Abstract
Summary: Generic delimitation in the Compositae is notoriously difficult, because a great many of the recognizable groups are connected by palpable intermediates. Neither lumping nor splitting leads consistently to well‐defined, conceptually useful genera, and the fruits of a determined pursuit of absolute monophylesis are equally poisonous. In order to have conceptually useful genera, we must accept the fact that many of them are inherently ill‐defined and unresponsive to efforts at precision.The number of recognized genera of Compositae has increased from 90 in 1753 to about 1200 today. By 1838 the number had reached 889, and was still rapidly rising. In 1873 Bentham shook the number down to 766. He submerged most of the genera that had been proposed half a century earlier by Cassini, but even so he continued to accept nearly a hundred of them, and some others have subsequently been resuscitated. Much of the increase since Bentham's time reflects new discoveries, but there has also been some splitting and refinement of Bentham's genera. Recent proposals to divide such large genera as Eupatorium and Senecio into hordes of minigenera have provoked a vigorous counter‐reaction, and the issue remains in doubt.