Report on the Bacteriology of Water. 267 



Examined under a low power (-in. objective), the radiating strands 

 are found to consist of closely woven and interwoven, wavy, and 

 coiled tresses of filaments, with a distinct yellowish colour such as 

 protoplasmic structures usually present by transmitted light. These 

 tresses (fig. 2) break up into finer and finer strands, as said, towards 

 the margin, and finally into single filaments. 



The coarser tresses may consist of twenty to thirty closely twisted 

 or woven filaments, and are perfectly evident even to the unaided 

 eye in older (e.g., six days) colonies, the margins of which, as shown 

 in fig. 3, remind one of a complex river-system as depicted on a map, 

 or even of the yellowish, radiately- veined plasmodium of some 

 myxomycete. 



I suppose this radiating meshwork of yellowish-white strands was 

 one character that suggested the resemblance to a network of fine 

 rootlets, creeping on a flat substratum, and so gave origin to the 

 German name ( Wurzel-bacillus) for this species. 



As we shall see subsequently, the individual filaments, twisted and 

 woven to form the tresses and meshwork referred to, are the very 

 long, wavy, coiled, or straight cell-series of the schizomycete. 



The liquefaction of the gelatine at 15 C. commences only slowly 

 about the second day, when the centre of the colony is soft beneath 

 the dense membrane of tresses ; it gradually progresses day by day, 

 until, about the fourth day, the plate can no longer be upturned 

 without liquid oozing from below. In about a week most of the 

 colonies are floating on the fluid mass of liquefied gelatine. The 

 liquid is slimy, and never seems to become completely watery, as is 

 the case with some species. At higher temperatures, such as 20 and 

 25 C., the liquefaction is more rapid, in proportion to the quicker 

 development and growth generally. 



Colonies submerged deep down in the solid gelatine of a tube grow 

 also, though more slowly than those at the surface ; this is enough 

 to show that the organism is partially anaerobic, though only to a 

 very slight extent. The stab-cultures to be described demonstrate the 

 same fact. The ultimate filaments themselves are long, straight, 

 wavy, or spirally- coiled cylinders, averaging about 1'75 p. in diameter, 

 and regularly divided into segments by transverse septa, at intervals 

 of from 3 to 5 or 6 p. It depends on the stage of development and 

 other factors how far these segments are isolated from one another. 

 In some cases, especially in the young colonies, the cylindrical fila- 

 ments are merely septate ; in others they are indented at the septa, 

 and we may then speak of the filaments as more properly seg- 

 mented ; while in yet other cases the segments are so nearly isolated 

 evidently by more or less complete fission of the septa that it is 

 almost necessary to regard the filaments as long chains of segments 

 end to end. 



D 2 



