Hedera in Australia

Hedera is a relatively small genus (there is disagreement between sources as to how many species, with some taxa being relegated to subspecies in some sources, but for what it's worth POWO currently lists 19 species and a hybrid) of climbing plants native to northern Africa, Europe, and then stretching in a band across Asia as far east as Japan.

For those outside the native range of Hedera (but also for many within), the most well-known species is Hedera helix, the 'English ivy' or 'common ivy'. It is widely cultivated across the world, and is commonly planted to cover walls, fences, and other vertical surfaces. Unfortunately, it is now also widely naturalised as an invasive species around the world, readily escaping cultivation into nearby areas.

Like elsewhere, this situation seems to very much be the case in Australia. As I write this, there are ~900 iNat records of Hedera in Australia identified as H. helix (without having looked at them all yet, I strongly suspect a large % of these represent specimens that are planted rather than naturalised populations and need to be marked as cultivated, but let's assume for now they're all wild), and the AVH currently holds 157 collections identified as H. helix from Australia. In almost any suburb in Sydney, it is fairly easy to find at least a few properties or public spaces with H. helix planted over walls or fences, and indeed Hyde Park in the city has large sections where planted H. helix entirely blankets the ground. I am also increasingly seeing this species escaping cultivation in Sydney and invading nearby bushland.

However, all is not as it seems. Enter, Helix hibernica ('Atlantic ivy', 'Boston ivy').

There is a recent preprint, currently under review, entitled Extensive misidentification of European ivy species (Hedera L.): How taxonomically reliable are online biodiversity databases?, which can be read here: https://doi.org/10.21203/rs.3.rs-3693710/v1. It has a lot of useful information for us here. Here are the most important elements from that manuscript (with numbered references removed for readability):

"This is the case for ivies (Hedera L.), a small genus of recent diversification affected by extensive hybridization, where species delimitation has long been controversial. Besides, ivy species identification relies on inconspicuous microscopic characters (trichomes) of difficult interpretation while macroscopic characters (leaves, flowers or fruits) tend to have low taxonomic importance. In fact, the first feature in an ivy species diagnosis is the identification of the type of trichomes, while leaves are regarded as secondary diagnostic characters or even useless (Fig. 2). Even in cases where leaves are used for species diagnosis, it is the combination of leaf and trichome characteristics that allows species identification."

"The most important diagnostic character in Hedera is the type of trichomes (scale-like, stellate-rotate and stellate-multiangulate; Fig. 2A), whose assessment constitutes the first step on the species identification for the European ivy species...Indeed, the single area in the world where the three types of trichomes converge is the Iberian Peninsula in Europe, with H. helix as the representative of the multiangulate type, H. hibernica as the representative of the rotate type and H. iberica representing the scale-like type. Although the trichomes are quite distinct among the three species, towards the regions where they contact across their ranges it is frequent to observe individuals with intermediate features. Indeed, across H. hibernica distribution (from UK to Spain) whenever it contacts with H. helix, there are populations whose individuals display intermediate forms of trichomes between the typical multiangulate of H. helix and the typical rotate of H. hibernica."

"In the case of the intermediate trichomes between H. helix and H. hibernica, leaves (or any other macromorphological trait so far analyzed) do not help species identification, as the two species show high variation and overlap (Fig. 2B)."

This third point is the crucial one here: you need to inspect the trichomes to reliably differentiate H. helix and H. hibernica. Why is this important? Because H. hibernica is also present in Australia. And indeed, it seems that a very large proportion of records, both photographic and vouchered, identified as H. helix are actually H. hibernica.

