The Action of Gravity upon Bacterium Zopfii. 311 



glass dishes gelatine to the depth of 1 inch or more, that at the 

 lower portion of the streak of inoculation the rami which penetrated 

 the gelatine were very distinctly more horizontal than those higher 

 up. This phenomenon is also very conspicuously marked in the case 

 of centrifugalised tubes. Further, our great difficulty in connexion 

 with centrifugalising gelatine is to keep it in position, for in a large 

 number of our experiments the gelatine is forced to the bottom of 

 the tube, and the cotton plug is often driven some distance into the 

 tube. We can see, further, that the compression to which the 

 gelatine is subjected increases towards the distal end of the tube. 

 Towards the distal end, however, the rami tend to become more hori- 

 zontal, so it seems very much as if the tension of the gelatine caused 

 the increased deviation to the horizontal. Having gone so far, we 

 may ask why the rami are not orthogeotropic. We have many times 

 called attention to the fact that the delicate filaments tend always to 

 dip under the gelatine, and it may possibly be that it is the resist- 

 ance of the gelatine which causes the deviation from the vertical to 

 the curious angle of about 45. We may here mention that this fact 

 of the tendency to penetrate the gelatine is in apparent contradiction 

 to the fact, that, as our experiments showed, C0 2 hinders and O 

 favours symmetrical growth. Now, as a matter of fact, very little 

 growth is obtained from a stab culture, so, therefore, it appears neces- 

 sary that some portion of the growth should be close to or freely in 

 contact with the air, and this obtains in streak cultures. To return to 

 the action of the resistance of the gelatine, we have called attention 

 to the fact that, next to the pinnate form of growth, the most striking 

 feature in the life of Bacterium Zopfii is the tendency to curl. Kurth 

 attributes this fact to the resistance of the gelatine. Now the 

 twisting and the geotropic position of the rami we find tend to go 

 hand in hand ; they both tend to disappear upon agar or hard 

 gelatine. Zopf noticed the absence of twisted forms in fluid. We 

 find, as has been above stated, that the majority of the twists are in 

 a direction contrary to the hands of a clock ; many of the photo- 

 graphs show this, and these were taken at a time before we had 

 directed our attention to the nature of the twist. Amongst the 

 higher plants the inward twist is the commoner, and various theories 

 have been made to account for the twisting. If we examine the rami 

 which penetrate the gelatine we note that if they tend to be curved 

 they tend to be convex on the lower surface, and that thus we tend to 

 have two surfaces, a dorsal and a ventral, and can imagine with the 

 botanist a difference in the tension of the protoplasm on these sides of 

 the filament. According to De Vries, whose paper from the Wurtz- 

 burg Institute we have not been able to obtain, geotropism is not an 

 improbable factor in the production of coils (Sachs). Wortmann* also 

 * " Theorie des Windens." ' 33ot. Zeitung,' 1886. 



