Showing posts with label taxonomy. Show all posts
Showing posts with label taxonomy. Show all posts

Tuesday, 28 January 2014

Wednesday Wildflower: tarweed

Tarweed is flowering at the moment.  Parentucellia viscosa is an erect herb with small yellow flowers and its leaves are covered with glandular hairs so dense they feel sticky to the touch.  You'll see it in damp patches beside roads and tracks, along the edges of ditches, and wet hollows in grassland.
Tarweed, Parentucellia viscosa, Karori, Wellington
It's a hemiparasite, which means it derives some of its nutrients parasitically from other plants, but it's also green and able to generate its own energy through photosynthesis.  Full parasites (holoparasites) usually lose the ability to photosynthesise and to make green pigments, so they are often brown or pale.

Tarweed flower.
It's related to Euphrasia, another genus of hemiparasites, of which we have a large number of native species in New Zealand, and to the introduced broomrape, Orobanche, which are holoparasites.

Orobanche minor, broomrape, near Nelson.
The hemiparasites Parentucellia and Euphrasia used to be classified along with Veronica in the family Scrophulariaceae, but it was discovered a decade or so ago ago that Scrophulariaceae as it was then drawn up wasn't a natural group (of related plants).  So that previously large family has been split up. Veronica was transferred to be classified with its relative Plantago (Plantaginaceae), and Parentucellia and Euphrasia joined their relatives the broomrapes in Orobanchaceae.

Ngaio, Myoporum laetum, Wellington
Although Scrophulariaceae has been dismembered into 7–10 different families to make a more natural classification, it still exists as a much smaller family, many of them African.  Our only native member is Myoporum, ngaio.

Wednesday, 27 November 2013

Wednesday wildflower: Diddillibah wildflowers.

This week I’m on the Sunshine Coast, Queensland, visiting family before the Australasian Systematic Botany Society’s conference in Sydney next week.  We arrived last night and this morning took a short walk to get a feel for our surroundings, from Diddillibah to the Maroochy River and back.
Maroochy River
We haven’t seen much natural vegetation yet, but plenty of wildflowers and a few native Eucalypts and she-oaks.
Mangroves, Maroochy River.
Along the river are mangroves, which I assume are the same as we have in New Zealand, Avicennia marina.  It reaches its southern limit—and the southern limit of mangroves generally—at Corner Inlet, Victoria.  Here, in the warmer climate, they grow taller.
Mistletoe in a Casuarina tree.
There were mistletoes in the she-oak (Casuarina) trees near the river.
Mistletoe flower buds.
I think this little weed is Emilia sonchifolia, something I’ve collected before, in Singapore; at least I think it’s the same.  
Emilia sonchifolia

Its resemblance to sow-thistle (Sonchus) is remarkable, but it’s convergence, because this isn’t in the same tribe.
Emilia sonchifolia flower head.
The single row of involucral bracts is characteristic of tribe Senecioneae, whereas Sonchus is in the Lactuceae.

And there was a pelican on the river. Nice.

Tuesday, 12 November 2013

A new classification for the southern beeches.

ResearchBlogging.org
In New Zealand, the forest we typically identify with—the “bush”— is the lowland mixed conifer-angiosperm forest, with a canopy usually of angiosperm trees like tawa (Beilschmiedia tawa) or kamahi (Weinmannia racemosa) and large emergent conifers of the southern families Podocarpaceae (e.g., rimu, Dacrydium cupressinum) and Araucariaceae (i.e., kauri, Agathis australis).  But in most montane parts of the South Island (Te Wai Pounamu) and on often drier ridges and hill country of the North Island (Te Ika A Māui), a very different type of forest is dominant.  This forest comprises a uniform canopy of often a single species of southern beech.  Usually the forest doesn’t have a dense understory, giving an open and well-lit appearance to the interior.  This is just as much an iconic New Zealand forest as the “bush”, and one that’s familiar to many trampers.
Southern beech forest, Orongorongo Valley, near Wellington
Similar forests are found in the southern part of South America, so that travelers there from New Zealand often feel it’s just like home.  Southern beech forest also occurs in Australia, New Guinea and New Caledonia.  It’s known from fossils in Antarctica too, going back to the Cretaceous, as well as in many of the places where it still occurs today.

The southern beeches were originally classified in the genus Fagus, along with their northern namesakes, but by 1850 their differences had been recognised and they were transferred to the genus Nothofagus (the name means southern beech [correction, 21 Nov 2013: it means "false beech"; southern beech would be Notofagus.  H/T Rosi]).  As Nothofagus, the southern beeches have been important trees in New Zealand ecology, conservation, forestry, and biogeography.  Whole books have been written about them.  Nothofagus is currently reckoned to have about 40 species.  In 1962, a Russian botanist, Lyudmila Kuprianova, went a step further and proposed a new family, Nothofagaceae, for the southern beeches.  This took rather a while to be accepted.

