60 



20 



24 



40 



13 1/3 



12 



40 



13 1/3 



12 



40 



13 1/3 



12 



Ocracoke Island on dune sand, and the other was situated on Hatteras Island 

 on hydraulic fill covered by a thin layer of windblown sand. Stands of 

 this grass planted in February 1965 were utilized. The experimental design 

 was a randomized block, 3 replications, with each replicate containing four 

 20-foot rows for each date of harvest. Each plot was harvested once and only 

 once so that the growth harvested represented the total growth to this date. 

 Harvests were made at intervals of about 2 weeks throughout the season be- 

 ginning approximately 30 days after the initiation of growth in the spring. 

 The entire harvest sample was dried, weighed, chopped, and subsampled for 

 chemical analysis. 



The fertilizer program was designed to ma i nta i n fully adequate supplies 

 of nitrogen, phosphorus and potassium throughout the season and was as 

 fol lows: 



_N PoOr K,0 



March 26 - Apri I 3 

 May 10-11 

 June 23 

 August 5-1 I 



Total 180 60 60 



All values given are lb/acre 



Data from these experiments are presented in Figures 6 through 8 and 

 Tables 10 and II. They present a fairly complete picture for this one 

 growing season. 



Obviously, it would be preferable to have data of this kind for several 

 years before attempting to draw any hard and fast conclusions. However, 

 since this is the first study of this sort that we are aware of, it seems 

 worthwhile to examine these data and draw such inferences from them as can 

 be justified. 



a. Growth 



On both Ocracoke and Hatteras the 1965 growing season was rather 

 dry during April and May, quite wet in June, July and early August, and was 

 again dry in September and October. This kind of rainfall pattern might be 

 expected to minimize both leaching and growth in the early part of the season 

 and perhaps promote more. than normal growth during the midsummer period. 

 Examination of the data in Figure 6 indicates that the growth rate was rather 

 slow during the first 30 to 40 days of the growing season, but formed very 

 nearly a straight line from that point on until very near the onset of cold 

 weather. 



These data certainly show a more rapid rate of growth during a wet mid- 

 summer than could be expected during an extremely dry summer. On the other 



18 



