4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 94 



For exposures of the material in the spectrum special boxes lo x 

 4x1 inches were made of brass and provided with parallel center 

 plates of monel metal about 4x1 inches, spaced at 0.4 inch. Two such 

 boxes placed end to end thus more than filled the visible spectrum, and 

 each of the 48 compartments was subjected to a band of radiation 

 ranging in width from the order of 50 A in the low violet to that of 

 200 A in the high red. The spectrum and boxes were provided with 

 a secondary light-proof housing, in which, at an elevation of about 



1 foot, were installed two 20-inch milk-glass lumiline lights wrapped 

 in red cellophane having no appreciable transmission for wave lengths 

 shorter than 6000 A. These lights were so arranged that in conjunc- 

 tion with end-mirrors no shadows were cast in any compartment of 

 the boxes. The intensity of illumination was regulated by a rheostat. 



The focal plane of the spectrum obtained was located by inserting 

 a plate of didymium glass between the condenser lens and the slit. 

 The sharp absorption lines of didymium also provided a convenient 

 means of establishing the wave lengths of all regions of the visible 

 spectrum. Wave lengths in the near ultraviolet region were established 

 by substituting a mercury arc as a light source and using uranium glass 

 to pick up through fluorescence the lines characteristic of mercury. 

 Wave lengths in the infrared region were established by following the 

 absorption characteristics of water vapor with a thermopile. The 

 radiation energies throughout the entire spectrum were established 

 by means of a thermocouple. 



The procedure in each experiment was as follows. Two boxes 

 were placed in the spectrum and half filled with tap water, which 

 served as the medium of germination. Dormant or light-sensitive 

 lettuce seeds were then scattered into the compartments, surface ten- 

 sion bringing about a fairly uniform distribution of the seeds over 

 the available water surface. About 100 seeds could be conveniently 

 accommodated in each compartment. 



After 2| hours presoaking the seeds were given exposures of spec- 

 tral light, of red lumiline light, or of both lights, depending upon the 

 particular objective. The red lumiline light, by suitable modification 

 of the duration or intensity of the exposure, was used for the most 

 part to effect a 50 percent germination of the seeds independent of the 

 spectral light — a feature ordinarily offering some difficulty, but en- 

 tirely feasible with the material at hand, as had been attested by tables 



2 and 3 of Flint's paper (3). Upon this base the nature and extent 

 of any promoting or inhibiting influence of the spectral light was 

 registered as a departure. After 24 hours the boxes were removed. 



