If a plant is positioned horizontally, the elongating region responds by bending upward within 10 to 12 h until it is vertical, forming a goo bend with the stem below. If a Xan thiv~ strumarium L. (cocklebur) plant is placed horizontally, but restricted to that position for 48 h and then released, the bend to the vertical usually takes place within 10 s, suggesting that bending energy is stored in restricted stems . Some plants that do not bend completely to 90 within 10 s do so within 5 min, and other plants can overshoot the 90 mark by as much as sao. Microscopic measurements show that cells on the bottom of stems that have been restricted and then released are longer and narrower than cells on the bottom of restricted stems; cells on the top of restricted-and-released stems are shorter and thicker than those on the top of restricted stems . Thus, stems bend upward rapidly after release in response to changes in cell dimensions, but apparently with conservation of cell volume (i.e., little or no movement of water in or out of cells during the rapid bending ). The increased diameter of the cells on the bottom of restricted plants indicates that the cells are taking up water before they are released {apparently accompanied by an increase in cell wall area), while they are not allowed to increase much in length. Any increase in length was accompanied by stretching of cells on top. Thus, energy for bending was stored in stretched upper cells and compressed lower cells that have taken up water. It was al so shown that graviperception takes place in the very tissue that bends, and this perception is not a perception of the tension and compression caused by the weight of a horizontal stem. Also, amyloplasts were found in a sheath also in the region of bending and were found to settle in the direction of gravity. The location of the sheath between the vascular tissue and the cortex lead to a proposed model of graviperception for green vegetative dicot shoots.