Let's skip across the Pacific to the US for a second, with a 2006 paper entitled Prevalence of different horticultural taxa of ivy (Hedera spp., Araliaceae) in invading populations (https://link.springer.com/article/10.1007/s10530-004-2424-6). They note that:

" Several similar forms are sold under the general common name of English ivy. They are groundcovers that have been used extensively in urban landscapes in the Pacific Northwest because of many desirable characteristics, including appearance, shade-tolerance, and easy propagation and growth. They have also been used for erosion control and slope stabilization, although their effectiveness in that role appears overrated (Parker 1996). As a result of their extensive use and biological characteristics, ‘English’ ivy has become one of the most ubiquitous invaders of urban and suburban forests in the Pacific Northwest (Reichard 2000)"

"As a result of both genetic analysis and morphological identification, 85% of the 119 samples were attributed to Hedera hibernica and 15% to H. helix. This indicates that H. hibernica is the taxon most responsible for the invasion by English ivy in the Pacific Northwest. Only 15% of the samples from the invading population were found to be Hedera helix"

Jumping back to Australia, it seems like we have a similar situation. VicFlora currently lists both H. helix and H. hibernica as naturalised in Victoria, and provides this explanation:

"Hedera hibernica is the more prevalent of the two species of Hedera naturalised in Victoria, but the name H. helix has been widely misapplied to this species."

"Hedera helix is the less common of the two species of Hedera naturalised in Victoria, known only from Ballarat, and Hallston Forest near Leongatha, but possibly more common that collections suggest. In South Australia H. helix is also rare, with H. hibernica being the prevalent species (Chris Brodie pers.comm. July 2020)."

Here is the VicFlora couplet for the two species:

Hairs on young leaves and stems in vegetative shoots stellate, usually stalked, hairs white or off-white, rays 4–8(–10), of different lengths and projecting at a range of angles from the leaf surface, giving an irregular appearance; leaves in vegetative shoots very rarely more than 8 cm wide, often lobed > 1/2 way to base with lobes usually longer than wide = H. helix

Hairs on young leaves and young stems in vegetative shoots stellate, generally sessile, the central part of the stellate hair occupying 1/6–1/3 of the diameter, hairs often pale yellowish brown, sometimes white or off-white, or orange or tan in the centre with rays mostly white or off-white, or indumentum a mixture of these colours, rays 4–12(–15) radiating in a plane parallel to the leaf blade, rays often of similar length; leaves in vegetative shoots occasionally > 8 cm wide, usually lobed < 1/2 way to base with lobes often as wide as long = H. hibernica

[reiterating per my comments above that the leaf characters noted at that couplet shouldn't be relied on, it's the hair characters that are diagnostic]

In addition to Victoria, the South Australia, Queensland and Western Australia herbaria also recognise that H. hibernica is naturalised in Australia in addition to H. helix. Indeed, NSW is the only one that still only recognises H. helix as being present.

What do the specimens in AVH say? As an aside, it's a bit annoying at the moment. Despite H. hibernica being unambiguously present in Australia, with four state herbaria recognising its presence as noted above, this species currently does not have the APC 'red tick' (I don't know if this is an oversight or a deliberate decision), so the AVH/ALA do not have a profile for the species. Because of this, all specimens IDed as H. hibernica in the AVH are being dumped into the genus. Anyway, there are currently 59 specimens with an ID of H. hibernica in the AVH. These are from:

Victoria (30 specimens)
South Australia (17)
NSW (3)
Queensland (4, including one that is not naturalised)
ACT (2)
WA (2)
Tasmania (1)

We have confirmed specimens from NSW, so why does NSW not recognise the species? Because the specimens are lodged at the Brisbane Herbarium, and a profile is only created for a species in PlantNET if the NSW Herbarium holds a specimen (I personally disagree with this policy, but that's an argument for another day).

So overall, it is clear that H. helix is not the only species present in Australia, with H. hibernica also present (there is actually a third species too, H. algeriensis. This one is a little bit confusing: APC lists it as naturalised in South Australia, and the South Australia eFlora agrees, yet the AVH has zero specimens from South Australia. Instead, it has 7 specimens from Victoria, yet VicFlora doesn't list it... So let's put this one aside for now). And at the very least in Victoria and South Australia, H. hibernica is the more commonly naturalised species. Yet on iNat, there are almost 900 Australian observations IDed as H. helix, and just 26 as H. hibernica!

