The history and distribution of the bulbil-bearing Ficaria verna subsp. verna (Ranunculaceae) in Britain

The distributions of the subspecies of Ficaria verna Huds. are under-recorded in Britain, but published county Floras indicate that subsp. verna is an introduction in some areas and is increasing in many areas. This evidence is examined and a hypothesis to account for the observed patterns is tested against an analogy with Allium paradoxum (M. Bieb.) G. Don in Berwickshire where the author has personally observed colonisation over forty years. The results suggest a link between the recent increase in Ficaria verna subsp. verna and the nineteenth century expansion of the horticultural trade. Earlier waves of colonisation are suspected but not substantiated.


Introduction
I have been intrigued by circumstantial evidence in published county Floras that, in some parts of Britain, Ficaria verna subsp. verna is increasing and may be an introduction, unlike the native subsp. fertilis.
My home is a former farmhouse at Clarilaw, Roxburghshire (v.c.81) which was built in about 1860 on the site of earlier farmhouses, in a sequence which may date back to around 600 AD. The naturalised plant introductions that are thought to have been deliberately introduced at an early date include Chenopodium bonus-henricus as a pot-herb and Myrrhis odorata as a flavouring. Unintentional introductions include Allium paradoxum, probably brought in with herbaceous plants about 1970 (before my wife and I bought the property). Unsurprisingly there is Ficaria verna in the garden and I was aware, when weeding, that at least some of the plants produced bulbils, but I had never given the matter further thought.
I have now found that all the Ficaria verna in my garden and nearby is subsp. verna and suspect that it is an introduction here, so I have set out to investigate whether this is likely and, if so, how and when it came to be introduced.

History
The earliest record known to me of bulbils in British Ficaria is in Gigi Crompton's historical Cambridgeshire online-flora (Crompton, 2001) as 'subsp. bulbifer, Cambridge, CC Babington, 26.5.1839, det. PD Sell, CGE'. This is picked up in Babington's Manual of British Botany (Babington, 1843) as 'St[ems] 3-8 in, long, weak, often producing bulbs in the axils of its leaves'. Despite the Manual being widely used, British botanists were slow to record the presence of bulbils, perhaps because Babington did not give the bulbil-producing plants any separate taxonomic rank.
Babington was such a discerning botanist that I do not believe he would have omitted to suggest the possibility of the bulbil-producing plants being an introduction if he had not been familiar with them over a wider area than the city of Cambridge. Similarly at least some other discerning botanists in Britain would have drawn attention to the bulbils had they considered them unusual. One botanist who did comment on the bulbils was James White in his The Flora of Bristol (White, 1912).
He noted that 'axillary bulbils are not of very rare occurrence with us' and goes on to describe in exemplary detail a luxuriant plant gathered in a sheltered lane.
The situation in Britain contrasts with that on the continent of Europe where subsp. verna, a tetraploid, is much more widespread than subsp. fertilis, a diploid, which is found mainly across a broad strip along the Atlantic fringe. Indeed subsp.
fertilis, but not subsp. verna, is very rare as close to Britain as Belgium, so, before proposing subsp. verna as an introduction everywhere in Britain, it would be necessary to discuss its likely history in Belgium and adjacent countries. If Ficaria bulbils were dispersed in plant containers, might they not have also been dispersed with the bare-rooted hedging plants (chiefly Crataegus) that were despatched by the thousand from plant nurseries to enable the massive hedgeplanting programme that was integral with the enclosures of the Agricultural Revolution from around 1750? It is certainly possible, but there is no evidence from various Floras (see below) to support it.

