Flora of North America species comparison

Genera ca. 65, species ca. 1810 (16 genera, 92 species; 9 genera, 19 species introduced, and various hybrid complexes in the flora).

Iridaceae are currently divided into four subfamilies (P. Goldblatt 1990, 1991). Subfamily Isophysidoideae Takhtajan is monotypic, comprising the Tasmanian Isophysis T. Moore with a superior ovary. Only subfamily Iridoideae is native in North America. The remaining Nivenioideae Schulze ex Goldblatt and Ixioideae Klatt are centered in Africa south of the Sahara. Iridaceae are of considerable economic importance in ornamental horticulture and the cut-flower industry, especially Iris, Gladiolus, and Freesia. Several other genera (e.g., Crocus, Dietes, Sparaxis, Tritonia, Watsonia) are cultivated in gardens in both tropical and temperate areas. Moraea and Homeria are poisonous and pose significant problems in cattle- and sheep-raising areas, notably in southern Africa.

In addition to the several genera and species escaped from cultivation and dealt with in detail below, the following are widely grown in areas of mild winter and may persist in and near abandoned gardens, sometimes reproducing successfully: Dietes Salisbury [D. iridioides (Linnaeus) Salisbury ex Klatt, D. grandiflora N. E. Brown]; Ixia Linnaeus (I. maculata Linnaeus, I. polystachya Linnaeus); Crocus cultivars and even some wild species.

(Discussion copyrighted by Flora of North America; reprinted with permission.)

Species ca. 80 (37 in the flora).

Sisyrinchium is a complex polyploid taxon in which the species are not always easily distinguished. When immature, plants of branched species appear to be simple-stemmed (internodes do not elongate until just before anthesis) and those of usually simple-stemmed species occasionally are branched. White flowers may occur in otherwise blue-flowered species, and vivipary occasionally occurs. Furthermore, vegetative characteristics, while distinctive in some species, may overlap greatly in wide-ranging species. Writers of past floras sometimes were unaware of such phenotypic plasticity, or were inconsistent in their use of terminology (spathes, for instance, have variously been called leaves or valves). Some taxonomists have thought differences too subtle and chosen to lump species (e.g., S. angustifolium and S. montanum have been considered synonymous by several authors).

Flowers and spathes provide the best characters but require fresh material or extreme care during pressing. Many older herbarium collections have been misidentified due to improper collecting/pressing techniques that obscured critical characters. Additionally, variability within populations makes accurate identification based on only one individual nearly impossible.

Chromosome counts can be useful in differentiating species but consistent numbers seem to be obtained only with meiotic counts. Some of the unusual counts reported in the literature were obtained with mitotic material or were from misidentified plants. A definite ploidal series seems to be evident, however, at least among the blue-flowered species. Limited breeding studies have shown that ploidal levels are strong barriers to reproduction (A. F. Cholewa and D. M. Henderson 1984; D. M. Henderson 1976). There is some indication, however, that true hybrids may exist (D. B. Ward 1959; D. S. Correll and M. C. Johnston 1970; G. Ajilvsgi 1984). Much more work, especially with molecular data and cladistic analyses, needs to be done to understand species relationships.

In eastern Texas (especially) and adjacent states, intermediates between Sisyrinchium ensigerum, S. pruinosum, S. langloisii, and perhaps S. sagittiferum can be found. These have been referred to S. texanum E. P. Bicknell (D. S. Correll and M. C. Johnston 1970), and Bicknell’s original description of that entity does mention a high degree of variability. Many of the specimens that we examined that were previously labeled S. texanum exhibit character states intermediate between those of S. langloisii and S. pruinosum or S. ensigerum. The types of S. texanum that we examined (Bray s.n., NY; Bush 32, NY; Hall 636, NY) would key to S. langloisii or S. pruinosum but exhibit several intermediate character states. Further, although K. L. Hornberger (1987, 1991) and D. S. Correll and M. C. Johnston (1970) considered S. texanum synonymous with S. sagittiferum (which they described as a branched species), E. P. Bicknell (1899, 1901) clearly indicated that the latter is unbranched, and thus it probably does not play a role in S. texanum. Much more work is needed to resolve the proper disposition of S. texanum and the true nature of this complex of southern species.

Characters critical for distinguishing species of Sisyrinchium are often found in floral material, requiring extra care in collecting and pressing. Accurate identification requires examining more than one individual in order to discount uncommon or atypical character states (generally not accounted for in this key). In branched individuals, leaves will be associated with one or more branches or peduncles that are narrower than the primary stem. Peduncles are morphologically indistinguishable from branches and are, hereafter, referred to simply as branches.

(Discussion copyrighted by Flora of North America; reprinted with permission.)