308 Prof. R. Boyce and Mr. A. E. Evans. 



well as the experiment of mixing a small quantity of the bacterium 

 in liquefied sterile gelatine and then pouring it into a Petri dish ; in 

 the latter case numerous zoogleaform colonies develop in the solid 

 gelatine, and send out branches which ramify in all directions. The 

 branching zooglea of Cladothrix dichotoma. commonly known as Zooglea 

 ramigera, follows closely the arrangement and branching of the 

 filamentous stage ; similarly, the " linear, globose, arborescent, reti- 

 cular, and tesselated aggregations " of the Bacterium rubescens of 

 Lankester probably represent local cell aggregation. As our figures 

 show, the mode of aggregation leads to various forms, but they 

 usually possess one feature in common, and that is spirillation, and a 

 large number may be aptly compared to the twisted strands of a rope. 

 The segments of which they are composed are small bacteroid or 

 coccal forms, and they appear to be formed by the deflection and 

 division of the minute segments into which the original single 

 filaments break up. They rapidly follow the filamentous growth in 

 the symmetrical pinnate cultivations ; like the filaments which pre- 

 cede them, they assume the geotropic position. They may extend 

 great lengths and be quite unbranched, or branching may occur as in 

 figs. 30 and 31. Like the filaments, they form in the gelatine ; we 

 have not observed them in agar or broth. In fig. 31 a sarcina-like 

 arrangement of the segments is seen ; this is of interest, because, as 

 Billet has pointed out, the sarcina of the lower bacteria probably is 

 the representative of the zooglea phase of the higher bacteria. 



Upon the Phenomenon of Negative Geotropism exhibited by Bacterium 



Zopfii. 



If a tube of gelatine is inoculated with a streak culture and kept in 

 a nearly vertical position, a regular pinnate growth will, in the 

 majority of cases, be obtained, identical with fig. 1. If a series of 

 such tubes are placed at intermediate positions between the vertical 

 and horizontal, it will be found that the growths become more and 

 more irregular as the horizontal is reached. These results we have 

 repeatedly obtained since May, 1892. But apparent failures do 

 occur : e.g., if the gelatine is too stiff, an irregular growth results, as 

 in the case of agar, although the tube is kept vertical ; microscopic 

 examination of these irregular growths shows the skein phase. Simi- 

 larly, if the temperature is low the symmetry is greatly interfered 

 with. Experiments in the same manner with Petri boxes give, when 

 sloped, the appearance already noticed in figs. 2 to 7. The pinnate 

 growth is, in these cases, not so striking on the surface as with test- 

 tube cultures, but, as we have previously remarked, the filaments 

 which penetrate the gelatine ran a perfectly parallel course upwards, 

 at an angle of about 45, into the substance of the gelatine ; the same 



