As the human genome program gets under way, we examine the progress made since the first human genome meeting in Santa Cruz in 1985. The lessons of the last 5 years demonstrate that progress has been much slower than anticipated. The new technology being developed in 1985 was fluorescent sequencing and multiplexing. These techniques are now established, but they still have to produce a substantial sequence to rival those determined by conventional technology. Inspection of the EMBL and GenBank databases shows few large sequences have been determined and that there is a large discrepancy between what is theoretically possible and what has been achieved so far.—Barrell, B. DNA sequencing: present limitations and prospects for the future.FASEB J.5: 40–45; 1991.