.

Understanding Knapweeds
Centaurea (Asteraceae, Compositae)

By Arieh Tal, 2016

.

This article applies to a small number of species in the genus Centaurea (knapweeds).  These species are found in northeastern North America, and points farther west and south.  All of these species have flowers with white, pink, magenta or purple corollas, involucres that are 15-18 mm high, appendages of involucral bracts that are truncated at the base, and are neither spine-tipped nor with tips that are very narrow and recurved.

These species, considered together, are sometimes referred to as the "Centaurea jacea complex", and include Centaurea jacea, C. ×moncktonii, C. nigra and C. nigrescens.  None of these species is native to North America.  They all have a well-deserved reputation for being difficult to distinguish from each other.

The characters most useful for identification of these species are characters of the flower heads ("heads") and the fruits (cypselae).  Stem and leaf characters are more or less the same for all 4 species, and thus are not considered very diagnostic.

x

Basic Flower Head Morphology

[ If you already know the morphology, you may want to skip to the discussion below. ]

.

A composite "flower" is not the entire flower head, though it may appear to be so.  The actual flowers are the small "florets" that comprise the head.

.

.

  
              Figure 1.  Schematic drawing of composite flower head with only disc florets.
.
.

Figure 1 depicts a flower head that has florets only in the center of the disc.  The disc florets are usually very small and not showy, like "rays".  Many species in the composite family have flower heads without rays, including some Centaurea species.

.


          
   Figure 2.  Schematic drawing of composite flower head with both ray and disc florets.

.

Figure 2 depicts a flower head that has both ray and disc florets.  Ray florets are often much larger than disc florets.  The "lobes" of ray florets are fused together to form an enlarged "strap" called a "ray".  Most of the showy composite family species, such as daisies, sunflowers and asters have heads with both ray and disc florets.

.

.
Figure 3.    (A) Flower head of Centaurea nigra, with only disc florets.  (B) Flower head of Centaurea jacea, with both enlarged, peripheral florets and smaller (central) disc florets.  (Pictures are not to same scale.)

.

Understanding the difference between knapweed flower heads with and without enlarged peripheral florets is important, but understanding differences in characteristics of the involucres and bracts of the involucre is more important.  In the Asteraceae, the involucre is a collection of small, leaflike "bracts" that partially surrounds and helps protect the collection of florets comprising a flower head.  The involucral bracts of Centaurea species consist of two parts:  the body of the bract (usually green) and an appendage at the top of the body.

.


Figure 4.   Flower head of Centaurea stoebe, showing characteristics of the involucre.  The appendages of the involucral bracts are the dark, triangular structures with comb-like fringes.

.

Centaurea involucral bracts are highly variable.  Understanding the differences are important for identification of the various species.

 

. .

Figure 5.   (A) Involucre of Centaurea jacea.  (B) Involucre of Centaurea nigra.  (C) Involucre of Centaurea nigrescens.  (D) Involucre of Centaurea ×moncktonii(Pictures are not to exactly the same scale.]

.

Discussion, Tips and Caveats

.
Centaurea jacea  (brown knapweed):  This species always produces heads with enlarged, showy, ray-like, neuter, peripheral florets and smaller, bisexual, fertile, central disc florets.  Appendages of involucral bracts are light brown to golden-brown, concave and roundish, at least at the top.  Margins of the involucral bracts are ragged, lacerated or sometimes with a prominent notch.  Bodies of involucral bracts are mostly obscured by appendages of adjacent bracts.

Centaurea nigra  (black knapweed):  This species never produces heads with enlarged ray-like, peripheral florets.  That is, all florets are bisexual, fertile, disc florets.  Sometimes the outermost disc florets lean outward beyond the circumference of the disc, due to pressure, but are not enlarged or showy.  Appendages of involucral bracts are dark brown to black (taller than 2.2 mm), at least the outermost and middle ones triangular in shape, with numerous, conspicuous comb-like fringes along the margins, which largely obscure the green bodies of the adjacent bracts above.  Bodies of the outermost bracts (the ones near the bottom of the involucre) may not be fully obscured.  The uppermost bracts (the ones at the very top of the involucre) may be roundish and lack comb-like fringes, but should be dark brown or black.

