The possibility that merging might largely consist of bursts of subsolar merging repeating every 8 minutes (Flux Transfer Events-FTEs) has been raised. We examine the maximum possible contribution to the cross polar cap potential from FTEs. Assuming each candidate FTE actually transfers flux, we calculate the contribution from: (i) FTEs which fit statistical magnetopause observations; (ii) FTEs which fit typical ionospheric observations; (iii) The largest reported FTEs in case studies at low and high altitude. We find that FTEs account for less than 10% of a normal active time polar cap potential (~100 kV); although some individual reported cases could, with optimistic assumptions, account for 20--33%. Two new limits on the size of FTEs are introduced. The requirement that FTEs form within ¿6 Re of the subsolar point limits FTEs to producing about 38 kV average potential. An even more fundamental limit is imposed by coherence: to represent a merging burst which starts and stops in 2 minutes, an FTE size is limited by the fast mode wave velocity. The coherence limit is about 14.5 kV, meaning that larger pulses consist of independently active merging sites. Simulation results in which bursty merging dominates are reconsidered, and shown to have geometrical assumptions that may be applicable locally, but not globally. A burst of subsolar merging every 8 minutes therefore cannot contribute most of the cross polar cap potential. ¿ American Geophysical Union 1993