Optical tweezers are commonly used for manipulating microscopic particles, with applications in cell manipulation, colloid research, manipulation of micromachines and studies of the properties of light beams. Such tweezers work by the transfer of momentum from a tightly focused laser to the particle, which refracts and scatters the light and distorts the profile of the beam. The forces produced by this process cause the particle to be trapped near the beam focus. Conventional tweezers use gaussian light beams, which cannot trap particles in multiple locations more than a few micrometres apart in the axial direction, because of beam distortion by the particle and subsequent strong divergence from the focal plane. Bessel beams, however, do not diverge and, furthermore, if part of the beam is obstructed or distorted the beam reconstructs itself after a characteristic propagation distance. Here we show how this reconstructive property may be utilized within optical tweezers to trap particles in multiple, spatially separated sample cells with a single beam. Owing to the diffractionless nature of the Bessel beam, secondary trapped particles can reside in a second sample cell far removed ( approximately 3 mm) from the first cell. Such tweezers could be used for the simultaneous study of identically prepared ensembles of colloids and biological matter, and potentially offer enhanced control of 'lab-on-a-chip' and optically driven microstructures.