Mark McDonough's reaction to the proposed reorganization of the genus Scilla is similar to mine, though I had not thought about it in such detail. In addition to Allium, one can easily think of many other genera which display differences in, e.g., seed morphology similar to those offered as the basis of genus differentiation in "Scilla nova." Calochortus is a good example for seeds, and for bulb and root form Iris is obvious. There are anomalous bulbs or corms in many genera -- for example, there are a few stoloniferous species in Crocus, Fritillaria, and Lilium. I would like someone who actually does taxonomy based on DNA studies to clarify something for me. As I understand it, these studies are based on certain selected portions, or segments, of a few certain chromosomes -- not on the whole genome of the plant, which would not be feasible with present technology. How does the researcher know that the sequence(s) selected represent the variation in the entire genome in a meaningful and statistically reliable way? I am not a scientist, but linguists employ rather similar statistical analyses to estimate the degree to which various languages are related and the time depth of their divergence from common ancestors. One currently popular method, called mass comparison, is capable of generating extremely dubious results if done on a database that is not, shall we say, perfectly understood by those conducting the comparison. I think that quite a few of us on this forum have studied enough statistics and science to understand a brief answer to these questions, and we probably all have some grasp of genetics, so if someone appropriate has the time, would you please enlighten us? Thanks, Jane McGary Northwestern Oregon