Although MBE technology is over a quarter-century old, and has been outstandingly successful in the growth of semiconductor heterostructures, it has a large reserve of as-yet unexplored capabilities left, many of which are likely to play a role in the future evolution of MBE. Developments that can be anticipated are the additions of OMVPE techniques to MBE, for example, for gas etching and surface cleanup. A central problem will be finding MBE-compatible ways to achieve lateral pattern control down to the nanometer scale. Nanoimprint techniques are a good candidate for that. Self-assembled quantum dots will probably give way to lithographically defined quantum dots with much better control over size and placement. Heterostructures of materials other than semiconductors will be increasingly explored, like magnetic and superconducting structures, and may be even organics. (C) 2002 Elsevier Science B.V. All rights reserved.