FRB 121102 Burst Pairs

The repeating FRB 121102 emitted a pair of discrete bursts separated by 37 ms and another pair, 131 days later, separated by 34 ms, during observations that detected FRB at a mean rate of $2 \times 10^{-4}$/s. Here I assume that these events are separate bursts rather than multiple peaks from longer single bursts and consider their implications. They then are inconsistent with Poissonian statistics. The measured intervals constrain any possible periodic modulation underlying the highly episodic emission. If more such short intervals are measured a period may be determined or periodicity may be excluded. Sources of narrow wandering beams predict pairs (or higher multiples) of bursts closely spaced in time compared to the mean event rate.