131 



species it is always wise to cany a controlled experimeut with as nearly 

 similar conditions as possible. A success in one case and a failure in the 

 other can then usually be relied on as representing the actual state of 

 affairs. If a plant fails to show any results within the reasonable time it 

 is best not to use it again until a sufficient number of days of gi-ace 

 have passed, as there is always a possibility of a belated infection. 



It will be seen that care in execution and accuracy of observation are 

 the main features in this work, costly apparatus not being required, and 

 it is hoped that this brief. description of the operations may be of service 

 to those interested in this modern method of studying and classifying this 

 group of fungi. 



