Flora of North America species comparison

Species ca. 150 (17 in the flora).

As noted by J. W. Thieret (1971), systematists have generally accepted the classification by G. Beck (1930) of a broadly circumscribed Orobanche comprising four sections, all of which are represented in the flora area. However, J. Holub (1990) and some other authors have questioned whether this classification accurately reflects the phylogeny and taxonomic complexity of the group and have proposed recognizing the sections of Beck as separate genera. More recently, cytological and molecular phylogenetic studies have added new data to the discussion but have not resulted in a well-supported, revised classification. G. M. Schneeweiss et al. (2004b) reviewed a large series of new and earlier chromosome counts and concluded that sect. Orobanche has a base number of x = 19, whereas members of the other sections are characterized by x = 24. As summarized by J. M. Park et al. (2008), molecular data have resulted in discordant phylogenies, depending on taxon sampling and whether plastid or nuclear markers were sequenced, but have suggested that five lineages should be recognized within the traditional Orobanche. The fifth lineage has been segregated by some authors as the monospecific Boulardia latisquama F. W. Schultz [O. latisquama (F. W. Schultz) Battandier], native to the Iberian Peninsula and Morocco. The molecular data support the group as monophyletic whether treated as one genus or several. Although the authors retain Orobanche in the broad sense, it seems inevitable that Orobanche will be split into three or more genera once the number of well-supported segregates is resolved. Most recently, A. C. Schneider (2016) has made a case for splitting all of the species native to the New World [sect. Gymnocaulis Nuttall and sect. Nothaphyllon (A. Gray) Heckard] into two sections of a single generic segregate, Aphyllon Mitchell, and has published new combinations for the constituent taxa.

As outlined in the key to species below, the two Old World sections present as adventives in our region are notable for their usually four-lobed or -toothed calyces and spicate inflorescences. Members of the two New World sections have five-lobed calyces and a variety of inflorescence types. Section Orobanche includes 120 to 130 species and is widespread in Europe, Asia, and Africa. The flowers lack bracteoles, the calyces are deeply two-parted with one or both halves toothed or lobed, and the corollas are usually yellow, red, or brown. The 12 to 16 species of sect. Trionychon Wallroth (Phelipanche Pomel) are most diverse in the Mediterranean region; their flowers have a small pair of bracteoles adnate to the calyx bases, the calyces are four-toothed or -lobed, and the corollas are usually purple or blue.

In the New World, the small sect. Gymnocaulis [sect. Euanoplon (Endlicher ex Walpers) Thieret; Aphyllon] comprises two or more species widespread in the flora area and is notable for its condensed inflorescence axes shorter than the elongate pedicels, flowers lacking bracteoles, and calyces shallowly to moderately lobed. About 15 members of sect. Nothaphyllon [sect. Myzorrhiza (R. A. Philippi) Beck-Mannagetta; Myzorrhiza R. A. Philippi] are widespread from temperate North America south to Guatemala, with a few disjunct species in western South America. They are characterized by usually elongate inflorescence axes usually much longer than the pedicels, flowers with a pair of bracteoles, and deeply five-lobed calyces.

Several of the Old World species are introduced widely, including the species treated below. They parasitize various crop plants, principally tomatoes (Solanum lycopersicum), tobacco (Nicotiana tabacum), clovers (Trifolium spp.), and hemp (Cannabis sativa) in the United States and are considered agricultural pests. In response, the Federal Government has placed all of the non-native taxa of Orobanche on the United States Department of Agriculture’s noxious weed list (http://www.aphis.usda.gov/plant_health/plant_pest_info/weeds/downloads/weedlist.pdf), which regulates their movement into, out of, and within the United States. Several states also regulate various non-native broomrapes as noxious weeds.

In 2014, an infestation of Orobanche aegyptiaca Persoon [Phelipanche aegyptiaca (Persoon) Pomel] was discovered in a tomato field in Solano County, California, the first report of this taxon from the flora area. Egyptian broomrape is native to the Middle East and adjacent portions of Europe and Asia but is a widely distributed weed in Europe, Asia, and Africa, and has also been reported from Cuba (R. Oviedo Prieto et al. 2012). Its principal negative economic impact is to crop species in the Solanaceae (tomatoes, potatoes, tobacco), but, like O. ramosa, it has a broad host range, including a variety of other crops. Although the appearance of the species in North America is not unexpected, a full treatment is not provided here, because there is no certainty that it will become established. Vigorous efforts have been undertaken to eradicate plants at the sole known location. Orobanche aegyptiaca is similar to O. ramosa in its branched, glandular-pubescent stems and general floral morphology, but it differs from that species in its usually more robust habit (stems to 40 cm), with larger corollas (20–35 mm) that are often darker purple, and by having densely villous versus glabrous anthers.

