118 ANNUAL REPORT SMITHSONIAN INSTITUTION, 19 30 



As artificial light, because of its possibility of rigid control, is used 

 to replace sunlight in most of these experiments, the problem of con- 

 struction of such special sources becomes a considerable part of the 

 undertaking. Because of his extensive experience in the commercial 

 construction of Neon lamps, as well as his unusual skill in glass bloAv- 

 ing, Mr. Clark was added to the staff. By October 1, his glass- 

 bl'OAving laboratory was sufficiently complete to permit the commence- 

 ment of construction of high vacuum systems to be used in the evacu- 

 ation of special lamps and thermocouples. The constructive develop- 

 ment of these sources will, undoubtedl}^, be a matter of experimenta- 

 tion which will extend over several years. 



PPIOTOTROPIC EXPERIMENTS 



As it was early realized that the construction of special growth 

 chambers would be an undertaking extending probably to two years, 

 and undoubtedly requiring more funds than were immediately avail- 

 able, another experiment dealing with the bending of plants in the 

 presence of light was undertaken. This experiment could be carried 

 out through the preliminary stages with equipment immediately at 

 hand. A long constant-temperature room, constructed out of a por- 

 tion of room No. 18, as has been mentioned, was ideally suited to this 

 purpose. For this phototropic experiment a special photometer box 

 was constructed. Two beams of light, originating at the extremities 

 of the box, played from opposite sides on a central plant. By the in- 

 troduction of water-cooled filters, these beams were restricted to a 

 narrow range of visible radiation, and provision was made for four 

 possible colors of light. 



Even before the completion of the laboratory, actual observations 

 on the phototropic influence of different colors were being made. 

 It was found immediately that beams of equal intensity but different 

 color affected the bending of the plant in markedly different degree. 

 A thermocouple was provided so that by the galvanometer deflection 

 an accurate determination could be made of the relative intensities 

 of the two beams. By adjusting the beam whose color had the 

 lesser influence, its intensity could be increased until a balance was 

 secured against the weaker, but more potent radiation, so that the 

 plant would grow vertically without bending. 



Early in the experiment it was realized that the most important 

 disturbing factor was the presence of small temperature fluctuations 

 in the air surrounding the oat sprout whose bending was being 

 observed. After many failures to remove this source of disturbance, 

 which vitiated many of the early observations, a scheme was devised 

 which overcame the difliculty. A double-walled cylindrical glass 

 shield was placed around the s]:)rout, symmetrically, and maintained 

 in continual rotation. Thus, if one side were unevenly heated, its 



