I 



314 BOTANICAL GAZETTE [april 



Inner and outer walls. — The records show a close similarity 

 in the width of the outer solid tissue of the culms of the control 

 and of the experiment. This is also true for the partitions between 

 the chambers. 



Palisades. — In experiment 1, one control culm developed one 

 layer of palisades and the other two; while in experiment 2 the 

 same thing occurred in the experimental culms. In the third 

 experiment both culms had two layers. Atmospheric pressure 

 seems therefore to have no effect upon palisade tissue. 



Experiment 4 



* 



This experiment extended from April 11 to May 10. The 

 apparatus used was the same as in the previous experiments, 

 except that a tube dipping into a dish of mercury was used to 

 indicate the pressure instead of the regular manometer. Cyperus 

 alternifolius (?) was used in this and the following experiment 

 instead of Scirpus. The time was divided into five periods, dur- 

 ing which an effort was made to keep the pressure at 1/6 

 atm. (i3o=fc=mm.), 1/3 atm. (25o=*=mm.), 1/2 atm. (380=*= mm.), 

 3/4 atm. (570=*= mm.), and 5/6 atm. (630=*= mm.) respectively. 



1 



The pressure varied very greatly, probably because the city water 

 supply varied more at this time than earlier in the year. Culms i , 

 2, 6, 7, and 8 were the control, and nos. 3, 4, 5, and 9 the experi- 

 mental culms. Nos. 1, 2, 3, and 4 started before the experiment 

 and were not used in the comparison. Culms 3 and 4 when 

 placed at 1/6 atm. grew very little and ceased growth in 3-4 days. 

 At the end of a week the pressure was raised to 1/3 atm., but with- 

 out effect upon these culms. Toward the end of this period no. 5 

 started to grow slowly. At the same time no. 6 started in the 

 control. With the rise to 1/2 atm. no. 5 grew better, but still kept 

 behind no. 6. The rates approached each other very closely when 

 the pressure rose to 3/4 atm. During this period no. 7 began to 

 grow in the control, and as it followed very closely the growth curve 

 of no. 6 (which had grown very tall) it was used for comparison 

 with no. 5. At 5/6 atm. no. 7 still maintained a higher rate than 

 no. 5, which by this time had about reached the limit of its growth. 

 About the middle of this period no. 8 started in the control, and no. 9 



