A field experiment was conducted with potato in hilly zone (Nilgiris) soils of Tamil Nadu, India to study the response of potato to N, K and Mg fertilization and to evolve the optimum fertilizer schedule for potato. A field experiment was conducted at Nanjanad farm, Nanjanad, Ooty during March to June 2002. The soil of the experimental field comes under Ooty series (Ultic Hapludalf). The soil was sandy clay loam in texture, non saline and acidic (pH 5.0) and CEC (29.87 cmol (p+) kg-1). The fertility status of the soil was high in organic carbon (21.0 g kg-1), medium in available N (360 kg ha-1), high in available P (178 kg ha-1) and K (280 kg ha-1). The exchangeable Ca, S and micronutrients (Fe, Zn, Mn and Cu) were above the critical level, but exchangeable Mg level (0.5 cmol (p+) kg-1) was below the critical level. Potato crop respond well in the soil with pH of 5.2 to 6.5, so in order to decrease the acidity, lime requirement was estimated and lime was applied @ 4.3 t ha-1. The treatment combinations (20) included N, P2O5, K2O and S at 120 : 240 : 120 : 80, 180 : 240 : 180 : 80, 240 : 240 : 240 : 80 kg ha-1 with 0, 6, 12, 24, 48 kg Mg ha-1, recommended fertilizer dose of 120 : 240 : 120 : 60 kg N, P2O5, K2O and MgSO4 ha-1 and an absolute control. Treatments with omission of N, P2O5 and K2O were also tried with higher levels of nutrient combinations. The twenty treatments were replicated thrice in a randomized block design.

The maximum tuber yield of 50.3 t ha-1 was obtained for the highest levels of N, P2O5, K2O and Mg (240:240:240:48 kg ha-1) tried in this study which was significantly higher than the currently recommended levels of N, P2O5, K2O and Mg (120:240:120:6 kgha-1) was only 22.4 t ha-1, which is just 44.6 per cent of the maximum yield obtained. The tuber yield of 5.74 t ha-1 for the absolute control (no fertilizer application) showed the poor nutrient supplying capacity of the soils for the potato crop. Among the nutrients Nitrogen (N) was found to be the most limiting nutrient in respect of tuber yield followed by P, K and Mg. Native N supply (along with application of 240 P, 240 K20 and 48 Mg ha-1) could result in only 42.3 per cent of the yield obtained for the fertilizer N supply of 240 kg ha-1. The higher yield at enhanced N levels could have been mainly due to the better assimilation of carbohydrates and their translocation to tubers which ultimately helped in enlarging the tuber size and increasing the tuber weight. Withholding of P application (and application of 240 N, 240 K2O and 48 Mg ha-1) resulted in 24.9 t ha-1 of tuber yield which was only 49.4 per cent of the yield when 240 P2O5 was applied. The tuber yield reduction (when compared to the maximum yield) due to non-application of K was 36 per cent and the native K supply along with 240 N, 240 P2O5 and 48 Mg ha-1 could yield 32.2 t ha-1 only. At 240N, 240 P2O5 and 240 K2O, non-supply of fertilizer Mg resulted in 82.3 per cent of the maximum yield. Since the experimental soil was deficient in exchangeable Mg potato crop responded significantly to added Mg. On the other hand the crop removal of Mg ranged from 16 to 135 kg-1. Hence there was marked increase in the yield due to Mg addition.

The applied Mg levels of 24 and 48 Kg ha-1 resulted in significantly higher tuber grade and seed grade yield. Seed grade tubers contributed maximum to the tuber yield to the extend of 50 to 68 percent. The proportion of higher grade tubers (tables, seeds) did not vary very much due to Mg application at 240 N : 240 K2O kg ha-1, showing that at these levels of N, K and Mg did not influence the tuber grades. Withholding of N supply (0 N, 240 P2O5, 240 K2O and 48 Mg ha-1) resulted in more of small sized tubers when compared to withholding of P (240N, 0 P, 240 K2O, 48 Mg ha-1) and K (240N, 240 P2O5, 0 K, 48 Mg ha-1). This indicates that N is the key element for getting higher grade tubers. The starch content was significantly influenced by the application N, K and Mg fertilizers. Reducing sugar and crude protein content of potato increased, whereas, total soluble solid content decreased significantly due to enhanced application of N, K and Mg fertilizers. The results clearly indicated that the presently followed state recommended level of 120 kg N: 240 kg P2O5: 120 kg K2O with 60 kg MgSO4 ha-1 is sub optimal. Hence, there is a need for upward revision of the existing blanket recommendation to 240: 240: 240: 48: 80 kg N: P2O5: K2O: Mg: S ha-1 respectively to ensure balanced fertilization and to get maximum tuber yield, grade and quality of potato.