Many Australians, especially those on the east coast, would be familiar with the plant genus Bidens, especially the very common and widespread non-native species B. pilosa. Referred to as cobblers pegs, pitch-forks, and shepherd's needles among a number of names, many bushwalkers (especially in more disturbed areas) will have brushed against one of these at some point and had their clothes covered in the annoying achenes. However, B. pilosa isn't the only species present in Australia. Also, Bidens in Australia has seen some taxonomic confusion between similar species.

@bean_ar provided a great overview of the situation in a 2020 edition of the Australian Systematic Botany Society Newsletter (see pp. 44-45 in https://asbs.org.au/newsletter/pdf/20-sep-184.pdf). As a summary of Tony's article, plus information from the various state herbaria, the species currently recognised as occurring in Australia are:

Bidens pilosa. Non-native. Common and widespread across the country, and recorded as occurring in all states and territories (including Lord Howe Is., Norfolk Is., and Christmas Is.) except Tasmania. Many (most?) Australian herbaria/sources recognise two varieties of B. pilosa, var. pilosa and var. minor, with the former meant to be lacking any ligules/ray florets ('petals') and the latter having very small, white, often ephemeral ray florets. Tony notes in the above article that "My field observations in several locations have clearly shown that the non-ligulate form (var. pilosa) and the ligulate form (var. minor) occur mixed within the same population, even side by side. For this reason, I suggest that varieties should not be recognised in B. pilosa." Interestingly, Plants of the World Online also takes this view and does not accept the varieties, however, both are currently accepted and used on iNat as deviations.

Bidens biternata. Native. As noted by Tony, this species was previously overlooked in Australia, being misidentified as either B. pilosa (both pinnate leaves) or B. subalternans + B. bipinnata (all yellow ligules). It occurs across the Kimberley in WA (although Florabase is yet to take up this change), the Top End of the NT, and in QLD, down to within ~10 km of the NSW border based on specimens Tony has looked at.

Bidens tripartita. Non-native. Relatively uncommon in Victoria and NSW.

Bidens aurea. Non-native. Seemingly rare, recorded only from Sydney thus far.

Bidens bipinnata/Bidens subalternans. Non-native. I've included these two species together here because this is a taxonomically contentious pairing. Most sources treat these two entities as separate (but similar) species. However, Tony believes that the two should be collapsed into a single species (see the linked article above for explanation). Florabase has collapsed the two, and so too has QLD, however, NSW and FloraNT still treat them separately. For completeness, here is how the two are separated by the PlantNET key:

Pappus awns erect; peduncles and margins of involucral bracts glabrous or with scattered small septate hairs; leaflet lobes usually linear to lanceolate ---> Bidens subalternans

Pappus awns spreading; peduncles and margins of involucral bracts conspicuously ciliate with spreading septate hairs; leaflet lobes rhombic to broad-lanceolate ---> Bidens bipinnata

Bidens alba. Non-native. This is the species that prompted me to make this journal post. Previously, this species was included within B. pilosa as 'var. albus', although it was often just treated/identified as B. pilosa. Recently, however, it was elevated to full species status (for anyone interested, see discussion here: https://www.inaturalist.org/flags/362565). Unfortunately, most Australian herbaria have not recognised/taken up this change yet. The variety in Australia is var. radiata.

Florabase (WA) - not recognised. There are only 6 observations of Bidens on iNat for WA, none of them alba, so I don't know whether alba does occur in WA.

SA Flora (SA) - not recognised. A brief skim of iNat records immediately shows at least one observation of B. alba, so it clearly does occur here.

FloraNT (NT) - not recognised. There's only a single Bidens observation on iNat for the NT, so I don't know whether alba does occur in WA.

VicFlora (Vic) - not recognised. A brief skim of iNat records immediately shows multiple observations of B. alba, so it clearly does occur here. The main image for B. pilosa on the VicFlora page for that species also actually depicts B. alba.

PlantNET (NSW) - not recognised. Clearly does occur in NSW, with many iNat records. There are also AVH collection records in NSW (all in northern NSW, but it clearly extends all the way down based on my records + those of others).

Queensland Herbarium - recognised, and does occur.

There is also a native lookalike genus, Glossocardia, that is often mistaken for Bidens.

The main reason I made this post is that not a lot of people realise that B. alba is a separate species from B. pilosa, there are misidentifications of alba as pilosa on iNat in Australia, and a few years ago when B. alba was recognised on iNat, all observations of B. pilosa in Australia were actually pushed back to genus, so there are now a lot of easily recognised records sitting at genus because people weren't aware of the change. I'm going to go through the Australian Bidens records and add IDs where I can to clean things up.

For now, I'm going to recognise B. subalternans as a legitimate species in NSW, and ID based on the key. If it does get collapsed into B. bipinnata in future, it will be easy to swap these records into bipinnata.

Most of my IDs will be pilosa + alba, the two common species. Here are the ways to separate them:

QLD Keybase key (Tony's):

Ligules 10-16 mm long; leaves/leaflets with 17-42 pairs of teeth; teeth 0.5-1.3 mm long on large leaflets; leaves 1-3 (rarely 5)-foliolate --> alba

Ligules 2-8 mm long or absent; leaflets with 8-20 pairs of teeth; teeth 1.1-4 mm long on large leaflets; leaves 3-7-foliolate --> pilosa (and biternata, but that has yellow ligules, and alba always has white)

Weakley 2020 and Ballard 1986 (US sources):

Ray florets 5-8, the ligule 3-18 mm long; cypselas 0-2-awned, the awns 1-2 mm long; outer phyllaries (8-) 12 (-16) --> alba

Ray florets absent (or if a few present, the ligule is only 2-3 mm long); cypselas 3 (-5)-awned, the awns 1-3 mm long; outer phyllaries 7-10 --> pilosa

CONABIO (Mexico):

Up to 8 ligules that are notably longer than wide. Relatively short external (green) bracts --> alba

Absent or reduced ligules. At least some fruits with three awns instead of two --> pilosa

So there are discrepancies + overlaps between the keys re the differences in ligule length. What we can take home regarding pilosa and alba (ignoring the other species for now):

If flowers are present, and there are no white ligules, only yellow disk florets, it is B. pilosa. A good example is here: https://www.inaturalist.org/observations/155421971

If ligules are present, and they're white (if they're yellow, it's not pilosa or alba), and they're very small, it is B. pilosa. A good example is here: https://www.inaturalist.org/observations/120001760

If ligules are present, and they're white, and they're large, it is B. alba. A good example is here: https://www.inaturalist.org/observations/124345528

Now of course the tricky part comes when it's difficult to gauge ligule length from photos, or if plants are non-flowering. This is where ideally you would use the vegetative characters from Tony's key, although unfortunately the two species can overlap on all three leaf characters for some values/measurements.

I'm going to go through the Australian observations, ID all the easy ones first, then go back and look more carefully at the 'edge' cases.

Tagging some people who might be interested:
@bean_ar @nicklambert @gregtasney @aavankampen @scottwgavins @alx4mtmel @russellcumming @wcornwell @cesdamess @jackiemiles @insiderelic @gtaseski @onetapir @reiner @pcopping_ecp @martinbennett

feel free to tag others that I've missed

Posted on April 18, 2023 10:45 AM by thebeachcomber