Abstract: The experiment was conducted to find out the effect of method of planning (row and haphazard) and five hill arrangements [1 (25x12 cm2), 2 (25x6 cm2), 3 (25x4 cm2), 4 (25x3 cm2) and 5 (25x2.4 cm2)] on the yield of late transplanted aman rice (cv. BR23) grown under different planting dates (1st, 15th and 30th Sept.). Yield components namely number of effective tillers m-2, number of grains and that of total spikelets panicle-1, weight of 1000 grains were notably decreased with the delay in transplanting which in turn resulted in the decreased grain yield. Number of effective tillers m-2, grain yield and biological yield were the highest in hill arrangement 5 (25x2.4 cm2). The grain yield gradually decreased from 1st September transplanting onwards and became the lowest when the crop was transplanted on 30th September. The highest grain yield was obtained with closest hill spacing (hill arrangement 5) on 1st September transplanting. The biological yield was the highest at haphazard method of planting at hill arrangement 5 on 1st September transplanting. The highest harvest index was observed with hill arrangement 1 (25x2.4 cm2) on 1st September transplanting.