Pondering Polygonums

Publication in 2010 of the inspiring Flora of Cardiganshire brought with it an account of the knotgrasses (Polygonum aviculare and arenastrum aggregates) that was vastly different from the account of the genus given in Stace. This has always been a tricky group for botanists to get to grips with as the variability in morphology seems at odds with the published descriptions of these broadly defined species.  Flora of Cardiganshire unfortunately did not have enough information to help with identification of the named species, that was not the purpose of this county flora, but while I waited for Sell & Murrell to catch-up, it did start me looking more closely to get a feel for the different forms present in my local area. This in itself was a revelation, with a number of discrete forms becoming apparent and popping up in the same locations year after year.

Lake drawdown habitats - as productive for Polygonum taxa as they are for Persicaria species

Volume 1 of Sell & Murrell subsequently arrived after a long wait, and at last there were descriptions and a key (of sorts, more on that later), ultimately leading to this account on what I think I know about Polygonum. I could be more confident, as I think I have found most of the inland taxa now, but its early days and I need to compare notes with others. Hopefully this blog is a starting point down that road.

So first impressions after a summer of contemplation - I like it. It seems widely accepted that the genus has not been split enough historically (Flora Nordica presents a very similar picture), and there seem to be some distinct entities that are easier to identify at a glance and defend than the overarching variable aggregate taxa (see the following photos). What is interesting about the account in Sell & Murrell is that it re-establishes taxa first proposed by Jordan and  Boreau back in the mid-1800's. In other words the same authorities who established P. arenastrum and rurivagum, taxa we have been more than happy to accept for some time.

I decided to take things slow this year and visit several carefully chosen locations near home, local arable fields and a lake with an extensive drawdown zone, at regular intervals across the season from June to October.  I wanted to give myself opportunity to re-visit and test my preliminary identifications given the challenges of the key, understand how the plants concerned varied across the season, and to understand the best time to attempt ID. I'd recommend this approach to anyone interested in giving this group a try. Its easy to make false starts, but if you choose some good hunting grounds then you should find most taxa quite easily and be able to correct and learn from any early mis-identifications.

So what do I think I have learnt:

1) Don't start too early in the season. The initial growth phase for several erect species seems to be lateral, and these prostrate shoots will flower. For example, Polygonum monspeliense will erroneously key to the coastal neolittorale before it develops its distinctive main stem. Once it has the latter it is unmistakable.
2) Mid-July to end of August seems optimal for recording all taxa. They are at peak growth and have established their mature morphology.
3) From September onward it is still possible to record, especially in late sown crops e.g. sugar beet where plants will still be fresh, but you need to be mindful that plants will be increasingly tatty and clothed in mildew, they may have a second growth spurt that dilutes there distinctive morphology (similar to, for example, Taraxacum; so not a problem specific to this genus, some taxa have narrow seasons), and environmental influences can influence morphology and/or flower colour (some of this is revisited below).

Does the key in Sell & Murrell work well. I would say yes and no, and its another reason why it is good to familiarise yourself with as much material as possible before fixing identifications of specific plants. This seems to be a genus that is exceptionally hard to build a key for given the morphology of the species concerned and the plasticity that can result from growing conditions. For example, a plant on moisture retentive or nutrient-rich ground will likely grow larger or more etiolated than one on less favourable ground, making it harder to build a definitive key. This does not in any way undermine the validity of the taxa which can usually be readily distinguished by eye. The key seems to use those traits that are most definitive based on optimal/typical material. But this may not be what you find in the field, so its really important to review the full description and then interrogate other options if this does not seem a good match. I found populations of small Polygonum monspeliense quite widely and in the early days mistakenly took them for P. agrestinum. I only realised my mistake when (a) noticing the plants were otherwise identical to good material of the former, and (b) finding the latter in the stubbles of recently harvested arable fields.

I look forward to picking this genus up again next year and looking further afield.

The species as I understand them follow below (P. rurivagum is mainly a southern species so is not included in this account based on West Yorkshire). I hope this is of help to others:

Common Knotgrass Polygonum aviculare sens. str.

