Greater Kanpweed (Centaurea scabiosa)
Above: Greater Knapweed, Centaurea scabiosa, may have similar colored florets to Centaurea nigra and C. debeauxii but is much larger, usually strikingly radiate with peripheral florets modified as pseudorays and the phyllaries forming the cup or involucre are quite different. It is however found in similar habitats and often grows with Centaurea debeauxii in particular on chalk meadows/grassland.
The name knapweed comes from 'knap' referring to an ornate knob, in reference to the capitula with their ornate involucres formed from scale-like bracts called phyllaries. The phyllaries protect the developing fruit inside from insect herbivores, and also play an important role in seed dispersal as discussed below.
When looking at the structure of the flower head it is important to realize that the phyllaries and florets all attach on the receptacle at more-or-less the same level. Thus, the upper phyllaries are innermost and have elongated green portions that are hidden by the outer phyallries. Thus the involucre or cup of phyllaries is lined on the inside by the innermost/topmost phyllaries. this is sometimes made clear when a head is preparing to disperse its achenes, the outermost phyllaries may open fully, leaving the innermost still enclosed around the achenes and withering phyllaries, as seen below:
Above: although the outer phyllaries are open this capitulum is not yet ready to disperse achenes. the withering florets are still attached and the inner phyllaries are still closed.
Centaurea scabiosa has a woody rootstock and is, like C. nigra and C. debeauxii, a perennial. The stem leaves are broader but much more deeply divided into lobes or segments in scabiosa. The leaves are divided into shallow lobes and are described as pinnatipartite (in which the lobes are divided to just over half-way to the midrib) or pinnatifid with less deeply divided lobes. Sometimes these two terms are used interchangeably and their is confusion over the distinction. In pinnatisect leaves the clefts between the lobes reaches the midrib, but the bases of the segments are not narrowed into leaflets. In pinnate leaves, the lobes are distinctly separated from one-another along the midrib and may have their own stalks so as to form a compound leaf. Thus we have a sequence from least to most divided as follows:
pinnatifid < pinnatipartite < pinnatisect < pinnate
I would describe the leaf above as pinnatisect, almost pinnate. In all these cases, the reference to 'pinna' implies a symmetry in which the lobes are in pairs like the branches of a feather.
Some authorities recognize different variants of Centaurea scabiosa, but varieties assigned to Centaurea species often lack clarity and may be better classed as forms with distinct morphological traits. Type descriptions for varieties often refer to several characteristics, present in the type specimen(s), in addition to the main variant-defining attribute and this causes confusion when plants are encountered that do not have all the attributes but which may possess the central one. Are such plants individuals of the variant or a hybrid of variants or a different variant? Variant might be warranted when there is something distinct about a population or plants within a certain geographical region (there is no reason to assume that plants with similar characteristics found in separate geographical regions would be related at the variant level.
Varieties. What is well-established is that Centaurea scabiosa differ a lot in leaf shape and the extent to which the dark triangular phyllary appendages hide the green part of the phyllaries. Both Sell & Murrell's flora and Britton recognized var. succisaefolia, which both sources agree have largely undivided leaves; var. nana, recognized by Sell & Murrell, has capitula on short peduncles; whilst Britton recognized var. Riddelsdellii which has small capitula on conspicuously elongated branches and no or rudimentary leaves on the upper parts. Britton also recognized var. silesiaca (var. spinulosa) in which the phyllaries terminate in a conspicuous spine. The standard type is of course var. scabiosa, which is itself said to be very variable. More work to clarify the variants and forms and their geographical distribution might shed light on their significance.
The peduncles (capitulum stems) are 'naked' as they are devoid of leaves or sizeable bracts (scale-like bracts may occur) in contrast to C. nigra and C. debeauxii. Above and below: different views of the same individual.
The radiating outer florets are divided into linear or strap-shaped segments also to the base of the corolla (they are pseduorays). These outer radiating florets lack stamens and a style 9they are neuter) and lack an expanded upper throat but the long petals tubes are directed outwards then divide into 5 similar strap-shaped or linear segments.
Above: a rayless form (although the peripheral florets are slightly elongated). Centaurea scabiosa often grows among Centaurea debeauxii, as here,and in this case some of these scabiosa were small and rayless and superficially resembled the debeauxii rather closely. Intermediate forms have been noted by several botanists, but breeding experiments have failed to get C. debeauxii and C. scabiosa to hybridize and the two species appear quite incompatible (as scabiosa is with the nigra-group as a whole). However, where populations have grown together for a very long time, I would not necessarily rule out horizontal gene transfer between the two altogether, as there is more than one mechanism of achieving this.
Above and below: Centaurea scabiosa is often strikingly rayed.
