197 



For this experiment nine plots, each one-hundredth of an acre 

 in size, were used and treated as follows : 



On Plots 1 to 3 the grass was left undisturbed. 



On Plots 4 to 6 the grass was burned off. 



On Plots 7 to 9 the grass was burned and the ground dug 

 up. 



Each plot in each set was then sown broadcast with seed of 

 the above tree at the rate, respectively, of 10, 20, and 30 lbs. of 

 seed per acre. In actuality this amounted for Plots 1, 4, and 7 

 to 1.6 ounces or 838 seeds which could be expected to grow; 

 for Plots 2, 5, and 8 to 3.2 ounces or 1677 seeds ; and for Plots 

 3, 6, and 9 to 4.8 ounces or 2515 viable seeds. 



On September 16, 1921, an examination of the plots was made 

 to determine what had happened by the end of the first half year 

 period. The six months intervening had been comparatively dry 

 and on the plots from which the grass had been removed the 

 ground was dry and had cracked open in places. The ground 

 on the three untreated plots was wet on account of the heavy 

 ntat of Hilo grass which retained the moisture. 



A count of the haole koa seedlings on the nine plots gave the 

 following results : 



The number of seeds that germinated did not amount to what 

 had been expected, but it is possible that further germination of 

 seed, still lying dormant in the ground, will take place, and that 

 a larger number of seedlings will show up in the next count. 



The results of the examination show that seeds sown in the 

 tall grass have no chance for germination, and that seed, sown 

 on the plots on which the grass had merely been burned, ger- 

 minated almost as well as on the plots where the grass had been 

 burned and the ground dug up. 



The largest seedlin,8f on the former set of plots was 5 inches 

 high with a tap root 10 inches long. On the second set of plots 

 the tallest was 14 inches high with a tap root 10 inches long. 



