140 

 spring, and tolerated drought less than the Pensacola bahiagrass. The 

 1978 season had an unusually dry spring and fall. During both years, 

 the first harvest in May consisted of 50 to 80% broadleaf weeds at 

 Green Acres. Subsequent harvests were dominantly bermudagrass . 



Neither S nor N had any effect on dry matter production during 

 1978, although there were significant differences in the S and N con- 

 centrations in the plant tissues (Table 31) . These differences were 

 also obvious in decreasing N:S ratios with increasing S rates. 



Harvest data from 1979 showed that higher N rates increased yield 

 by 19% (Table 32, Fig. 14). Sulfur fertilization produced a signifi- 

 cant difference in total dry matter yield for the season. The 20 and 

 40 kg/ha S rate increased yields by 10% over the check plot at high N 

 rates. No significant differences were observed between S treatments 

 at the low N rate. 



The previous experiments as well as research by others (Ensminger, 

 1958; Neller, 1959; Anderson and Futral, 1966; Lund and Murdock, 1978; 

 Mitchell and Gallaher, 1980) have pointed out the significance of sub- 

 soil S on crop nutrition in Ultisols. Extractable and total S with 

 depth in selected plots in this experiment show that extractable S in 

 the surface horizons are comparable to that in the Myakka soil while 

 t ~tal S is generally less (Table 33) . Extractable and total S in- 

 creases sharply at the argillic horizon. Because of the sandv, well- 

 drained nature of this soil, one would expect roots of the bermudagrass 

 to reach this large source of available S. 



In the process of sampling selected plots for subsoil material, 

 the author noted considerable variation in depth to the argillic hori- 

 zon. The depth to the argillic horizon was measured in each plot to 



