NO. 21 CO2 ASSIMILATION AND WAVE LENGTH HOOVER 3 



diameter and 2 inches thick were used in order to illuminate the plants 

 on four sides. The temperature of the filters was controlled by placing 

 them in thermostated water baths. A very simple optical system was 

 used in this case. A looo-watt Mazda lamp was placed near the focus 

 of a concave mirror in order to obtain a magnified image of the 

 source at the plant. The light passed twice through the filter before 

 reaching the plant. Figure i represents the energy distribution ob- 

 tained with these filters under the condition of the experiment. The 

 distribution of energy in the bands was not entirely satisfactory, owing 



Fig. I. — Energy transmission curves for 6-inch Christiansen filters. 



to imperfections in the concave mirrors and to the fact that the filters 

 were too thick to obtain good temperature control. 



In the second experiment with the Christiansen filters sunlight 

 was used as the source of radiation. Sunlight was obtained for the 

 experiment by using a large coelostat located on the south side of the 

 Smithsonian building. The moving mirror of the coelostat was 30 

 inches in diameter. Light reflected from this mirror was reflected by 

 a second mirror to a third mirror located at one end of a long room. 

 This mirror was about 60 feet from the plants. This combination of 

 mirrors illuminated an area at the plant chamber about 15 inches in 

 diameter. A Christiansen filter 12 x 14 inches and i inch thick was 

 made in order to use most of the energy in this beam. The filter was 

 placed 50 feet from the plant chamber. The distribution of energy in 