Andrew Lack carried out an interesting series of studies on British Centaurea and examined competition between scabiosa and the nigra-group, especially with C. debeauxii, for competition for pollinators. Centaurea scabiosa produces more nectar (about 4 times as much, Lack 1981b) and with its rays and generally larger flowerheads is likely to be more attractive to pollinators. However, in most regions of southern England where the two species coexist, C. debeauxii is also rayed (but not generally in the southeast interestingly) and Lack suggested that this may be due to competition with scabiosa or perhaps it is mimicking scabiosa. This idea of mimicry is plausible since C. debeauxii has a late flowering time which peaks slightly later than scabiosa so pollinators that have learned to forage on scabiosa may then carry on foraging debeauxii, even though it produces less nectar, especially when the two look similar. Centaurea nigra usually grows apart from C. scabiosa has a much earlier flowering time and debeauxii may have evolved later flowering to reduce competition with scabiosa (Lack, 1981a). Bees are the most important pollinators of both.
However, the tubular flowers of Centaurea scabiosa are longer than those of the C. nigra group which will favor those pollinators able to reach the nectar, perhaps avoiding a certain amount of pollinator competition.
Above: the Chalkhill Blue butterfly (Polyommatus coridon) alighting on Centaurea scabiosa and, below, on Centaurea debeauxii. Centaurea scabiosa is visited by bees such as Apis and Bombus, flies, beetles and wasps, but bees are considered to be the most important pollinators.
Seed Dispersal - wind and ants
Above: achenes of Centaurea scabiosa have a white, brown or black pappus consisting of 3 rows of filaments. The innermost ring of short blade-like filaments typically angle in to form a cone around the foot of the style which remains as the stylopodium (foot of style and remnants of nectary) in the center. The central ring forms the fan of long filaments and the outermost row form a fan of shorter filaments on the outside. This style is well-adapted for wind dispersal and also for 'walking upwind' when on a substrate.
Above: the achenes also have a well-developed elaiosome (yellow) surrounded by bristles. The elaiosome is nutritious and these structures generally attract ants who bury the achenes in their nests to eat the elaiosome when needed, leaving the achene in a suitable place to germinate. The function of the bristles around the elaiosome is uncertain, but perhaps they are a tactile signal to the ants to help them locate it.
Above: the phyllaries open widely in dry weather when the achenes are ripe, but will close again in high humidity, such as in overcast weather. The pappus also responds with similar hygroscopic movements, opening in dry weather and closing in damp weather. Clearly the pappus can not work in the rain so this makes sense. The open phyllaries provide an open pltaform from which the achenes are dispersed by the slightest breeze. However, the achenes of this species are heavy for their pappus and are only carried a short distance on average. Note the loose white hairs, these are the receptacle scales (see introduction to Asteracea) which detach easily so as not to obstruct the dispersal of the achenes (and perhaps contribute to the lightness of the mass and its ability to catch the wind).
Above and below: the phyllaries continue to open and close even when all the achenes have been dispersed (hygroscopic movements such as these require no active supply of energy by the plant and will continue when the shoot is dead). Note this dispersal mechanism is quite different from the shaker mechanism in Centaurea debeauxii and Centaurea nigra.
The attachment scars where the achenes were attached can be seen on the receptacle surface. one achene still remains in the capitulum above. Some of the detached receptacle scales can be seen as whitish hairs, these surround the developing achenes but by detaching they offer no obstruction to achene dispersal and may even help catch the wind as the whole mass blows away.
Centaurea cyanus (Blue Cornflower)
All the florets in the flower heads of Centaurea cyanus are disc florets, in which the petal tube is not extended into a strap-shaped ligule, not even in the peripheral florets and so lacks true rays. Nevertheless, the petal tube is quite extravagant, being elongated and widened into a frilly funnel (with 5 to 8 lobes) which is larger in the sterile outer florets, forming radiant heads with pseudorays. Towards the center of the disc the petal-tubes become smaller and less extravagantly developed. There are about 30 florets in total (20 - 50). The phyllaries that form the involucre (cup forming the base of the flowerhead) resemble those of Centaurea scabiosa, with dark tips bearing teeth, but the whole flowerhead is generally smaller.
Found in grassland, open woodland and on disturbed ground, C. cyanus is an annual native to southern Europe. It is partially self-incompatible. The fruit are dry single-seeded achenes each with a pappus of bristles up to but generally much shorter than the achene itself. Such a pappus is likely most useful in attachment to animal fur/feathers for dispersal. It is invasive in much of the northern hemisphere and parts of the southern and is often derived from cultivated forms.
The flowers are generally blue, but may be white, pink or purple.
Centaurea montana (Mountain Centaurea)
