Plant of the Week – 5th April 2021 – Few-flowered Leek (or Few-flowered Garlic) – Allium paradoxum

If you live in Lowland Scotland you will be very familiar with this plant, flowering now (late March to end of April). Figure 1 below shows a pungent carpet of it spreading as far as the eye can see on the slopes above Wadingburn near Lasswade. Most people will recognize it as wild garlic by its smell, but many do not know that it is not our native garlic, A. ursinum, or they may confuse it with A. triquetrum, which is more common in the South. Strictly speaking what we are talking about is Allium paradoxum var. paradoxum. It can be ID’d easily. The single leaf is narrow and linear, and the plant has few flowers, as the common name suggests. Many Alliums have aerial bulbils, but the green, shiny, nutty ones of A. paradoxum var. paradoxum, borne in sequential spathes, seem distinctive.

The species is currently placed in the Amaryllidaceae, but it was not always so. Earlier floras placed in in the Liliaceae, but the views of taxonomists have wavered over the last decade; Stace variously placed it in the Liliaceae (2nd edition), Alliaceae (3rd edition) and finally Amaryllidaceae (4th edition).

Figure 1 A. paradoxum in Wadingburn, near Lasswade (photograph © Richard Milne)

Allium paradoxum var. paradoxum was introduced to the UK from the Caucasus, where it is native. It was first recorded in the wild near Edinburgh in 1863 (Balfour, 1866). Since then, it has spread widely around the Lowlands of the British Isles (Figure 2). When I was an undergraduate, over 50 years ago, I carried out a small undergraduate project on this species with my supervisor, Dr Adrian Dyer of the Botany Department in Edinburgh University. Sadly, I have not retained the notes, so my memories will have to suffice. An informative article on the species was published by Dyer (2017) in our BSS Newsletter, but I thought a lockdown blog might deliver a wider awareness. Dyer called it “the Beast”, in contrast to its more ornamental cousin Allium paradoxum var. normale, which he called “the Beauty”.  I mention the latter briefly later in the blog, but I will mostly be referring to ‘our’ familiar weed, Allium paradoxum var. paradoxum.

 I remember the small and rather precious patches of it in the Hermitage of Braid. Adrian said, it’s spreading, it’s invasive. You will see how it will become a problem in years to come. And how right he was. In 2005 the species (and I think they mean Allium paradoxum var. paradoxum) was added to the Schedule 9 list of Wildlife and Countryside Act 1981 which lists non-native invasive species which it is an offence to plant or cause to grow in the wild (Wildlife and Countryside Act 1981). Figure 2 shows its current distribution, and Figure 3 shows its distribution in the 1960s, which is when I first developed an interest in this species. The large clump of records around Edinburgh corroborates Balfour’s observation that Edinburgh was the place of its first sighting in the British Isles.

Dr Henry Noltie, an expert on monocots, provided some interesting historical vignettes to supplement those of Dyer (2017). The species was first described as Scilla? paradoxa by Marschall von Bieberstein (1819) in his Flora Taurico-Caucasica dealing with the Crimeo-Caucasian region.  Why choose ‘paradoxical’? It could be that Bieberstein might have found it strange for something he thought of a Scilla to smell of garlic – to say nothing of the bulbils. Curiously, the details Bieberstein gives are ‘In Iberia D[ominus] Wilhelms’. When George Don (son of G. Don one-time Superintendent at the Royal Botanic Garden, Edinburgh) transferred it to Allium in 1827 (in his Allium monograph in the Edinburgh Memoirs of the Wernerian Nat Hist Soc) he took this to be a misprint for ‘Siberia’. Iberia is in fact not Spain or Siberia, but a region in the Caucasus, currently Georgia. But who was Wilhelms? Don says it was introduced a few years before 1827 by Mr Otto (i.e. C.F. Otto, Inspector of the Berlin Botanic Garden) as a gift to Mr William Anderson, Curator of the Chelsea Physic Garden who was born in 1766 at Easter Warriston, Edinburgh, i.e. very close to the present RBGE (but less close to Leith Walk, where the RBGE was situated at the time). Did he bring it home to his parents’ garden, or plant it in the RBGE before or after it moved to Inverleith in 1820?  But why not Chelsea? Anyway, by the 1860s, the species was already extensively naturalised in several sites within 12 miles of Edinburgh centre (Dyer, 2017).

