Redbuds and Legumes Subfamilies W

University of Richmond UR Scholarship Repository

Biology Faculty Publications Biology

Spring 2013 Redbuds and Legumes Subfamilies W. John Hayden University of Richmond, [email protected]

Follow this and additional works at: http://scholarship.richmond.edu/biology-faculty-publications Part of the Botany Commons

Recommended Citation Hayden, W. John. "Redbuds and Legumes Subfamilies." Bulletin of the Virginia Native Plant Society 32, no. 2 (Spring 2013): 1, 4-5.

This Article is brought to you for free and open access by the Biology at UR Scholarship Repository. It has been accepted for inclusion in Biology Faculty Publications by an authorized administrator of UR Scholarship Repository. For more information, please contact [email protected]. Vol. 32, No. 2 Spring 2013 ISSN 1085-9632

A publication of the VIRGINIA NATIVE PLANT SOCIETY http://www.vnps.org Conserving wild flowers and wild places First Legume WOY Redbuds and Legume subfamilies Although legumes constitute one of should be familiar to most readers of numerous small leaflets. Further, the largest families of flowering plants the VNPS Bulletin, the so-called mimosoid flowers are individually in the world, and despite 25 years of mimosa, Albizzia julibrissen, na- small and radially symmetric (Figure celebrating Virginia’s wildflowers, red- tive to western Asia, but now wide- A), but they are most easily recognized bud (Cercis canadensis) is the first le- spread in much of North America. by their occurrence in tight head-like gume to be recognized as a VNPS Wild- Most mimosoid legumes are woody clusters dominated by numerous elon- flower of the Year. This article addresses plants of the tropics and subtropics; gate styles and stamens—the overall the relationships of Cercis with the rest relatively few occur in temperate re- effect resembling a powder puff. There of the legumes (family Fabaceae, or gions. Leaves are often bipinnate with (See Redbud, page 4) Leguminosae in older literature). In general, legumes can be recognized by their usually compound, stipu- late leaves bearing hinge-like swollen pulvini on petioles and petiolules (leaf- let stalks); floral details vary from group to group (see below), but there is always a single pistil with a superior ovary that matures into a dry fruit that is usually flattened, elongate, multi-seeded, and dehiscent along both sides. The characteristic fruit is known botani- cally as a legume—which leads me to the nearly tautological truism, “Le- gumes make legumes.” Traditionally, legumes (the plants) have been parti- tioned into three well-defined subfamilies—Mimosoideae, Caesalpinioideae, and Papillionoideae—distinguished largely by details of floral structure. Mimosoid legumes are most diverse in the tropics and subtropics, but Pea pods? Well, sort of...You are looking at the seed pods of the VNPS at least one member of this subfamily 2013 Wildflower of the Year, redbud. (Photo courtesy John Hayden) Bulletin of the Virginia Native Plant Society •Redbud (Continued from page 1) A. Subfamily Mimosoideae are two mimosoid legumes native to Vir- ginia: Aeschynomene virginica, sensitive or Virginia jointvetch, a rare plant of freshwater tidal marshes, and Mi- mosa (Schrankia) microphylla, littleleaf sensitive-briar, a species of the South- east that reaches its northern limit in Vir- ginia where it, too, is considered rare. Acacia Papillionoid legumes are cosmo- politan, well developed and diverse from the tropics to temperate regions. All growth habits are well represented among the papillionoid legumes, but B. Subfamily Caesalpinioideae herbaceous forms are especially common in temperate regions. Leaves are mostly once-pinnate or trifoliolate. Flowers are bilaterally symmetric with two keel petals that enclose 10 stamens and the pistil, two laterally divergent wing petals, and an erect standard (or banner) petal (Figure C). Further, the 10 stamens can be completely separate from each other, fused in a ring, or, most frequently, nine Cercis are fused along their filaments forming a partial tube surrounding the pistil, with the 10th stamen separate from the rest, ex- tending parallel to the upper edge of the C. Subfamily Papillionoideae pistil. There are many papillionoid legumes native to Virginia. A few of the more familiar genera include Desmodium (beggar’s ticks), Robinia (locusts), Tri- folium (clovers), Vicia (vetches), and Wisteria. The traditional definition of caesalpinioid legumes—to which red- Cytisus buds belong—suggests some degree of intermediacy between the other two subfamilies. Like the mimosoids, most are woody and tropical or subtropical, but Sketches and diagrams of legume flowers representing the three traditionally there are temperate zone examples, in- recognized subfamilies. Key to floral organs in the diagrams (right hand side cluding some herbs. Floral symmetry of figure): calyx/sepals are shaded, corolla/petals are black, stamens are 4- varies from radial to extremely bilateral; lobed and white, pistils are the centermost element of each diagram; note that in some cases, redbud being a good ex- papillionoid stamens are linked indicating fusion; sepals in each subfamily ample, flowers appear superficially very may be fused at the base, a detail not depicted in these diagrams. Redrawn by much like papillionoid flowers. How- Nicky Staunton from images on the Watson and Dallwitz web site (http:// ever, when possessing bilateral symme- delta-intkey.com/angio/). try, caesalpinioids always differ from papillionoids in one respect: the upper- gumes is that roots lack the nodules Chamaecrista (partridge pea), Senna, most petal (standard or banner) is cov- containing nitrogen-fixing symbiotic Gleditsia (honey locust), and ered by the two wing petals in flower bacteria that are widespread among Gymnocladus (Kentucky coffee tree). buds whereas among papillionoids the other legumes. Though few in number, the Virginia uppermost petal is always outermost in In addition to redbuds, some caesalpinioids form a heterogeneous the bud (Figure B). Another interesting prominent caesalpinioid legumes to be group: Chamaecrista species have characteristic of many caesalpinioid le- found in Virginia include species of (See Redbud, page 5) Page 4 Spring 2013 Bulletin of the Virginia Native Plant Society

