In Flora of North America, volume 25, it is stated that the aerial spikelets of Amphicarpum do not set seed.  Dr. James Quinn has pointed out that this is not correct, at least, not for A. purshii. He provided the following information to Wipff and Barkworth for placement on the Web site, as it was too late to amend the comment in the published volume.

James McNamara and I (Amer. J. Bot. 64: 17-23) documented germination of "aerial seeds" of Amphicarpum purshii in the field and under three laboratory regimes; using the tetrazolium chloride method, caryopses from aerial spikelets showed a range of viability of 76 to 82% among the five populations, and the range of germination response was 0.7 to 42.7%.

Subsequently, Gregory Cheplick and I investigated the "pessimistic strategy" of early reproductive allocation to the larger subterranean propagules and the later allocation to the smaller aerial propagules, and the various causes and significance of shifts in the aerial/subterranean fruit ratio (Oecologia 52: 327-332; 57: 374-379); these and subsequent studies included the number, growth, survival, and reproduction of plants arising from "aerial seeds", and the factors affecting their success and ecological significance.

We found that large plants arising from "subterranean seeds" (as would occur at low densities, or following fire) produced large numbers of "aerial seeds" which contributed significantly to the increase in plant numbers the following year (Oecologia 73: 459-464; J. Ecol. 76: 263-273).

We also investigated the fitness and variation of progeny from aerial panicles (Amer. Midl. Nat. 116: 394-402) and used twelve families raised from seeds collected from the aerial panicles of twelve parents grown in the greenhouse to investigate the quantitative variation of life history traits and its evolutionary significance (Amer. J. Bot. 75: 123-131).

We thank Dr. Quinn for enabling us to present this correction

Mary Barkworth and J.K. Wipff

Received 17 April 2003