

254 BOTANICAL GAZETTE [October 



means of a refrigerating brine coil under thermostatic control. 

 The freezing point of the water in the tank was slightly 

 lowered by the addition of alcohol in order to avoid accumulation 

 of ice about the brine coil. The circulation of water from and to 

 the tank was maintained by an electrically driven pump. The 

 flow from the tank was led to a horizontal feeder pipe 6 ft. above T 



the experimental table, and in this feeder six outlets were tapped 

 about 30 cm. apart. At the table level a similarly tapped collec- 

 tor pipe returned the water to the tank. The flow for each plant 

 was led from the feeder by one of the six insulated outlets, 

 through rubber tubing to the loop or coil about the plant, 

 through the coil and back to a corresponding tap on the collector 

 pipe. With this apparatus as many plants as there were pairs of 

 outlets (supply and return) could be placed under experimental 

 conditions at the same time. Each supply pipe was fitted with a 

 valve, making it possible independently to regulate or stop entirely 

 the flow of cold water through any of the six coils. By means of 

 these valves temperatures differing by several degrees could be . 



maintained in different lines without affecting appreciably the tem- 

 perature of the general supply. All exposed metal piping was 

 covered with non-conducting material. During the winter months, 

 while the temperature of the city water was so low that it could be 

 used directly, a second similar system was also arranged for use 

 with the city water. Each system was supplied with six con- 

 necting lines, so that twelve experiments could be conducted at 

 the same time. With a little care it was possible to control within 

 about i° C. the temperature to which the cooled region was sub- 

 jected. This temperature apparatus was devised by the junior 

 author, and for its maintenance in proper working order during 

 the experiments he is largely responsible. 



The temperatures found to be effective for the purpose ranged 

 from 2.5 to 6° C . , according to the species of plant used and the 

 region subjected to cooling. In the bean seedling, in which regions 

 of the main stem were cooled, the more basal levels of the stem 

 required a lower temperature than more apical levels to bring 

 about physiological isolation and outgrowth of buds below the 

 cooled zone. 



f 





