One of the causes for declined sugarcane yields is low soil nutrient levels, especially the most limiting ones, nitrogen (N) and phosphorus (P). This is further aggravated by inadequate or lack of nutrient replenishment. The study conducted at KESREF, Kibos, experimental field on a Eutric Vertisol soil evaluated the effect of four P sources, single superphosphate (SSP), triple superphosphate (TSP), diammonium phosphate (DAP), and rock phosphate (RP), and four P levels (0, 17, 34, and 52 kg P ha–1) on yield and quality of sugarcane varieties KEN 82-808 and CO 421. P sources significantly influenced yield of variety CO 421 at second ratoon harvest, the trend being DAP > RP > TSP > SSP. Effect of P sources on quality was not significant. Application of P increased population of millable stalk and yield compared with the control (0 kg P ha–1) in both KEN 82-808 and CO 421. Highest yield was recorded when P applied was 34 kg P ha–1 and lowest in control (0 kg P ha–1), the trend being 34 > 17 > 52 > 0 kg P ha–1. Effect of P rates on quality was not significant. It is concluded that fertilizer P sources can be applied to supply P. P plays a significant role on yield parameters than on quality parameters. The level 34 kg P ha–1 (80 kg P2O5 ha–1) is appropriate to maintain the crop to second ratoon harvest for increasing the yield.