Many botanists have wrestled with the relationships of the species within Nothofagus, using sometimes single or few characteristics, other times multiple ones.  The advents of (1) cladistic thinking (using explicit evolutionary trees) and (2) molecular characteristics from DNA sequencing have been of major help to this enterprise, because DNA has provided a wealth of new characters that are independent of the morphological ones and because the analysis and interpretation are out in the open for everyone to evaluate.  Pretty quickly, the understanding of relationships in the southern beeches has converged on a single well-supported arrangement, which was arranged into a classification by Australian botanists Bob Hill and Jenny Read, who recognised one genus (Nothofagus) with four subgenera.
Southern beech forest near Eastbourne, Wellington.
Beyond the southern beeches, DNA sequence data were also telling us a lot about the relationships of southern beeches to the oaks, beeches, chestnuts and she-oaks and it became pretty clear that Kuprianova was correct in isolating them in their own family.  It turns out that the ancestor of the beech order (Order Fagales) first divided into two species: one that was the common ancestor of the northern sweet chestnuts, beeches & oaks, she-oaks, myrtles, and more—seven families in all—and the other that was the common ancestor of just Nothofagus.  If you were to divide Fagales into two suborders, one would have seven families and many genera, the other would have just one family, and that family would have just one genus.  Thus Nothofagus and Nothofagaceae have different ranks (their place in the hierarchical classification) but identical circumscriptions (the species they contain); that redundancy means we're not using the available hierarchy of ranks to full advantage.

What’s more, the current classification of all the southern beeches in one genus Nothofagus can be a bit misleading.  Most biologists agree that it’s absolutely essential that every genus or family should contain closest relatives.  In other words, a species shouldn’t be more closely related to a member of another genus than it is to a species that’s classified in its own genus.  Nothofagus doesn’t break that rule: every species of Nothofagus is more closely related to every other species than it is to any species that’s not placed in Nothofagus.  So far, so good.

But it’s easy to assume that our New Zealand species—black, hard, red, mountain and silver beeches—might be each other’s nearest relatives, and often people are surprised to find that’s not the case.  In fact, hard, black, mountain, and red beeches are related, but silver beech’s nearest relative is in Australia.  Wouldn’t it be better if their classification and their scientific names could reflect that?

This week two New Zealand botanists, Peter Heenan and Rob Smissen from Landcare Research, have revisited the classification of the southern beeches (Heenan & Smissen 2013).  They brought together everything that’s been published so far, from both morphology and molecular systematics, and added some new data and analyses of their own.  Their findings are pretty much the same as several previous reports, but they can now place greater levels of confidence in the groups they recognise.  They comprehensively discuss alternative classifications and alternative criteria and come down with what I think is the most sensible classification. 

Nothfagaceae now contains four genera.
  • Nothofagus comprises just five species from temperate South America.  The rest of the family is no longer classified as Nothofagus.
  • Lophozonia is a reinstated genus, containing seven species from South America, New Zealand, and Australia.  
  • Fuscospora has six species and a very similar distribution; it’s a newly recognised genus, although like the others it has been treated as a subgenus in the past.  Additionally in Fuscospora, this paper promotes mountain beech to species rank as F. cliffortioides.  I look forward to reading the evidence for that change, because it was previously treated just as a variety of black beech.
  • Finally, Trisyngyne is the largest genus (25 species) and found today in the tropics: New Caledonia, Papua New Guinea and extending into Indonesia.  
These genera are strongly supported by both molecular, morphological, and chemical characteristics, and they have symbiotic fungi and parasitic fungi and insects that also seem to recognise their relationships.
Red beech, Fuscospora fusca.
Black beech, Fuscospora solandri.
Mountain beech, Fuscospora cliffortioides, near Cass, Canterbury.
Hard beech, Fuscospora truncata
Silver beech, Lophozonia menziesii.
When we use these new names for the New Zealand plants, we see immediately that we have two natural groups represented here: Fuscospora and Lophozonia.  New Zealand no longer has any species of Nothofagus.
Red beech, Fuscospora fusca, Tunnel Gully near Wellington.
A final word of a more general nature.  Some people will want to reject this change, perhaps because they feel nostalgic about the name Nothofagus, or perhaps because they feel name changes are disruptive.  But taxonomy is science and there are scientific criteria involved.  Like climate change, evolution, and vaccination, you can’t simply reject sound science because you don’t like it.  
Silver beech, Lophozonia menziesii, Haast Pass.
It’s very rare in science for there to be two equally well-supported positions such that users are free to choose whichever one they prefer.  Rather, scientists make decisions after critically considering the evidence.  It's true a classification is a human construct, but it's based on facts about evolutionary history.  Those facts are hard-won data from the field, herbarium, and genetics lab.  If your opinion contradicts those facts, then you're at risk of denying the science.

