There have been excellent, long-term results in identifying high yielding sugarcane (Saccharum spp.) varieties for organic (muck) soils in Florida. However, newer varieties for sand soils in Florida have not had yield improvements during a recent 33-year period. The purpose of this study was to compare the performance of 31 sugarcane genotypes and one erianthus genotype on sand soils with and without mill mud added at the rate of 1510 cubic meters per ha. A by-product from the sugarcane milling process, mill mud is an organic waste produced from the clarification of sugarcane juice, and it is rich in N, P, K, Ca, and Mg. The experiment was planted with soil treatment as main plots and genotypes as subplots on a Margate sand soil in 2007. Commercial recoverable sucrose (CRS) (g sucrose per kg cane), cane yield (Mg per ha), and sucrose yield (Mg per ha) were determined in the plant-cane through second ratoon crop cycles during the next 3 yr.  The mean CRS on sand soil with added mill mud was much lower  than that without mill mud (111 vs. 127 g sucrose per kg cane). However, yields of cane and sucrose per hectare were much higher as a result of the added mill mud (150 vs. 80 Mg cane per ha, and 17 vs. 10 Mg sucrose per ha). Yields of CRS, cane, and sucrose of three, nine, and nine genotypes, respectively, were differentially affected by whether or not mill mud was added to the sand (P = 0.05).  CP 01-2390 was the most adapted sand genotype; it had a larger increase in CRS on sand than most genotypes and its TCH yields on sand with and without mill mud were 147 and 141 Mg per ha, respectively. The Erianthus genotype was also well adapted to sand with tonnage yields of 84 Mg per ha with mill mud and 75 Mg per ha on sand without mill mud. Except for CP 01-2390, no other sugarcane genotype was well adapted to sand without mill mud. Six genotypes—CP 00-1748, ‘CP 00-2180', CPCL 00-6756, CPCL 01-0571. ‘CP 72-1210', ‘CP 72-2086', and ‘CL 90-4725'--were well adapted to sand with mill mud. Of these, CP 00-2180 had been preselected as a genotype that would yield well on sand, and CP 00-1748, CP 72-1210, and CL 90-4725 were expected to yield well in both soil treatments. It was expected that CP 01-2390 would yield well on sand without mill mud and that CP 72-2086 would have poor yields on sand without mill mud. Relative yields by soil type were not as predicted for 6 of the 32 genotypes suggesting that multiple locations are needed for identifying successful commercial sugarcane cultivars on sand soils.