A plant that seems to defy biogeographical and evolutionary rules: Dodonaea viscosa (Sapindaceae), part 1

THE PUZZLE

Dodonaea viscosa (https://www.inaturalist.org/taxa/122711-Dodonaea-viscosa and https://prota.prota4u.org/protav8.asp?g=pe&p=Dodonaea+viscosa and https://www.anbg.gov.au/gnp/interns-2007/dodonaea-viscosa.html and https://anpsa.org.au/APOL27/sep02-1.html) is biologically puzzling in at least six ways.

Firstly, it is uniquely cosmopolitan for a woody plant (https://www.mq.edu.au/__data/assets/pdf_file/0005/1275341/Plant-of-the-week-Dodonaea-viscosa-Hop-Bush-ed-2.pdf). Long before human influence, it spread naturally across several oceans and then far inland on several continents.

Secondly, this wide natural distribution was achieved despite the fact that 60 other members of the same genus remain restricted to the same continent - and in many cases the same landscapes - from which D. viscosa originated.

Thirdly, its original occurrence was on a landmass, namely the 'island continent' of Australia/Sahul (https://en.wikipedia.org/wiki/Sahul), that has never been connected to any continent by a land-bridge.

Fourthly, its diaspores seem not to be particularly adapted for long-distance dispersal.

It is true that the seed-capsules have papery wings, suggesting anemochory (https://www.dictionary.com/browse/anemochory), and the seeds survive immersion in sea water (West, 1984, https://www.publish.csiro.au/sb/BRU9840001).

However,

the capsules tend to remain attached to the plant for months (https://plantpono.org/pono-plants/dodonaea-viscosa-%CA%BBa%CA%BBali%CA%BBi/), as opposed to falling off and blowing away,
the seeds possess a small elaiosome (clearly illustrated in https://spapps.environment.sa.gov.au/SeedsOfSA/speciesinformation.html?rid=1513#:~:text=Seed%20collecting%3A,for%201%20to%202%20weeks. and https://spapps.environment.sa.gov.au/SeedsOfSA/speciesinformation.html?rid=51), indicating local burial by ants, and
most populations occur far from rivers or the coast, making it unlikely that the diaspore is particularly adapted for dispersal and sowing by means of flotation.

Fifthly, D. viscosa is 'polymorphic', in a way incongruent with subspeciation/raciality/ecotypy (https://en.wikipedia.org/wiki/Ecotype and https://digital.library.adelaide.edu.au/dspace/handle/2440/119801 and https://onlinelibrary.wiley.com/doi/10.1111/j.1365-2699.2009.02176.x).

Please see https://www.inaturalist.org/posts/99976-a-plant-that-seems-to-defy-biogeographical-and-evolutionary-rules-dodonaea-viscosa-sapindaceae#activity_comment_81def12b-fd99-44ea-9b4a-a264e1b52b8c.

In a given population, the leaves may take a distinctive form (e.g. https://wtlandcare.org/details/dodonaea-viscosa-subsp-cuneata/) without any geographical isolation from other forms (see 'key to the subspecies' in https://plantnet.rbgsyd.nsw.gov.au/cgi-bin/NSWfl.pl?page=nswfl&lvl=sp&name=Dodonaea~viscosa).

By the same token, the leaves may remain true to form despite having been subjected to environmental pressures and 'genetic drift' on a completely different continent for hundreds of thousands of years.

Sixthly, D. viscosa has remarkably dense wood (https://www.inaturalist.org/journal/milewski/100396-a-plant-that-seems-to-defy-biogeographical-and-evolutionary-rules-dodonaea-viscosa-sapindaceae-part-2-wood-density#).

DISCUSSION

The literature lacks any satisfactory hypothesis on how D. viscosa achieved its unique spread, either ecologically or mechanistically.

The following review of fruit-form and seed-dispersal in the family Sapindaceae, more generally, highlights the puzzle.

No sapindaceous plant seems to be adapted for long-distance dispersal.

This is because

the typical fruit-form in this family is arillate (https://en.wikipedia.org/wiki/Aril), involving seed-dispersing animals ranging from ants through birds to mammals,
those fruits adapted - by virtue of papery wings on the mature, dry capsule - to dispersal by wind tend not to move farther than a few hundred metres from the parent individual, and
the seeds are not particularly small in any member of the Sapindaceae.

In Dodonaea, the aril is an outgrowth of the funicle (West, page 33). It is

always so small that it is better-described as an elaiosome (https://en.wikipedia.org/wiki/Elaiosome) than as an aril, and
so small in D. viscosa that some authors have claimed that it is absent.

Two relevant mysteries are

the adaptive value of the bright hues of the mature capsules of various spp. of Dodonaea, including D. viscosa (https://www.friendsgbg.org.au/uploads/images/Plants-in-Focus/Dodonaea%20viscosa.Lucy%20Pope.PiF.2018-11.1.150.PS.pdf and https://www.aboutgardendesign.com/plant-guide/item/dodonaea-viscosa-purpurea and https://resources.austplants.com.au/plant/dodonaea-boroniifolia/), and
the natural lifespan of D. viscosa, for which no data seem available.

Please also see https://www.inaturalist.org/journal/milewski/100103-a-comparison-of-sapindaceae-in-the-mediterranean-and-adjacent-arid-climates-of-australia-and-southern-africa#.

to be continued in https://www.inaturalist.org/journal/milewski/100396-a-plant-that-seems-to-defy-biogeographical-and-evolutionary-rules-dodonaea-viscosa-sapindaceae-part-2-wood-density#...