This seems to me a very clear-cut case of a pervasive set of misidentifications driven by a combination of factors, including CV suggestions and the fact that a lot of people probably don't know H. hibernica even exists as a species, let alone knowledge of how similar it is to H. helix and the characters that are required to differentiate the two. From even just a brief skim of Australian observations, it is immediately clear that the vast majority, probably close to 100%, do not contain images that allow a legitimate ID to be made of either H. helix or H. hibernica, and that almost all of these records should be kept at genus. My plan is to therefore go through all Australian Hedera observations and, unless images of trichomes are present (eg this observation by @nomennudum: https://www.inaturalist.org/observations/141113195), I will be pushing them all back to genus, as I don't think it is useful to have so many observations IDed as H. helix when a) you cannot actually make that ID from the characters present, and b) a large proportion of these records are almost certainly actually H. hibernica. If noone has any objections to this, I'll do this in the next day or two.

Posted on December 14, 2023 10:04 AM by thebeachcomber thebeachcomber


Thanks Thomas. Your proposed action makes sense.

Posted by bushbandit 7 months ago

By the way, the reference images for these two species may need some attention too.

Posted by bushbandit 7 months ago

I haven't been photographing them much in the last couple of years because it seemed a bit difficult.

I think this one might be my only verifiable one

I did try to get some details in a few others.

Reverting all 800 manually seems like a big job!

Posted by reiner 7 months ago

I have no objection, Thomas.

Posted by margl 7 months ago

Good work Thomas. Busting them all back to genus for now does seem like the best way to go. Looks like I better get out and photograph hairs, and see if we can't get some verifiable images up there.

Posted by jackiemiles 7 months ago

Sure. Sounds reasonable.

Posted by pennywort_man 7 months ago

I'll also begin working on it.

Posted by cs16-levi 7 months ago

@cs16-levi if you look at observations outside Australia, the situation is probably different to what I've described above, so make sure to check what the local state of the playing field is

Posted by thebeachcomber 7 months ago

Of course! I've read the journal post - i'll only be doing the ones in Australia.

Posted by cs16-levi 7 months ago

thanks for that
(also if you can, it's helpful to link this post when you're adding IDs so the observer + other IDers know why you're adding a disagreeing ID)

Posted by thebeachcomber 7 months ago

I'll be sure to do that every time I am asked why I am disagreeing.

Posted by cs16-levi 7 months ago

thanks all, I've now finished reviewing all Australia obs
sorry for spamming some of you with many notifications :)

@bushbandit good point re the reference images, none of them are from Australian obs so I'll leave for now as I suspect they're probably ok
@reiner wasn't too bad, just a few hours with the cricket on in the background...

aside from the legitimate ones that are still sitting at species, there are a fair few records still at species due to 3 or more species IDs previously being added, so if you see any of those would be appreciated pushing them back

Posted by thebeachcomber 7 months ago

I think I may have an explanation for the widespread use of H. helix. I checked back on my reference books from before the everything on the internet days. I used "Weeds. An illustrated botanical guide to the weeds of Australia" by Auld and Medd, which is a 1987 book and more recently I bought "Weeds of the South-East. An Identification Guide for Australia. Third edition" by Richardson, Richardson and Shepherd. This is a 2016 edition of a book originally published in 2006. Both of them have only H. helix under ivy as a weed. A Google search for ivy as a weed in Australia likewise gives predominantly H helix information, even on government agricultural sites. So in some ways it's a tradition and most people looking to identify the stuff will just find the H. helix name.

Posted by elizabethhatfield 7 months ago

@elizabethhatfield Elizabeth yes this will be the case for a lot of naturalised species and inaturalist seems a good tool to discovering these. Unfortunately it tends to work the other way for some native taxa with people identifying things to species that cannot be separated from 1 or more other species using the photo.

Posted by margl 7 months ago

Add a Comment

Sign In or Sign Up to add comments