Cirsium altissimum
Cirsium edule
roadside thistle, tall thistle
cardon, edible thistle
single, erect, villous with septate trichomes, sometimes ± glabrate, sometimes distally thinly tomentose;
branches few–many, ascending.
usually 1, erect, simple to openly branched in distal 1/2, ± villous with jointed trichomes, sometimes finely arachnoid, sometimes ± glabrate;
branches 0–many, ascending.
blades oblanceolate to elliptic or ovate, 10–40 × 1–13 cm, margins flat, finely spiny-toothed and otherwise undivided to coarsely toothed or shallowly pinnatifid, lobes broadly triangular, main spines 1–5 mm, abaxial faces white-tomentose, adaxial faces green, glabrate to villous with septate trichomes;
basal usually absent at flowering, winged-petiolate, bases tapered;
principal cauline well distributed, gradually reduced, bases narrowed, sometimes weakly clasping;
distal cauline well developed.
blades oblong to elliptic or oblanceolate, 5–50 × 1–10 cm, plane to moderately undulate, coarsely dentate to deeply pinnatifid, lobes 5–10 well separated, linear, narrowly to broadly triangular, spinulose to spiny-dentate or shallowly lobed, main spines 3–10 mm, abaxial faces thinly to densely villous along major veins with septate trichomes, sometimes thinly arachnoid-tomentose, sometimes glabrescent, adaxial glabrous to sparsely villous or shaggy-tomentose along midveins with septate trichomes;
basal often absent at flowering, spiny winged-petiolate or sessile;
principal cauline well distributed, only gradually reduced, bases auriculate-clasping;
distal moderately to strongly reduced, thin, often spinier than the proximal.
0–5 cm (leafy-bracted).
0–5(–30) cm.
ovoid to broadly cylindric or campanulate, (2–)2.5–3.5(–4) × (1.5–)2–3(–4) cm, thinly arachnoid.
narrowly ovoid to hemispheric or campanulate, 1.5–3.5 × 1.5–4 cm (including spines), loosely to densely arachnoid with fine, non-septate trichomes.
tips 4–6 mm.
pink to purple (white), 20–35 mm, tubes 10–16 mm, throats 5–12 mm, lobes 5–9 mm.
purple (pink or white), (15–)18–22(–33) mm, tubes 7–11 mm, throats (4–)5–8.5(–13) mm, lobes linear but not filiform, not knobbed at tips, (2–)4.5–7(–10) mm;
style tips 3–4(–5) mm, conspicuously exserted beyond corolla lobes.
in 10–20 series, strongly imbricate, greenish with subapical darker central zone, ovate (outer) to lanceolate (inner), abaxial faces with a narrow glutinous ridge (milky when fresh, dark when dry), outer and middle entire, bodies appressed, spines slender, abruptly spreading, 3–4 mm;
apices of inner phyllaries spreading, narrow, flattened, entire, spines spreading, slender, 3–4 mm;
apices of inner phyllaries spreading, narrow, flattened, ± dilated, ± erose or finely serrulate.
in 4–8 series, subequal, green or often purplish, bodies short, appressed, abaxial faces without glutinous ridge, apices stiffly radiating to ascending, straight or flexuous, narrowly linear, plane to acicular, spines straight, slender, 1–10+ mm;
outermost spiny-fringed or entire, mid entire or minutely serrulate;
apices of inner straight, sometimes expanded and erose, flat.
1–many, in corymbiform or paniculiform arrays, (± elevated above principal cauline leaves), not subtended by ring of spiny bracts.
1–many, erect, often crowded and ± sessile in tight clusters at stem tips, closely subtended by clusters of leafy bracts or not, collectively forming corymbiform or paniculiform arrays.
tan to dark brown, 4–5.5 mm, apical collars stramineous, 0.5–1 mm;
pappi 12–24 mm.
dark brown, 3.5–6.5 mm, apical collars not differentiated;
pappi 9–19(–25) mm.
= 18.
Cirsium altissimum
Cirsium edule
Plants of Cirsium altissimum ranging from southern Minnesota to Texas often have more deeply divided leaves than do populations in other portions of the species’ range. Some botanists (e.g., R. J. Moore and C. Frankton 1969; D. S. Correll and M. C. Johnston 1970) have treated those plants as C. iowense. Others (e.g., R. E. Brooks 1986; H. A. Gleason and A. Cronquist 1991; G. B. Ownbey and T. Morley 1991) have treated them as C. altissimum. Still others considered them to be derivatives of hybridization between C. altissimum and C. discolor (J. T. Kartesz and C. A. Meacham 1999) and treated them as C. ×iowense. Indeed the existence of these plants blurs the distinction between C. altissimum and C. discolor, and herbarium specimens are often difficult to assign.
