400 MINNESOTA BOTANICAL STUDIES 



results are strikingly similar to those above recorded and need not 

 be given. 



Experiments with Sunflowers 



While the foUovi^ing experiments are only of a preliminary na- 

 ture, still they throw light upon the rust-transpiration problem and 

 may be briefly noted. 



Three sunflowers of the same age, height, and approximately 

 similar leaf area were selected from stock material in 4-inch pots 

 filled with ordinary potting soil. One of these plants was inoculated 

 on November 14 with the teleutospores of Pnccinia Hcliaiithi Schw. 

 found on a wild sunflower, probably HeliantJius annus. 



On November 29 and when only two rust areas had appeared, 

 the plants were well watered and then placed in Ganong's aluminum 

 shells. These were covered with sheet rubber and the w^ax seal in 

 such a manner as to prevent loss of water except by transpiration. 

 They were then weighed on a new "Cenco" trip balance sensitive 

 to 0.1 g. and the transpiration losses determined for five intervals 

 during the following 16 days. 



From time to time during the experiment the plants were taken 

 from the aluminum shells for aeration, and at these times the water 

 lost by transpiration was replaced. At the end of the experiment 

 when the transpiring area (exclusive of the stems) was calculated, 

 the plants were 27 to 30 cm. tall, and they had from eight to ten 

 leaves each. The diseased plant had rust on the five lower leaves. 

 By cutting out the pustular areas and dividing the weight of these 

 by that of the total leaf weight, the rusted area was approximated 

 to be 17 per cent. 



The results showed that during the first interval of the experi- 

 ment the diseased plant lost 63 per cent more water than the healthy 

 plants. This excess transpiration over the normal gradually dimin- 

 ished however, to 44, 36, 22, and 29 per cent respectively, at the suc- 

 cessive intervals between determinations. 



A second experiment was started on November 17, and when the 

 rusted areas of a similarly diseased sunflow^er were just beginning 

 to appear. The experiment w-as carried on for a period of eight 

 days and exactly like the preceding, only 3-inch pots were used. 

 The rusted area w^as about 15 per cent. 



Here again, the diseased plant showed at first a decidedly in- 