This erect species is widespread, but not necessarily abundant, in a wide-range of habitats. It is often to be found in field gateways and at the edges of paths, where it can form dense stands. Very distinctive if narrowly defined. It's not very heterophyllous for an aggregate species defined by this characteristic! It is also one of the few species easier to identify earlier in the year (June to early-July) before it runs to flower and becomes more elongated. Until then it has a relatively dense cover of relatively uniformly sized large leaves.



Broad-leaved Knotgrass Polygonum monspeliense

For such a large, distinctive and widespread species I found it surprising that I had overlooked it consistently till now. It seems inarguable as a species. It is present in arable fields (often with at least two other species), on drawdown zones, and at the edges of roads. In the latter habitat it is typically much smaller but retains the same distinct jizz. It is strongly heterophyllous, with a small number of large broad leaves contrasting with a number of very small leaves, but the foliage is relatively sparse given the long internodes. It has a tall erect stem with long branches, sometimes flopping over late in the season. Early in the season it starts as a prostrate plant with a low number of robust branches at ground level, before throwing up the single erect main stem.

Note that for the photograph the lowest branches have been swept up adjacent to the main stem but were originally prostrate, they are comparable in length to the latter.

A smaller plant with otherwise comparable morphology. It would likely key using Sell & Murrell to P. agrestinum, but is too heterophyllous for that species.

Typical habit and habitat. The best that can probably be said of this photo is that it shows how easy it can be to overlook this species amongst arable crops when it is not being specifically searched for

Creeping Knotgrass Polygonum chamaechyton

This species is strictly prostrate. It is the grey-colour of the plant, its long internodes and typically rather chunky ridged stem that give this species its jizz. The foliage can be rather variable in size/width between plants/populations, but the species otherwise seems relatively straightforward. It is abundant in arable fields, and other bare ground. The flowers, especially in bud, are typically a brilliant rose-pink.

Arable Knotgrass Polygonum agrestinum

Common in arable fields and urban habitats. An erect to ascending species with more than one main stem, weakly heterophyllous, the leaves are rather narrow and relatively small, The general impression is a sprawling plant with intertwining stems. Quite easily overlooked amongst cereal stubble and associated arable flora, including P. aviculare, chamaechyton and monspeliense.

This is the same plant as photographed above. It looks rather different out of context.

Naked Knotgrass Polygonum denudatum

This is the one that seems least plausible given it is such an oddity. I have only seen it twice on heavy ground, once on a drawdown as a small colony and once as a single large plant in a sugar beet field. It doesn't appear to have been defoliated by insects and the absence of leaves on the main stem contrasting with the tufts at the tips is rather striking. It is a prostrate plant, and I assume it is more leafy earlier in the year. If so it provides a further note of caution against attempting identification of Polygonum species too early in the summer. I'll be happier about this one once the weight of records and images is there to provide greater confidence in its existence, but I am not in position to disagree with its validity given the history behind the name and my limited experience. I look forward to seeing more of this quirky plant.

Moving on to the Polygonum arenastrum agg.

Bushy Knotgrass Polygonum polychnemiforme

This seems to be the most common species, present anywhere there is compacted ground. It is certainly the commonest species in the Yorkshire Dales where it is in virtually every gateway. It can get lankier as the season goes on, especially where there is enrichment, but otherwise can be recognised at a glance. It forms dense cushions of prostrate stems with ascending tips with dense leaves.

Dense-leaved Knotgrass Polygonum arenastrum sens. str.

This is the one that bothers me most, as I don't believe I have seen this prostrate species. Early in the summer I thought it was present along the edge of paths in my local park, but by the optimal recording season most, bar perhaps one or two plants, had turned into polychnemiforme. Come the autumn the top growth was lost again leaving plants resembling arenastrum. I would be over-reaching myself if I was to imply arenastrum does not exist as a species distinct from polychnemiforme, it may simply be that it is not common this far north. Its something I need to look for in VC31 next year. I am reserving judgement for the time being, on the expectation it will turn up when I look more widely.

Small-fruited Knotgrass Polygonum microspermum

This is a striking prostrate species with small distinctly blue-tinged foliage. Seen as a population, as I did this year in its hundreds, it seems very distinct. The fruit and flowers are very small, with the characteristic short white-bordered perianth lobes scarcely distinct from the tube and not reliably reflexing. One note of caution, as I found this year in my population, late in the season the perianth can take on a deep rose-pink hue. Flora Nordica is very useful for supplementing Sell & Murrell on this species.