Posted on October 7, 2024 10:19 AM by

milewski

Comments

In North America, there are two 'subspecies' of Dodonaea viscosa that occur nowhere else, viz. elaeagnoides (https://repository.naturalis.nl/pub/524969/) and arizonica.

The following illustrates the narrow leaves of arizonica:
https://www.inaturalist.org/observations/210564468

For comparison, the following illustrates angustifolia, in the same species, in South Africa:
https://www.inaturalist.org/observations/240383919.

What is the difference?

In southern Africa, there are two 'subspecies', viz. viscosa and angustifolia (https://treesa.org/dodonaea-viscosa/).

Posted by milewski 10 days ago

@iancastle @lloyd_esler @alan_dandie @tonyrebelo @arthur_chapman

The interpretation of variation within D. viscosa in the literature seems garbled and nonsensical. I suspect that this is mainly because the whole concept of subspeciation (https://en.wikipedia.org/wiki/Subspecies) does not apply to this species in the first place.

Various 'subspecies' of Dodonaea viscosa are described in https://anpsa.org.au/plant_profiles/dodonaea-viscosa/, where the spelling 'spathulata' is used.

Further information is presented in https://www.anbg.gov.au/gnp/interns-2007/dodonaea-viscosa.html and https://www.florabank.org.au/static/34de8e9cf148e1ad3883d6a8c378a317/f9412fca697f7dd0b9c87c6def4556ef.pdf.

As an example of the incoherence of the literature, w.r.t. the so-called 'subspecies' of Dodonaea viscosa, please consider 'spatulata'.

This 'subspecies' a) is reported right across Australia, from Perth to Sydney, yet b) remains sympatric with several other 'subspecies'.

In Perth, Powell (1990, https://catalogue.nla.gov.au/catalog/362451), on pages 90-91, describes 'spatulata' as the only form present. This classification is inconsistent with the accompanying illustrations, which do not show the leaf-shape claimed.

In and near Sydney, 'subsp. spatulata' is again reported, with the claim of synonymy with 'var. arborescens', and the caveat that it 'shows a great amount of leaf morphological variation, with leaf characters intergrading with those of subsp. angustissima, subs. cuneata and subs. mucronata' (https://plantnet.rbgsyd.nsw.gov.au/cgi-bin/NSWfl.pl?lvl=in&name=Dodonaea~viscosa+subsp.~spatulata&page=nswfl).

In https://profiles.ala.org.au/opus/foa/profile/Dodonaea%20viscosa%20subsp.%20spatulata, 'subsp. spatulata' is described as occurring over most of Australia, including Tasmania.

In Victoria (https://vicflora.rbg.vic.gov.au/flora/taxon/9fe71aed-2b5c-46df-b229-a52c25b23518), 'subsp. spatulata' is confusingly described as showing certain variations within the subspecies (e.g. between the Grampians and Warrandyte), yet somehow elsewhere also intergrading with subspp. cuneata and angustissima.

Overall, the current treatment seems to confuse adjective with taxonomy, not so?

Posted by milewski 6 days ago

A POSSIBLY NEW CONCEPTUAL FRAMEWORK FOR SO-CALLED SUBSPECIFIC VARIATION IN DODONAEA

In Dodonaea, there is a pattern of remarkable inconsistency in the form of the leaves. This applies both within Dodonaea viscosa, and interspecifically.

Interspecifically, I refer mainly to the distinction between simple and complex (pinnate) leaves.

Among the nearly 70 spp. of Dodonaea, many have complex leaves (https://bie.ala.org.au/species/https://id.biodiversity.org.au/node/apni/2905673), categorically different in form from simple leaves. Certain spp. have 'lobed' leaves of a shape intermediate between simple and complex (https://www.inaturalist.org/taxa/353810-Dodonaea-lobulata).

Is it possible that the variation is largely phenotypic, not genotypic? In other words, is it possible that Dodonaea is heterophyllic/heteroblastic at a generic level, as opposed to merely intraspecifically or within a given individual?

Heterophylly/heteroblasty are, by definition, phenotypic, not genotypic, phenomena, in which the variation in the form of the leaves reflects the switching on/off of certain genes.

As far as I know, no species of Dodonaea shows heteroblasty in the sense of 'juvenile' foliage.

However, even in D. viscosa, there are traces of complexity in the leaves, which are occasionally 3-toothed or minutely dentate (https://vicflora.rbg.vic.gov.au/flora/taxon/26a2dd86-3afa-43cf-8f95-83d8cf125ec2).

Posted by milewski 5 days ago

Reference: Harrington and Driver (1995, https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1442-9993.1995.tb00573.x).

Location: http://bonzle.com/c/a?a=p&p=240847&cmd=sp and https://en.wikipedia.org/wiki/Bourke,_New_South_Wales

Seed-harvesting ants (Pheidole spp., https://www.inaturalist.org/observations?place_id=6825&subview=map&taxon_id=47343) are important in regulating the abundance of seeds of Dodonaea viscosa.

The ants are initially attracted to fallen seeds by virtue of the elaiosomes. They remove the small food-body, and then discard the seeds in middens outside the entrance of the nest. The ants later take the seeds back into the nest during the cool season, storing them in chambers 2-30 cm deep.

Germination of D. viscosa is plentiful at the study site, both after fire (which breaks the dormancy of those seeds that remain on the surface, having escaped initial collection by the ants), and after unusual rains (which percolate deep enough to reach the seeds stored deep in the nests of Pheidole).

Posted by milewski about 6 hours ago