10 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 82 



and 12 inches high, lay on a tahle (E) hy the window at the south 

 end of the room. A wire frame- work, supporting a circular lawn 

 sprayer (P), was suspended from the ceiling directly over the hox ; 

 and a garden hose (G), running to a water spigot, was connected with 

 the sprayer. Two large bath towels (H), put around the frame-work 

 above the sprayer reached half-way to the bottom of the box, which 

 was kept two-thirds full of water. A hygrothermograph (/) lay on 

 another table at the north side of the room, and near this table there 

 was a centigrade thermometer (/) the scale of which was marked 

 in fifths of degrees. Before starting an experiment in this dark- 

 room, water was turned on at the spigot, passed upward and outward 

 from the spray, struck the bath towels, ran down them into the 

 box, and finally passed through a pipe (K) to the outside of the room. 

 In order to have good ventilation and a more humid atmosphere, air 

 from an electric fan (L) was directed against the wet towels and water 

 in the box. If the temperature and relative humidity could not have 

 been satisfactorily controlled for the tests planned, the original idea 

 was to use ice and water in the box for cooling the air, and a stove 

 for heating it. Since small variations in temperature and relative 

 humidity did not seem to affect the resix)nses of the bean beetle, the 

 ice and stove were not used. 



Experience showed that the temperature and relative humidity inside 

 the dark-room were influenced only slightly by outside climatic con- 

 ditions. During the forenoons the temperature was often held nearly 

 constant, and never varied more than two degrees centigrade, but 

 during the afternoons the variation was usually larger. Table i 

 shows that the greatest variation of temperature, between 9:00 

 o'clock a.m. and 4:30 o'clock p.m., was 4.2° C. and the smallest 

 variation was 1.8° C. When the outside relative humidity was ex- 

 ceedingly high, water was not run through the sprayer, because it 

 made the inside humidity too high. 



The testing apparatus consisted of a tall box (fig. 2, M and fig. 3), 

 a shallow water screen (N), and a lamp (O). The inside dimensions 

 of the box (fig. 3, A), called a photo-geotactic box, were 12 inches by 

 5.5 inches by 1.75 inches. It had two sets of shutters, one of glass (a) 

 and the other of wood (b), which fitted snugly in slots. The four 

 sides inside the box were covered with a dead-black cloth, while one 

 of the glass shutters was covered with cheesecloth. The inside of the 

 box was marked with white lines into ten equal sections, numbers I 

 to 10. When all four shutters were in place it was almost totally 

 dark in the box. 