Plants of Orobanche hederae Duby (ivy broomrape), another Eurasian species in sect. Orobanche, have been documented since 2000 parasitizing planted Hedera (Araliaceae) in Alameda County, California, on the campus of the University of California-Berkeley. It can be difficult to distinguish from the morphologically variable O. minor, differing in its often slightly more lax inflorescences and corollas with a constriction on the abaxial side just behind the throat. For recognition, most botanists focus on the difference in hosts, as O. hederae parasitizes species of Hedera. The taxon is excluded from the flora for the present, because its host specificity has limited its ability to spread beyond the immediate vicinity of the initial infection site.

The native North American species have had a negligible impact on agriculture. Orobanche cooperi and O. riparia (as O. ludoviciana) have been reported on cultivated tomato crops in southern California and on tobacco crops in the Ohio River Valley, respectively, with O. cooperi having required control measures (G. H. Starr 1943; S. Wilhelm et al. 1958).

Some of the western North American taxa were recorded as a minor, but apparently widely utilized, food source for western American Indians and were also used medicinally, mainly for treating colds and other pulmonary ailments (D. E. Moerman 1998).

The North American Orobanche species are more host-specific than has been reported previously (A. C. Schneider et al. 2016b), and understanding of host relationships continues to be refined. Many collectors merely note the nearest living plant as the host species or have neglected to note host species entirely, thus perpetuating imprecise knowledge of host relationships in this group. In this treatment, a host species is noted as a primary host when it has been confirmed multiple times by excavation and is frequently reported as the host. Less frequent, but reliable, reports of other hosts are noted as occasional hosts. Localized shifts in host preference in Orobanche species may reflect an indication of lineage divergence.

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

Genera ca. 100, species ca. 2000 (27 genera, 292 species in the flora).

Orobanchaceae are now defined to include both the holoparasitic members traditionally included in the family (A. Cronquist 1981) and the hemiparasitic genera formerly included in Scrophulariaceae. Although multiple research groups focus on members of the Orobanchaceae, a widely accepted infrafamilial classification of the family in the sense of Angiosperm Phylogeny Group (2016) has not yet appeared.

The classification by J. R. McNeal et al. (2013), who found that Orobanchaceae comprise six clades, is followed herein (their named clades are roughly equivalent to tribes). The autotrophic Lindenbergia Lehmann (12 species in the Old World) corresponds to the basal clade sister to the rest of the clades. Species in our region are distributed among the remaining five clades: Cymbarieae D. Don (genus 1), Orobancheae Lamarck & de Candolle (genera 2–6), Rhinantheae Lamarck & de Candolle (genera 7–12), Buchnereae Bentham (genera 13 and 14), and Pedicularideae Duby (genera 15–27). Within the family, genera are arranged alphabetically within tribes, or within Pedicularideae, in subgroups within the tribe.

Parasitic plants attach to their hosts via haustoria (L. J. Irving and D. D. Cameron 2009). Haustoria are produced by both hemiparasitic and holoparasitic Orobanchaceae (E. Fischer 2004). In hemiparasitic taxa, haustoria usually tap their host’s xylem, mostly taking up water, mineral nutrients, and nitrogen from their host, and sometimes also carbon. Holoparasitic taxa derive all of their growth requirements predominantly from the host’s phloem (Irving and Cameron).

Parasitism has evolved once in the family (N. D. Young et al. 1999; J. R. McNeal et al. 2013); holoparasitism has arisen independently three times from the hemiparasitic condition (J. R. Bennett and S. Mathews 2006; McNeal et al.).

Some Orobanchaceae are serious pests, primarily on legume and grain crops in warmer and drier areas, especially in sub-Saharan Africa. Striga is a particularly serious pest that parasitizes mostly monocots; S. gesnerioides attacks eudicots (K. I. Mohamed et al. 2006). Orobanche parasitizes eudicot crops primarily in temperate parts of the world (E. S. Teryokhin 1997). All Striga species and non-native species of Orobanche in the flora area are listed on the Federal Noxious Weed List (http://www.aphis.usda.gov/plant_health/plant_pest_info/weeds/downloads/weedlist.pdf) in the United States.

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