•Redbud form a single distinct lineage. Rather, 19th century, a time when gross mor- subfamily Caesalpinioideae is com- phology dominated how botanists per- (Continued from page 4) posed of six (maybe more) discrete lin- ceived relationships. Nowadays much symbiotic nitrogen fixation in root eages, several of which (for example, more data are available to systematists; nodules, while the others do not; tribe Cercideae) form the lowermost it is now commonplace to integrate tra- Chamaecrista and Senna are herba- branches of the legume evolutionary ditional gross morphology with micro- ceous, but the others are woody; and tree, while the remaining branches are scopic structure, comparative chemis- flower symmetry varies from essen- interspersed among the well-defined try, and vast amounts of gene sequence tially radial in Gleditsia and mimosoids and papillionoids. By the data in order to generate classifications. Gymnocladus to slightly bilateral in modern philosophy of systematics, this Not only is there much more informa- Chamaecrista and Senna to extremely situation is a mess . . . but exactly how tion available, but the principles of cla- bilateral in Cercis. best to resolve it is not yet clear. One distics, now firmly ascendant in sys- Within Caesalpinioideae, Cercis may expect proposals to define addi- tematics, alter how hypotheses about is classified in tribe Cercideae tional subfamilies that will, in effect, relationships are evaluated. As it (Wunderlin 2010), a group of 12 gen- dismantle the traditional broad defini- turned out, the morphology used by era that, except for the redbuds, are tion of Caesalpinioideae. In all likeli- 19th-century botanists to distinguish native to tropical regions of South hood, a much more narrowly defined mimosoid and papillionoid legumes America, Africa, and Australia. To Caesalpinioideae will emerge, and Cer- correlates well with patterns revealed most Virginians, these redbud rela- cis will be placed elsewhere. Stay tuned! via gene sequence data and the prin- tions are obscure plants but some It may be disconcerting to learn that ciples of cladistic classification, con- readers may be familiar with the large classification of eastern redbud, a plant firming, at least in broad overview, genus Bauhinia (so-called orchid named by Linnaeus more than two and these two very old hypotheses of le- trees), frequently seen as ornamentals a half centuries ago, is currently in flux. gume relationships. The problem with in conservatory collections and tropi- There are lessons to be learned here. The the caesalpinioids is not simply fail- cal landscapes. first lesson is that naming a plant and ure of morphological characters to de- The preceding sketch of legume placing that plant in a classification are fine the subfamily. The problem stems relationships will almost certainly two different enterprises. While many from the very fine resolution of rela- undergo some revision in the not of the names coined by Linnaeus are tionships revealed by gene sequence too distant future. Molecular ge- considered valid and enjoy widespread data coupled with the relatively new netic studies confirm, at least in use today, the very concepts of plant requirement that taxonomic groups be broad outline, the composition of family and subfamily that are such in- monophyletic (not merely descended subfamily Mimosoideae, subfam- tegral parts of modern plant taxonomy from a common ancestor but also in- ily Papillionioideae, and tribe do not appear at all in the formal clas- cluding all descendants of that ances- Cercideae. However, subfamily sifications of Linnaeus. Naming and tor) that makes the subfamily problem- Caesalpinioideae, as traditionally classifying are not the same. The sec- atic. Finally, there is a third lesson to defined, is untenable in the light of ond lesson is that all classifications are be taken from the impending failure of current knowledge. The problem is hypotheses. The traditional definition traditional Caesalpinioideae: we just that caesalpinioid legumes do not of legume subfamilies arose during the don’t know everything there is to know about biodiversity. That’s true for plants in general and the legumes in particular. Yes, Cercis, the redbuds, are reasonably well known plants, but to place them properly in a robust classification (a hypothesis likely to with- stand rigorous testing), requires that we also know all the potential redbud relatives, i.e., all the caesalpinioids, equally well. We are not there yet! So, for now, redbuds are classified in subfamily Caesalpinioideae, but that is a temporary situation, an old concept retained out of expediency for lack of a better alternative. Wunderlin, R. P. 2010. Reorganization of the Cercideae (Fabaceae: Caesalpinioideae. Phytoneuron 2010-48: 1-5. John Hayden, VNPS Botany Chair Spring 2013 Page 5