8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 92 



time interval to note the resulting growth curvature. If the light 

 adjustment was very much out of balance as indicated by the plant, 

 a bending similar to that shown in plate 2 occurred in 20 to 30 min- 

 utes. An adjustment was then made in the proper direction and the 

 used seedling discarded for a new one. As the balance point was 

 approached the exposure time necessarily increased. Finally on mov- 

 ing the standard light back and forth through a distance of i cm, the 

 plants could be made to curve repeatedly toward one light then toward 

 the other. The balance point was taken to be the midpoint between 

 these two positions. Care was always used not to expose the fresh 

 seedlings to any light but red in the preliminary handling. Priestley 

 (1926) has shown that light afifects normal and etiolated shoots very 

 diiferently. The amount of light required to induce phototropic curva- 

 ture in normal light-grown shoots is greater, and must be continued 

 longer, than that required to bring a similar curvature in etiolated 

 shoots. 



After a balance point had been determined and tested by using 

 several seedlings, a specially constructed thermocouple was inserted 

 into the glass cylinder occupied by the seedlings and the light intensi- 

 ties measured at the balance position. The junction of the thermo- 

 couple was made of a short length of fine bismuth wire and one of 

 bismuth-tin alloy, each about 25 microns in diameter. The alloy was 

 made up of 95 percent bismuth and 5 percent tin. Utmost care was 

 needed in measuring the light intensities since the plants were found 

 to be much more sensitive to the light than the best physical instru- 

 ments available. It should be remembered, however, that the seedling 

 integrates the effect of radiation over a relatively long period, while 

 the thermocouple responds in a few seconds. 



The results of this experiment are presented in table 2. The ratio 

 of the intensity of the monochromator light to that of the standard 

 light is given in the third column for corresponding wave-length 

 ranges shown in the first column. Where filters were used in combina- 

 tion with the monochromator they are indicated in the second column. 

 No phototropic responses were obtained in any of the first six wave- 

 length ranges. The first quantitative measurements that could be 

 made were for the range 5040 to 5160 A. In the last column of the 

 table the relative phototropic effectiveness of the different wave- 

 length ranges is given. The ratio 29.10 was arbitrarily taken as 

 unity. 



With unilateral illumination through the monochromator and a 

 number 'j'j Wratten filter in the region 5430 to 5670 A, bending oc- 



