Rosaceae

Abstract

Woody or herbaceous. Leaves usually alternate, sometimes distichous, rarely opposite, simple or compound; stipules on the twig or on the base of the petiole, free or adnate to the petiole, rarely O. Inflorescences various, usually terminal, usually (compound) racemes. Flowers actinomorphic, mostly (4)5-merous, mostly bisexual, rarely unisexual and then the plants monoecious or dioecious; hypanthium usually well-developed (not evident in some staminate flowers), from saucer-shaped to tubular or camp anulate, the epicalyx, sepals, petals, and stamens inserted on its rim, its inside usually lined by nectariferous tissue; disk sometimes distinct, intrastaminal; epicalyx + in some genera; sepals free; petals free, from large and showy to small and not or hardly distinct from sepals, rarely 0; stamens few to numerous, often their number distinctly related to the number of perianth parts; filaments free; anthers bilocular, dehiscing longitudinally; carpels 1-many, free or variously connate with each other and/or adnate to the hypanthium, forming 1 or more superior to inferior ovary(ies); stylodia (in monocarpellate ovaries styles) +, these sometimes (some Maleae) fused into a common, branched style; ovules 1-several (often 2) per carpel, anatropous, ascending or pendulous. Fruits various, fleshy or dry, dehiscent or not; seeds 1-several, testa usually firm, endosperm 0 or a thin layer, cotyledons fleshy or flat.

In his contribution to this series the late Dr. Kalkman (†19 January 1998) had included Quillaja in, and excluded Lyonothamnus from Rosaceae, although he had been aware of the results of the rbcL analysis by Morgan et al. (1994) that were in favour of the contrary. I have found no reason for rejecting these molecular findings and consequently have changed Dr. Kalkman’s manuscript and included Lyonothamnus in Rosaceae, although its position within the family remains unclear for the time being, whereas Quillaja has been removed from it and will be treated as a separate family in a further volume. After completion of Dr. Kalkman’s manuscript, some important papers dealing with the morphology and molecular systematics of Rosaceae have appeared, which may indicate that the maintenance of tribe Exocordeae is probably not justified. I have inserted references to these papers into the bibliography but in general have left untouched Dr. Kalkman’s argument, which to a large degree is supported by recent findings.