Figure 2 (left-hand side) shows current distribution – green squares are from 2020 onwards. Figure 3 (right-hand side) shows distribution 1950-1969. Maps © BSBI from

The species spreads vegetatively by bulbils. These are produced within an umbellate inflorescence, which usually consists of a basal spathe (made of 2 hyaline bracts) subtending several bulbils, one or sometimes two perfect white flowers on slender pedicels and one or more peduncles bearing secondary spathe-covered inflorescences with more bulbils.  Flower and bulbil number per plant are variable, depending at least partly on the size of the bulb and the accumulation of assimilates. The flowers are hermaphrodite with 6 tepals, 6 epipetalous stamens, a trilocular superior ovary with a trilobed stigma. A small percentage of flowers have 4 -locular ovaries. Flies and bees buzz around, so the potential is there to achieve cross pollination.

The plants have one narrow leaf, about 30 cm long; the flower stem is triangular is section. The plants are perennial and mature plants have small round bulbs about 2cm in diameter buried in soil 10cm deep; they bear more bulbils and flowers than the ‘youngster’ plants produced from last year’s bulbils which germinate on the surface of the soil (Figure 7).

Seeds are rare, though I did find a few when I did the project in the 1960s. I remember making thin microtome sections of the ovaries embedded in wax, and they were perfect. In the flower primordia of Babington’s Leek the carpels develop as three U-shaped swellings on the surface, which grow towards the centre, where their in-turned edges meet and fold to form the ovules – two in each carpel (Harding, 2004). From memory of my undergraduate project that is exactly what happens in A. paradoxum too.

So, the question which still remains unanswered is if ovule development proceeds normally, why are seeds rare, and if and when they do occur, are they in fact viable? I do not think I ever tried to grow them. I would like to urge readers of this blog to look out for seeds in early summer, collect some and maybe try some viability tests – bearing in mind that vernalisation may be required. However, you must be careful not to introduce the plant into your garden. If you do, it will take over, and you will curse me.

I know the plants from the Hermitage of Braid population are diploid (2n=16), with lovely chromosomes, which I counted both in root tips and in pollen grain mitoses.  Barling (1971) reports evidence of chromosome inversions, detectable as bridges at meiosis, which might account for the low fertility. It could also be that low seed fertility may arise automatically, not owing to any special lethal factor in the gametes or zygote but to bulbil competition for nutrients (Gustafsson, 1947 p. 296). Experiments, carried out in several species of Alliums by Levant (1937) which showed that the removal of bulbils promoted seed production, would seem to support this deduction. However, if there is plenty of land to colonise, why would you want to alter the genetic recipe through meiosis and fertilisation, especially as the seeds are small, and it may take several years to produce a bulb mature enough to flower. Retaining the ability to produce seed through outcrossing is nevertheless a useful strategy for times of environmental stress. Even a small number of sexual individuals per generation is sufficient to make an apparently asexual population highly genotypically variable (Bengtsson, 2003). Combining the two reproductive methods in varying proportions gives the populations the best of both options, offering the opportunity to adjust the balance between them, with the environment playing the part of switch-hitter. We do not know whether this happens in A. paradoxum, but it would make evolutionary sense.

Figure 6 Close up of this population shows that most plants have one perfect flower. Photograph © Chris Jeffree

Although the term apomixis is usually taken to mean clonal reproduction through seeds, some people use the term vegetative apomixis in plants in which flowers are replaced by bulbils. A. paradoxum and other bulbil bearing Alliums fall into this category.  Flora Europaea (Tutin et al., 1993) lists twenty species/varieties/subspecies, which produce bulbils. Bulbil development was examined in several species of Alliums, including A. paradoxum by Cottrell (1999).  With the help of a low vacuum scanning electron microscope and light microscopy he showed that bulbils arose as domes on the inflorescence meristem and that early inflorescence bulbil primordia were indistinguishable from floret primordia, only becoming distinguishable later when approximately 0.4mm in diameter and 0.38mm in height. At some stage in the developmental process of the umbel the genetic determination to produce a bulbil or a flower or a secondary inflorescence must be made, with the environment or maybe resource availability deciding on which pathway to follow. It would be interesting to research how this happens.

