270 Journal of Agricultural Research voi. xiii, No. s 



the leaf, may be explained by the tendency of the rain water to run down 

 and accumulate in the axils of the leaves, washing the spores with it. 



The appearance and location of the newly infected spots on the leaves 

 led to the belief that the dashing rains in the field and the use of hose for 

 watering or syringing in the greenhouses were accountable for the spread 

 of the disease from plant to plant. The following experiment was per- 

 formed to determine the relation of the splashing of water from plant 

 to plant to the dissemination of the fungus. 



Two plots of lettuce in a bench were separated by a wooden partition 

 covered with tar paper, and the rows parallel to and nearest the walk 

 were inoculated with M. panattoniana in wounds made with a sterile 



@(o)(o)(o)@(o)(§)(o)(o)(o)(o)(o) 



(S)(§)(D(«)(§)®(»)(§)(§)(5)(§)(§) 

 ®(5)(«)(«)(§)(§)(5)(§)®(§)(§)^ 



®($)(§)®'^(§)(S)(S)(§)(§)(§)(g) 



(§)(D(DO®(§)(§)(§)(§)0(§)(§) 

 000(§:'«'(5)0(§)000(§) 

 0000(§)0(§)(§)000(§) 



@@@(Q)@(0)(0)(0)(Q)(0)(0)(§J 



(§)(D00®®(§)0®000 



oooooo®ooooo 

 ®oooo 0000000 



(§)00000000000 



000000000000 

 000000000000 



/^ZOT- ^ 



Pig. 2. — Diagrams representing results (plot i) of watering with the hose and (plot 2) by subirrigation. 



Spatula. Plot I was watered by spraying with the hose. In plot 2 a 

 trench was dug between each row, transverse with the length of the bench, 

 and covered with a long flat board. A notch was cut in each of these 

 boards at the end nearest the walk, and the plants were watered by 

 inserting the hose through this notch and filling the trench with water. 

 The surface of the ground and the plants above ground were thus kept 

 dry during the watering process. 



In the plot which was watered by spraying with a hose 55 of the 72 

 plants were infected in one week. In the plot watered by subirrigation 

 only 9 plants out of the total 72 had been infected in the same period. 

 The results of this experiment are shown very clearly in the accompany- 

 ing diagrams (fig. 2). The objection may be raised that it was the wet- 



