Papers on biology and agriculture 83 



University of Illinois, Urbana, Illinois, between Febru- 

 ary sixth and April tenth, 1920. Only a few measure- 

 ments were taken out of doors because it was necessary 

 to finish the work before the end of the college year. 

 This is regretted by the authors because they believe im- 

 plicitly that ecological work should be done in the field. 

 However, since the objective was comparative and not 

 quantitative results it is believed that the data in this 

 case are as valuable as though obtained in the field, foi 

 the differences between greenhouse and field environmen- 

 tal conditions should affect all plants used comparably. 



materials and methods 



The corn used in this experiment was of the variety 

 known as Seed's Yellow Dent, the seed being purchased 

 from Vaughan's Seed Store, Chicago, Illinois. The 

 weeds used were Polygonum pennsylvanicum, Sida spi- 

 nosa and Setaria glauca, the seed of which was collected 

 from cornfields near Urbana, and Ambrosia artemisii- 

 folia and Abutilon Theophrasti, the seed of which was 

 kindly supplied by the Agricultural Experiment Station 

 of the University of Illinois. 



These five kinds of weeds were grown in the green- 

 house side by side with corn in such a way as to insure 

 practically identical environmental conditions for all. 

 The well known cobalt chloride paper of Stahl (18) as 

 described by Livingston (12) and more recently improved 

 by Livingston and Shreve (13) was used for making 

 the tests, care being taken to use the same leaves each 

 day, or, if not the same, those having the same relative 

 position on the stem. This latter precaution was taken 

 because of the fact pointed out by Bakke and Livingston 

 (3) that leaves of different ages and occupying different 

 positions on the stem may vary markedly in their trans- 

 piring power. The lateral leakage of moisture which 

 undoubtedly takes place, as stated by Shive and Martin, 

 when tripartite cobalt paper slips such as used by Liv- 

 ingston and Shreve are employed may be disregarded 

 in this experiment since any such leakage that occurred 

 must have affected readings from all plants used to the 

 same extent. 