 Of course, bulbils are heavier and less likely to be dispersed far and wide than are the tiny, shiny seeds, but widespread bulbil dispersal is now made easier with modern agricultural/horticultural practices which provide both the dispersal mechanism and disturbed ground for easy colonisation. The plant is commonly found on riverbanks, is distributed by water and colonises flood plains. In the summer, when no parts of the adult plant remain overground, one can see thick scatters of the bulbils lying on the surface of the soil (Figure 7). As the plants mature over several years, their bulbs become buried deeper in the ground, its dense carpets smother and outcompete native species, including Allium ursinum, Primula vulgaris and Adoxa moschatellina which live in similar habitats. Eradication is possible if you start to dig up the plants in very early spring before a new crop of bulbils is produced. That is possible (but difficult) in small private gardens, but in the wild, such a measure would also eradicate the very plants you wish to promote. The best course of action has to be prevention. Here on our urban Edinburgh street many gardens are infested with the plant, but happily we don’t have it (yet). I told my neighbours to make use of the plants they remove for pesto. They even deposited a bag of the plants in my front garden for me to make some as well. Ouch!!!!! I hope none of the bulbils were lost on the way to my kitchen…., but the pesto was good (recipe from BBC Good Food, 2015). Bulbils and flowers also formed a delicious and decorative garnish to a mixed salad. 

Figure 7 Germinating bulbils of A. paradoxum with Lathraea clandestina Photograph © Chris Jeffree

In Iran A. paradoxum (locally called “Alezi”) is grown as a cultivated vegetable and spice, especially in Mazandaran province, where it is used to prepare a variety of local and special foods. Studies by Iranian researchers have shown that extracts of the plant are rich in iron, manganese and antioxidants and have marked antihemolytic effects (Ebrahimzadeh et al., 2010).  It has also been found it to be hepatoprotective (Nabavi et al., 2012)and anti-toxoplasmotic (Ebrahimzadeh et al., 2017).  Rezaee et al., (2018) found it to contain saponins, which have leishmanicidal activity. Several ‘nature cure’ sites also cite that the juice of the plant is effective as a moth repellent. Maybe worth trying to deter out clothes moths?

Gardeners often grow, the non-bulbil-forming variety, Allium paradoxum var. normale Stearn, which Adrian Dyer called “the Beauty” (Figure 8) in his excellent Newsletter article (Dyer, 2017).   It originates from the wild from Iran’s Goubad Province, Gholidagh (Stearn, 1987; Pacific Bulb Society, 2019). It has more flowers per umbel (3-17), is more ornamental, equally edible, and not so invasive. It produces seeds which have an oil-bearing appendage (elaiosome) which is attractive to ants. The ants carry the seed away to eat the oil and then discard the seed, thus aiding dispersal of the plant (Plants for a Future, no date). I am not sure whether seeds of ‘our’ plant which we must now call A. paradoxum var. paradoxum, if, and when they occur, also bear such appendages.

When I submitted the first draft of this blog to our editors, John Grace wondered whether the Schedule 9 included var. normale.  It appears to get no mention. Adrian Dyer (pers. comm.) thinks that a major reason for this is that few people know of its existence and even fewer grow it, and those only in recent years. As a consequence, it hasn’t yet been reported as a naturalised alien in the wild, but the potential may be there, particularly as it is now available by mail order from nurseries. It’s only a matter of time before it too escapes, like the better known non-bulbil forming A. triquetrum.

