without cross-walls, arc formed in a more 

 elaborate, probably syngamous type of cell division, which occurs in many 

 difterent bacterial groups. These filaments eventually fragment to form new 

 bacterial units. 



In all forms of cell division, the new cell wall material, whether it appears 

 as a constriction or as a complete cross wall, is secreted by a basophilic septum 

 derived from the cell membrane, which appears across the cell as the first 

 sign ot incipient division, exactly as in certain types of plant cell. The con- 

 strictive ingrowths are secreted at the points of junction of the cell wall and 

 cross-walls or septum, and in the septate bacilli and cocci it is here that the 

 main growth oi the cell wall proceeds. The new surface is, as it were, passed 

 outwards around the edge ot the internal division between the cells to the 

 (Hitside surface. The work of Tomcsik (1954) confirms that the main growth 

 of the capsule is also at this point, where a distinct thickening of the cell wall 

 can be observed by his methods. A similar thickening can be observed also 

 in sections of bacteria prepared in this laboratory. In the unicellular, non- 

 septate types the new cell wall is formed at the point of division and also at 

 the tips of the cell ; usually only at one tip, the growing point. The point 

 of division becomes the growing point of one or both of the daughter cells, 

 so that there is no fundamental difference between the septa and the growing 

 points. 



In very young cultures division may be so rapid that a single bacillus is 

 subdivided into three or four cells, by septa derived from the cell membrane, 

 while the process of ingrowth of the wall is incomplete. 



The evidence of the growth of " smooth " bacteria, and the formation of 

 the cell wall at a growing point at one tip is derived from several sources. 

 The first is the arrangement of the surface structures and especially the flagella 

 in electron micrographs of growing and dividing bacteria and germinating 

 microcysts. In such cells the flagella appear progressively shorter towards 

 one pole where the wall is relatively thin, flexible and electron-transparent. 

 Frequently one daughter cell may be provided with a full quota of well- 

 ' developed flagella, whereas the other has very short flagella or none. It 



