ON A PEACH-COLOURED BACTERIUM. 35 



racter (indicated by the uniform distribution of the colouring 

 matter) and grow equally at the periphery so as to form 

 small discs. These are represented in fig. 11 when of small 

 size. 



This uniform homogeneous growth may proceed until 

 single plastids attain the very lai-ge diameter of -^i-oth of an 

 inch or more. This is accompanied frequently by great in- 

 tensification of the colour, indicating a special physiological 

 activity in the development of Bacterio-purpurin. Not 

 unfrequently, however^ the process which leads to the ela- 

 boration of Bacterio-purpurin is so modified as to give rise 

 to the brown variety of that pigment. Such discs, on account 

 of their large size, I would call " macroplasts." These macro- 

 plasts are certainly of a somewhat permanent character. I 

 have kept one under observation for fourteen days without 

 observing any change in its size or structure. 



The conditions which are necessary for this change are not 

 known, but clearly enough they exist, and at one time or 

 another the macroplast falls under their influence and takes 

 on a new development. This may occur either when it is of 

 comparatively small size or when it is of larger growth. The 

 addition of fresh nutrient material to the jar containing the 

 macroplasts led to their ultimate disappearance and their 

 replacement by colonies of plastids of types similar to those 

 previously described. Hence I am inclined to infer that 

 under favorable circumstances as regards nutrition the 

 macroplasts break up into immense numbers of small plastids. 

 The first step in this process is seen in the excessively deli- 

 cate granulation of such specimens as figs. 17, 18, and 21. 

 In fig. 12 we have a small specimen in which the newly 

 forming plastids have attained larger size and so on through 

 the series — a gelatinous exudation being produced as the 

 process of segregation advances (figs. 13, 9, and 15). 



If this be a correct interpretation of the appearances de- 

 picted in the plate (Plate III), we have an instance of 

 multiplication of " centres of organization " which does not 

 conform to the type observed in ordinary cell-division, and 

 which is comparable to free cell-formation in a blastema. 

 Each plastid in any growth or aggregation of Schizomy- 

 cetes may be regarded as a unit or centre of organization — 

 just as much so as a true cell, though the Schizomycetous 

 plastid differs from a cell in not possessing a central mass 

 corresponding to the nucleus. The transverse division of an 

 elongated or filamentous plastid of one of the Schizoraycetes 

 is equivalent to an act of cell-division and so far the proto- 

 plasm of these organisms conforms to the general mode of 