Figure 8 Allium paradoxum var. normale in Adrian Dyer’s garden. Photograph ©Adrian Dyer

Much work remains to be done on this interesting and rather understudied species. The questions which I would like to ask are:

  1. How often (if ever) does the wild Allium paradoxum (var. paradoxum) set seed?
  2. If it does, are the seeds viable?
  3. If you find seed have a look whether it has any appendages for ant dispersal.
  4. If you have molecular biology facilities at your disposal (you may be embarking on a research project for an undergraduate or Master’s dissertation), it would be interesting to collect bulbils from around Scotland and see whether they all belong to a single clonal population. Bulbils are present and easily sampled throughout all the seasons, so there would be no shortage of DNA.
  5. Look out for Allium paradoxum var. normale in the wild. There is at present only one dot (near Manchester) on the BSBI map, but if it becomes more common, then steps might have to be taken to control it.

I shall leave you with a quote from William Shakespeare: “Mine eyes smell onions: I shall weep anon.’’ All’s Well that Ends Well.

Maria Chamberlain


Barling, D. M. (1971). Studies on Gloucestershire populations of Allium paradoxum (Bieb.) G. Don. Watsonia 8, 379-384.

Barling, D. M. (1958). Reproduction and cytology of Allium paradoxum. B.S.B.R. Exhibition Meeting 1957. Proc. Bot. Soc. Br. Isl., 3: 85.

BBC Good Food (2015). Wild Garlic pesto (Retrieved 25/03/2021)

Bengtsson BO (2003) Genetic variation in organisms with sexual and asexual reproduction. J Evol Biol 16,189–199. 

Bieberstein, Marschall. von., F. (1819). Flora Taurico-Caucasica 3, 267-268. in Latin, as Scilla paradoxa.

Cottrell VM (1999). Vegetative reproduction by bulbils in Alliums. MSc thesis, Coventry University and University College Worcester.

Gustafsson. A. (I947). Apomixis in higher plants Ill. Biotype and species formation. LUl1ds Univ. Arsskr., 44 (2): 183-370.

Dyer, A.F. (2017). Illustrations of non-native invaders -6 – Allium paradoxum (M.Bieb.) G. Don (Amaryllidaceae). BSS News. 109, 45-53.

Ebrahimzadeh, M., Nabavi, S and Eslami, B. (2010). Antihemolytic and antioxidant activities of Allium paradoxum. Central European Journal of Biology 5, 338-345.Ebrahimzadeh , M (2017). Anti-Toxoplasma Effects of Methanol Extracts of Feijoa sellowiana, Quercus castaneifolia, and Allium paradoxum. J Pharmacopuncture.  20, 220–226.

Harding, S. (2004). Inflorescence Development in Allium ampeloprasum var. babingtonii (Babington’s Leek). PhD thesis, Cardiff University.

Levan, A. (1937). Cytological studies in the Allium paniculatum group. Hereditas, 23, 317-370.

Nabavi, S. et al., (2012). Hepatoprotective activity of Allium paradoxum. Eur Rev Med Pharmacol Sci 16, 43-46.

Pacific Bulb Society (2019). (retrieved 25/03/2021).

Plants for a Future (no date) (retrieved 25/03/2021).

Rezaee F., Zolfarghari, B. and Dinani, M.S. (2018). Isolation of dioscin-related steroidal saponin from the bulbs of Allium paradoxum L. with leishmanicidal activity. Res Pharm Sci.  13, 469–475.

Stace, C. (1997, 2nd edition; 2010, 3rd edition; 2019, 4th edition). New Flora of the British Isles. Cambridge University Press.

Stearn, W.  (1987). A paradoxical onion, Allium paradoxum var. normale. Bot. Mag. (Kew Mag.) 4, 196

Tutin, T.G. (ed) (1993). Flora Europaea Vol 5. Cambridge University Press.

Wildlife and Countryside Act 1981. Schedule 9 Part 2. (retrieved 25/03/2021)

One thought on “Plant of the Week – 5th April 2021 – Few-flowered Leek (or Few-flowered Garlic) – Allium paradoxum

  1. I moved to Egremont on the Wirral a few years ago and this wild garlic grows merrily alongside roads and in a local church garden.Its delicious and I have been growing in pots.It certainly is vigorous.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: