2.1. Location The experiment was carried out at the Bangladesh Agricultural University (BAU) Germplasm Centre, Department of Horticulture, Mymensingh, Bangladesh which is located at 240260 latitudes and 900150 longitudes with an altitude of 8.3 m above sea level. Biochemical analyses were carried out in the Department of Biochemistry of Bangladesh Agricultural University (BAU), Mymensingh, Bangladesh. 2.2. Soil and Climate The soil at the Germplasm Centre is sandy loam which belongs to the Old Brahmaputra Flood Plain Alluvial Tract. The selected soil samples of the experimental area were analyzed at the Laboratory of the Soil Science Division, Bangladesh Agricultural Research Institute, Joydebpur, Gazipur. Soil pH was 6.5 and 6.3 at a depth of 0–15 cm and 15–30 cm, respectively. The total %N, available P, and available K were 0.056, 38 µg/mL and 0.10 meq/100 mL at a soil depth of 0–15 cm, respectively, and 0.036, 13 µg/mL and 0.07 meq/100 mL at a soil depth of 15–30 cm, respectively. The experimental site enjoys a sub-tropical climate characterized by heavy rainfall (means of 361 mm/month and 2166 mm total), high temperature (mean 28.34 0C), a mean of 161 h/month of sun during April to September, and scant rainfall (mean 79 mm/month and 555 mm total), low temperature (22.6 0C) and a mean of 187 h/month of sun during October to March (Source: Weather Yard, Department of Irrigation and Water Management, BAU, Mymensingh, Bangladesh, 2005). 2.3. Experimental Material The cultivar used in the study, Amrapali (BARI Mango-3), is precocious, medium dwarf, annual, and prolific in bearing and has good fruit quality. The age of the plants was 11 years at the initiation of the experiment. Plant spacing was 5 m × 5 m. 2.4. Experimental Design The experiment was laid out in a Randomized Complete Block Design (RCBD) with 3 replications. The treatments were: (1) the best (BT) individual treatments of fertilizer (37.5 kg cow dung with 518.43 g N, 120 g P, 187.5 g K, 67.45 g S, and 8.09 g Zn per plant) applied in three installments, a 4% KNO3 spray application on 15 November 2007 approximately 1.5 to 2 months prior to anticipated flowering, irrigation on 15 October and 15 November 2007 and then continued from fruit set (6 March 2008) to maturity at 7 day intervals, soil drench application of paclobutrazol at 7.5 g/L of active ingredient on 15 July 2007 (all of the best treatments from prior studies); (2) the 2nd or next best (NB) individual treatments of fertilizer (same as above) applied in two installments, a 4% urea spray application on 15 November 2007 approximately 1.5 to 2 months prior to anticipated flowering, irrigation on 15 October and 15 November 2007 and then continued from fruit set to maturity at 14 day intervals, and a soil drench application of paclobutrazol at 10 g/L of active ingredient on 15 July 2007; and (3) untreated control. 2.5. Application of Fertilizer and Paclobutrazol Fertilizer was applied in two or three applications. The first application was the total amount of cow dung, triple superphosphate (TSP), gypsum, and zinc sulfate applied on 15 September. If in two applications, 1 2 of the urea and muriate of potash (MoP) were applied on 15 September and the remainder was applied on 15 March. If in three applications, 1 4 of the total amounts of urea and MoP were applied on 15 March and again at the 3rd application on 15 May. Very light irrigation was provided at each time of fertilizer application. Paclobutrazol (Syngenta, India) was prepared from a 25% (w/v) stock solution for final concentrations of 7.5 and 10 g/L of active ingredient. Treatments were applied as a soil drench in which 10 small holes (10–15 cm depth) were prepared in the soil around the collar region of the trees just inside the fertilizer ring. One liter of the prepared solutions was poured into the holes of each tree, and the soil was reworked after application. Only water was applied to the control plants. 2.6. Tree Growth and Yield Measurements The length and number of leaves of 10 randomly selected terminal shoots at flowering on 20 December 2007 were measured. The individual leaf area of 50 leaves was measured, taking 5 from each of the 10 above-selected shoots, using a leaf area meter and was expressed as cm2. Starting at the first appearance of a panicle, the number of panicles per plant was counted at 10-day intervals up to the completion of panicle emergence. Ten panicles were randomly selected from each treatment on 18 February 2008, and the length and number of secondary branches per panicle was recorded. The initial number of fruit of each panicle, and the fruit retained per panicle were counted at 10 day intervals starting from the pea stage up to harvest starting from 6 March 2008. The average date of harvest (when one or two ripe fruit dropped from the tree, all the fruit were harvested at one time), number of fruit/tree, fruit weight, and total yield/tree was recorded. 2.7. Fruit Quality Measurements After harvest, 10 randomly selected fruit per tree were ripened at room temperature, and fruit quality was determined. Each fruit was peeled, the stone removed, and each was weighed along with the remaining pulp. The edible (pulp) portion, stone-to-pulp ratio, peel-to-pulp ratio, shelf-life (the difference between the harvest date and the date up to which the fruit remained edible was considered as shelf-life), total soluble solids (TSS), titratable acidity, vitamin C, dry matter, and reducing, non-reducing, and total sugar content. TSS of the pulp was measured using a hand refractometer. The titratable acidity of mango pulp was determined [25] using 0.1 N NaOH solution. Vitamin C content was determined. Reducing sugar content of mango pulp was determined by the dinitrosalicylic acid method. The total sugar content of mango pulp was colorimetrically determined by the anthrone method. 2.8. Statistical Analysis Treatment means were analyzed by analysis of variance and separated by Fisher’s Least Significant Difference (LSD) test at p = 0.05.