OS A PEACH-COLOURED BACTERIUM. 39 



ditions under which the larger and the smaller form respec- 

 tively occur in the tAvo cases are similar. The smaller in 

 both cases occurs where nutrition is abundant and the sur- 

 roundings rendered special by human agency; the larger 

 occurs in natural waters (ponds and streams ); where putre- 

 scence is going on in the midst of conditions favorable to the 

 maintenance of other forms of life. 



The Causes of the Aggregation of Schizomycetous Plastids 

 to form Compound Structures of regular pattern. — I devote 

 a separate paragraph to this subject, not because I have much 

 to say with regard to it, but in order to draw attention to the 

 great peculiarity (already noticed above) in the mode of for- 

 mation of many of the aggregates of Schizomycetes. There 

 is no doubt that the zoogloea or glceogenous masses of spherical 

 or cavernous (Clathrocystis-like) form increase in bulk by 

 the transverse division of their constituent plastids. But 

 there is reason to doubt whether division along the 

 longer axis of these plastids ever takes place. In the case of 

 the delicate films of Schizomycetous plastids Avhich have been 

 spoken of as' the " mycoderma-phase " of growth, it results 

 from observations of the arrangement and connections of 

 the plastids that they must take up their positions to a large 

 extent as the result of a spontaneous movement of apposition 

 — a kind of mutual attraction, similar to that exhibited by 

 mammalian blood-corpuscles when uniting to form " rou- 

 leaux." 



The operation of this process of adhesion is modified no 

 doubt by the continued growth of the plastids and their self- 

 division at right angles to their long axes. 



The definiteness and beauty of the patterns resulting is 

 very striking. Some of the most curious, exhibiting a very 

 elagant combination of curved lines, are due to colourless 

 species of Schizomycetes, and have never, as far as I know, 

 been figured. The most striking of these patterns due to the 

 " fortuitous concourse " of the plastids of B. rubescens are 

 those seen in the " regular tesselate " form, Plate III, fig. 5, 

 and Plate XXIII of my former paper, figs. 19 and 21. 



Addendum. — I have omitted to describe two additional 

 frond- forms, or aggregates exhibited by Bacterium rubescens. 

 The first is not uncommon in very abundant growths of 

 small biscuit-shaped plastids. The plastids are arranged in 

 star-like groups consisting of 10 to 20 plastids, meeting one 

 another at a centralpoint. This form of aggregate I mentioned 

 in my former paper as observed in the case of the acicular 

 form of plastid (see Plates XXII, XXIII, of former memoir, 