Centaurea nigrescens  (short-fringed knapweed):  The involucre of this species more closely resembles that of C. nigra, except that the appendages of the involucral bracts are shorter (up to 2.2 mm tall) and possess fewer (approx. 5-8) comb-like fringes.  The appendages may be dark brown to black, but do not fully obscure the bodies of the adjacent bracts.  The uppermost bracts (the ones at the very top of the involucre) may be roundish and lack comb-like fringes, but should be dark brown or black.  This species produces heads with bisexual, fertile disc florets, but may also produce heads with enlarged, showy, neuter, peripheral florets.  (All the specimens I have seen have the enlarged, peripheral florets, but the appendages of the involucral bracts will look more like those of C. nigra than C. jacea.)

Centaurea ×moncktonii  (hybrid knapweed):  Because this species is a hybrid, it tends to exhibit variable morphology.   Showy, enlarged, neuter, peripheral florets may be present - or not.  (All the populations I have observed have the enlarged, peripheral florets.)  Disc florets will be bisexual and fertile.  Appendages of the involucral bracts are usually intermediate between those of typical C. jacea and typical C. nigra.  That is, they will tend to be more roundish than triangular, lighter brown rather than black, with fewer comb-like fringes.  Thus, they often don't obscure as much of the bodies of the adjacent bracts (more green showing).  Some specimens may have a large percentage of bracts with appendages similar to those of C. jacea, while others have fewer such appendages.  In sum, there is room for ambiguity.  Some of these plants may look very similar to C. jacea, but not quite.  If there are any bracts with comb-like fringes, then chances are it is the hybrid, rather than C. jacea.

.

Tips, etc.:  The fruits (cypselae) of these 4 species are similar in size, shape and color.  Most of the differences pertain to the nature of the pappusCentaurea jacea cypselae do not possess a pappus, whereas the other three may produce a pappus of variable-length (but short), minute bristles.  However, a pappus of the other three species may either not be produced, or may tend to be quickly deciduous.

Corollas and cypselae of these species also tend to disperse quickly once mature, as the flower heads are tossed about by the wind; or the cypselae are eaten by birds.  Many of the dried specimens I have observed were difficult to identify to species because the corollas and cypselae were gone (i.e., deciduous).  All that remained was the involucres.  So, it wasn't possible to check the florets to determine whether they were enlarged or not, or fertile.  Similarly, during pressing of the specimens the involucral bracts may have been pushed upwards due to pressure, such that the bodies of the involucral bracts became exposed, giving the impression that the appendages of the adjacent bracts are not obscuring them.

So in sum....  Collect and examine specimens when they are in flower, if possible.  It is much easier and more informative to work with live material.  Immature fruits of live flower heads will still be present and the requisite character states will still be observable.  Also, I don't feel that involucre height/width measurements, are diagnostically very reliable.   It seems preferable to measure the height of the appendages of the involucral bracts, and note the number of comb-like fringes, esp. when working with pressed material.

The nominclatural and taxonomic history of these species is well documented in Keil and Ochsmann (2006).  See below.

.

Selected references:

.

Haines, A.  2011.  Flora Novae Angliae:  A Manual for the Identification of Native and Naturalized Higher Vascular Plants of New England.  Yale University Press, New Haven, Conn.

Keil, D.J. and Jörg Ochsmann.  2006.  Centaurea, in Flora of North America Editorial Committee, eds.  Flora of North America, 19:181. Oxford University Press, New York, NY.

 

 

Copyright  botphoto.com  and Arieh Tal, 1990 to 2016.  All rights reserved.  ( Terms of use )

Page updated 2016.08.31