Two field experiments were carried out at two locations, i.e Shandaweel Agricultural Research Station (Souhag Governorate) and El-Mattana Agricultural Research Station (Qena Governorate), during the two successive seasons 1995/1996 and 1996/1997 to study the effect of three nitrogen fertilizer levels (180, 230 and 280 kg N/fed), three potassium fertilizer levels (0, 48 and 96 kg K20/fed) and their interactions on four sugar cane varieties, i.e G.T.54/9, G.85/37, G.84/47 and F.153 on yield and quality of sugar cane. A split plot design with four replications was used. Sugar cane was planted during the 1⁵' week of April and harvested at 12-month age. The important results could be summarized as follows: Applying 280 kg N/fed significantly produced the highest number of millable cane/fed at Shandaweel location. Meanwhile, applying 230 kg N/fed caused the same result at EL-Mattana location. Application of 48 kg K₂0/fed attained a significant increase in the values of millable cane at shandaweel location. - The combination between G.85-37 variety x 230 kg N/fed x 48 kg K20/fed significantly produced the highest value of cane yield (70.730 tons/fed) at Shandaweel. Also , the same variety with 180 kg Nifed and 96 kg K₂0/fed produced the highest cane yield at EL­Mattana (54.890 tons/fed)- The combination between nitrogen and potassium fertilizers was the most important combination that significantly affected sugar recovery percentage in the tow locations. The highest values of this trait were detected from adding 280 kg N/fed and 48 kg K20/fed at Shandaweel (14.1 %) and from adding 180 kg N /fed without potassium application (12.4 %) at EL-Mattana location.- The interaction between G.85-37 variety and 48 kg K,O/fcd at Shandaweel and EL-Mattana significantly produced the highest value of sugar yield (9.229 and 7.867 tons/fed, respectively).- Growing sugar cane variety G.85-37 under 230 kg N /fed and 98 kg K20/fed produced the highest sugar yield at Shandaweel and EL-Mattana locations (9.575 and 8.147 tons/fed, respectively).