Hypothesis to account for colonisation
It is evident that if Ficaria verna subsp. verna is brought to a garden, it can propagate itself freely and be dispersed within a confined area by soil disturbance. It is another matter for it to disperse more widely. In considering such dispersal, I have sought an analogy with a more closely-studied species.
Testing the hypotheses against the colonisation of Berwickshire by Allium paradoxum I have chosen the bulbil-bearing Allium paradoxum in Berwickshire to test the hypothesis because I have witnessed its ongoing colonisation there over forty years and because a 1 km distribution map ( Fig. 1) is available (albeit based on sample surveys).
Allium paradoxum has very much the same ability to propagate, by bulbils in the inflorescence and bulb offsets underground, as Ficaria verna subsp. verna has with its leaf-bulbils and root-tubers. Moreover, both taxa are shade-loving and occupy similar habitats. After fifty years the Allium in my garden has only colonised an area of about 0.25 hectares, though it is dominant under trees. It has failed to cross the drive to another wooded strip. Footwear seems very ineffective as a means of dispersal. This year, 2020, it has for the first time appeared under a hedge 100 m down the road, probably from a bulbil picked up by the tread on a car tyre and soon ejected. It must now be expected to colonise that hedge bottom. In Berwickshire, following its presumed introduction as a weed with garden plants a little before 1950, it has been spread only slowly along roads by bulbils caught in the treads of car tyres, colonising hedge bottoms, and as a weed in herbaceous plants passed from garden to garden, so that colonisation of these habitats remains modest. Meanwhile, slowly but inevitably, the bulbils have reached the rivers in a number of places and have been spread down their flood plains by winter floods. This has enabled massive colonisation of riverside woodland. In about 70 years the colonisation of the rivers is almost complete and a number of burns have also been colonised. I used to think that only the parts of the woodlands on Records 1987-99 Records 1970-86 Records pre 1970 Allium paradoxum (Few-flowered Garlic) NT NU  Webster, 1977), Arthur Chater's Flora of Cardiganshire (Chater, 2010), Paul Green's Flora of County Waterford (Green, 2008), Geoffrey Halliday's A Flora of Cumbria (Halliday, 1997)  Arthur Chater has written an extensive account of Ficaria verna in his Flora of Cardiganshire. He has acquired the enviable ability to distinguish the two subspecies vegetatively from a distance, utilising differences in leaf texture and mat-forming capability. He considers both subspecies to have increased between 1995 and 2005 in two urban and suburban areas around Aberystwyth, but that subsp. verna has increased much more than subsp. fertilis. He estimates a thousand-fold increase at these localities. This suggests a particular dispersal event (such as the contamination of council grass-cutting machinery) followed by bulking up. I have observed such a phenomenon with Allium paradoxum where bulbils have been sparsely scattered over an area of riverside woodland by floods and have bulked up over a decade. Chater (2010) found subsp. verna to be much more localised than subsp. fertilis and to be found chiefly near habitation and along river systems but to be rare in ancient woodland and other wild sites and to be quite absent from the coastal slopes. This contrasts with the situation on the continent where subsp. verna is especially characteristic of ancient woodland. Thus the distribution pattern in Cardiganshire is again consistent with my hypothesis.
Paul Green's Flora of County Waterford is also informative. He finds subsp. verna (as Ranunculus ficaria subsp. bulbifera) to be much the rarer of the two subspecies though he observes that it can often form dense patches under trees. He states that it is often found on dumped soil, which is aiding its spread to some of the more remote parts of the county. Once again there is evidence of ongoing colonisation.
The shaded habitats, chiefly in the east' [and thus away from the coast]. These are usually, but certainly not always, anthropogenic, for example gardens, where it can be a pernicious weed, and by woodland paths'. The distributions in Cumbria are strikingly similar to those for North Northumberland, though, in the absence of historical records, no comment is made on any apparent increase. It has also been instructive to search for pre-1950 herbarium records of both subspecies in the BSBI database and the herbaria@home website (http://herbariaunited.org/atHome/). There are records in the database of twenty such specimens of subsp. verna from BM (the Natural History Museum). While most are from the south and east of England, there are specimens from Northwest Yorkshire, East Perthshire and the settlement of Tarbert on the island of Harris in the Outer Hebrides. There are a similar number of specimens for subsp. fertilis. The specimens retained in the herbarium are presumably a selection aimed to cover morphological variation and geographical distribution, so I do not think any conclusions can be drawn of the relative frequencies of the two subspecies. There are very many specimens for Ficaria verna in herbaria@home but only two are determined to subspecies. I have examined a selection of the others and find most to be indeterminate at subspecies level, often because they were collected early in the season. I did notice two more subsp. verna specimens as they are much easier to identify than subsp. fertilis, for which I did not notice any further examples, though my best guess was that most of the remaining specimens were subsp.  (Simpson, 1982) considers subsp. verna (as subsp. bulbifera) as the less frequent subspecies favouring shady habitats, while Martin Sanford's A Flora of Suffolk (Sanford & Fisk, 2010) also considers it the less frequent subspecies but states that it is usually found in more disturbed sites, often associated with habitation or churchyards. This suggests recent expansion of the subsp. verna population in anthropomorphic habitats. In Hampshire, however, Martin Rand (pers. comm.) states that 'I thought a long while ago that it might be possible to make some distinction between habitats [of the subspecies], but the more I've looked, the less convinced I've become' (pers. comm..)