Natural hybrids between Cirsium altissimum and C. discolor are well documented (R. A. Davidson 1963; G. B. Ownbey and Hsi Y.-T. 1963; Ownbey 1964; S. Dabydeen 1997). Ownbey and Dabydeen both reported that apparent F1 hybrids between the two species have low seed set in comparison with the parental taxa. W. L. Bloom (1977) reported that the chromosomes of the two species differ by several rearareaments. Dabydeen reported a count of 2n = 19 with multiple meiotic irregularities for an apparent F1 hybrid. However, the presence of numerous individuals and populations seemingly intermediate between C. altissimum and C. discolor indicates that although F1 hybrids have low fertility, long-term processes may have stabilized hybrid derivatives of higher fertility. Ownbey and Hsi reported mitotic counts of 2n = 18 and 20 from a population that they treated as C. altissimum. In their discussion they noted that their plants represented “the segregate called C. iowense” and had been collected a short distance from that taxon’s type locality. R. J. Moore and C. Frankton (1969) reported a chromosome number of 2n = 18 for a plant from Texas that they considered to be C. iowense. Further investigation of morphologic variation, chromosome number, meiotic behavior, and fertility is needed of populations named as C. iowense to determine how those plants should be treated.
(Discussion copyrighted by Flora of North America; reprinted with permission.)
Varieties 3 (3 in the flora).
The edible thistle has had a convoluted nomenclatural history. The labels on the type collection (BM) bear the following information: “Circium [sic] * edule” and “R. Mountains & plains of Columbia.” When Nuttall published the name, he listed the range as “The plains of Oregon and the Blue Mountains.” The type specimen closely resembles plants from western Oregon, but plants of Cirsium edule are not known to occur in the Blue Mountains. J. T. Howell (1943) noted that the name C. edule had long been in use for two distinctly different species, one with a long, slender corolla tube, short lobes, and a barely exserted style and the other with a stouter corolla with a shorter tube, longer lobes, and a long-exserted style. He applied the name C. edule to the species with the slender corolla, and took up the name C. macounii for the second species. After examining the type of C. edule, A. Cronquist (1953) pointed out that Howell had erred in applying that name to the short-styled species, and described the latter as C. brevistylum. Cronquist expressed doubt as to the collection locality of the type of C. edule, focusing on the Blue Mountains and not noting the duality of the location data on the specimen and in Nuttall’s publication. It is likely that the plants Nuttall observed in the Blue Mountains were C. brevistylum.
J. T. Howell (1960b) resurrected the name Cirsium hallii to apply to a group of Oregon thistles growing west of the Cascade Range, attributing to it a type locality of Salem, Oregon. Howell (1943) had noted that he had borrowed “the type” of C. hallii from the Gray Herbarium. This species had been described (as Cnicus hallii) by Gray based upon three syntypes (one each cited from California, Utah, and Oregon). The specimen examined by Howell was apparently the Oregon collection by [Elihu] Hall that Gray cited. The Utah and California specimens are different taxa. After examining photographs of the holotype of C. edule and the Oregon specimen of C. hallii, and various specimens collected in the area that Howell described as the range of C. hallii, I have concluded that C. hallii and C. edule are clearly conspecific.
Cirsium edule is a polymorphic species much in need of an in-depth field-based investigation. R. J. Moore and C. Frankton (1962) noted that in the northern part of its range, C. edule occurs mostly at elevations from 300 to over 2100 m. However, along the Oregon coast the species occurs on sea bluffs a few meters above the surf. Populations from montane sites are often rather different in appearance from those of lowland areas, and coastal plants differ from those of nearby more interior areas. Both montane and strictly coastal plants tend to be compact with heads tightly crowded and usually with very densely arachnoid involucres. Plants of non-montane interior sites tend to be taller and more openly branched. Plants of interior sites in southern Washington and Oregon have smaller heads with less densely arachnoid involucres than those farther to the north or along the seashore. The spiny tips of the phyllaries may be ascending or may radiate from the head forming a dense, spiny ball.
Hybridization may have played a role in the diversification of Cirsium edule. Hybrids between C. edule var. macounii and C. brevistylum in southern Canada have been named as C. ×vancouveriense R. J. Moore & C. Frankton. Cirsium edule and C. brevistylum overlap extensively in parts of their ranges and hybrids may occur throughout their area of sympatry. Some of the variation in the southern part of the range of C. edule may be a result of past introgression with various forms of C. remotifolium.
(Discussion copyrighted by Flora of North America; reprinted with permission.)
1. Heads mostly borne singly; peduncles 10–30 cm | var. wenatchense |
1. Heads crowded at stem tips; peduncles usually 0–1 cm | → 2 |
2. Phyllary apices plane to acicular, ascending or ± spreading, spines 1–5 mm | var. edule |
2. Phyllary apices long-acicular, widely spreading, spines 5–15 mm | var. macounii |
Go Botany
IL Wildflowers
KS Wildflowers
LA Plants
MD Biodiversity
MI Flora
MN Wildflowers
MO Plants
WildflowerSearch
iNaturalist
USDA Plants Database
BC
OR
WA
Flora NW
PNW Herbaria