Discussion
It is as well to remember just how widespread and abundant both subspecies are across almost all of England and some parts of Scotland and Wales, and to reflect that this severely limits any attempt to trace distributional or habitat changes over time. There does seem however to be agreement that subsp. verna is the lessfrequent subspecies and to be the one that is increasing in many areas. This might reflect the increasing level of human activity, but it might possibly indicate a later date of arrival in Britain, with subsp. fertilis a native taxon and subsp. verna an archaeophyte which is still in the process of colonising Britain.
While the horticultural trade is proposed as a key driver of colonisation since 1800, there are likely to have been earlier waves of colonisation, at least in England. The medicinal herb gardens of the monasteries and medieval hospitals are possible sources of introduction, especially as the root-tubers of Ficaria verna were commonly used medicinally. Indeed, I have recorded subsp. verna by the walkway along the massive stone-faced ramparts that are the Elizabethan town walls of Berwick upon Tweed and abundantly in a tiny dean adjacent to the site of a medieval hospital at Dalcove (Berwickshire NT63) destroyed by the English in 1544, where Hyoscyamus niger still grows on an eroding bank.
It is not necessarily a bar to native status to suppose subsp. verna to have had a southern and eastern distribution until recently. Its bulbils do not favour longdistance dispersal (in the absence of human assistance) and it would be but one of a considerable number of native species that have never managed to colonise all the parts of Britain that have suitable habitat and climate. Thus the Cheviot Hills proved a barrier to the colonisation of Berwickshire by Glyceria maxima and Lysimachia vulgaris, but both colonised strongly after their introduction from England a little over a century ago. For short-distance dispersal the bulbils of subsp. verna are probably dispersed in mud by animals as effectively as the achenes of subsp. fertilis.
For long-distance dispersal the achenes have the advantage of being smaller and more robust. In open grassland, particularly on the coast, they may be dispersed by wind. It is to be expected that they are occasionally swallowed by birds, pass through their intestines unharmed and are evacuated at a distance. Such events, though rare, play a key role in the dispersal of many species.

Conclusion
While the early history of the subspecies of Ficaria verna remain ambiguous, there is circumstantial evidence of an increase in the range and abundance of subsp. verna since about 1800. The hypothesis that subsp. verna was dispersed as a weed with garden plants marketed by plant nurseries and gradually colonised more widely is supported by the observations of Flora writers and by analogy to the observed recent colonisation of the similarly-dispersed Allium paradoxum in Berwickshire. It is not suggested that the association between subsp. verna and garden plants has been the only cause of an increase in its range. Rather, it may well be but one of a series of associations with human activity over an undefined period of time.
In view of the long history of my own garden, it follows that the date of arrival of Ficaria verna subsp. verna remains unknown.