In this case however, both the old classification and the new one do pass the most important test, that of classifying related species together, so we can't rule out one or the other on that ground.  The question here is, "what's the appropriate rank for these four well-supported groups?"  But is one answer better than the other?  Heenan and Smissen argue strongly and in detail that there are good reasons to prefer their new scheme over the old one.  For instance, they show that the newly-recognised genera are at least as old, diverse, and distinct as established genera in the other families of the order, that the new names are more informative about relationships among the southern beeches, and that a redundant grouping has now been eliminated.  They conclude, and I agree, that these benefits far outweigh the temporary disruption of having new names to learn.
An Australian beech, Lophozonia moorei, growing at Eastwoodhill.

Heenan, P.B.; Smissen, R.D. (2013). Revised circumscription of Nothofagus and recognition of the segregate genera Fuscospora, Lophozonia, and Trisyngyne (Nothofagaceae) Phytotaxa, 146 DOI: 10.11646/phytotaxa.146.1.1

Thursday, 19 September 2013

The Great Veronica Hunt — part 5.

(Note: I've updated this post on 28 September, giving the name of the botanist whose advice led me to these two Veronicas and whose collections in New Zealand herbaria verify those discoveries.  The changes are underlined.)
 
If you've been paying attention, and I'm sure you have, you'll notice I haven't posted the Great Veronica Hunt part 4, but that's what I should have called this post a couple of weeks ago.  So, skipping part 4, here's part 5.

In part 1, I described trying to find Veronica peregrina last year.  That was frustrating, because although I had a very accurate description of the location and the habitat, I was there too late in the season. To make it worse, the original collector—Whanganui botanist Colin Ogle— hadn't seen it there for a few years and doubted it would still be present.  Still, Colin had told me last autumn of a site for another species I need to photograph, V. chamaedrys, so yesterday I went after them both.

Veronica peregrina plants, Kakariki.
It took a while to find V. peregrina, but it is still there.  It was growing in silty gravel at the edges of dried up puddles in a rough vehicle track.  The biggest plants were about 75 mm tall, and the small white flowers weren't fully open on a rather dull day.  I brought some plants back to photograph, some to grow, and some to make a couple of herbarium specimens.
Veronica peregrina
This is an American plant, and it seems to be often associated with railways in the States, so it's interesting that this site is right beside the main trunk railway, at Kakariki, near Marton.  I don't know whether the activities of railways spread seeds around or whether they create suitable habitats, or maybe it's just a coincidence.

V. peregrina plants are bright green and either have no hairs or very few long glandular ones.  Their flowers are pure white, an unusual colour for a northern hemisphere Veronica (most are blue), but a common colour among our native species (only a few of which are blue).

While at Kakariki, I'd promised a colleague I'd look for spore-bearing cones on Equisetum arvense, which is naturalised along the banks of the Rangitikei River.  I'd seen it there in abundance last trip, so I confidently went down to the river.  However the river banks have been extensively sprayed, and, while it hasn't completely cleared the infestation, it's knocked it back pretty severely.  Eventually I managed to find a single cone, and took photographs and a specimen.
Equisetum arvense

Equisetum (horsetail) is an odd plant, now known to belong among the ferns. The cones produce not seeds, but spores (pine cones produce spores too: male cones make male spores that develop into multicellular pollen grains before they're dispersed, and in the familiar female cones the spores are retained, develop there, and after fertilisation each develops into parts of a seed).  Horsetail spores are formed in cylindrical sporangia underneath the hexagonal umbrella-like scales on the cone, which spread apart to release them.
Equisetum arvense, spore-bearing cone.
Then it was on to Marton for lunch and through Whanganui to the hill country inland from Kaiiwi. Colin Ogle had told me of a locality for Veronica chamaedrys, a plant I'd seen and photographed in England and France, but one that's naturalised in a few scattered localities in New Zealand.

Veronica chamaedrys,  St. Léon sur Vézère, Dordogne, France.
Here in the bush it grows around the edges of a small clearing in an old waterworks reserve.  How it got here is anyone's guess, but it's well-established in a small area.  We were too early for flowers, but it's a vigorous plant and I'm sure we can grow it on at home in a semi-shaded spot.  If this works, I'll post photos later.
Veronica chamaedrys at the edge of the clearing
The roadside cliffs through the bush were covered in flowering plants of Ourisia macrophylla subsp. macrophylla, and some of them were pink-flowered, at least in the bud.  I'd never seen such colour in New Zealand Ourisia, but in South America there are both red- and pink-flowered members of this genus.
Pink Ourisia.
It's always odd going back to Whanganui.  That's where we first settled when we emigrated to New Zealand in 1955.  I started school there (this is me on the left end of the middle row), and we used to swim at Kaiiwi Beach.

Tuesday, 3 September 2013

Wednesday wildflower: Veronica hederifolia

Last week I was at Lincoln, near Christchurch, working in the herbarium at Landcare Research.  I was checking my descriptions and identifying specimens towards my Veronica treatment for the new on line Flora of New Zealand, the eFlora.
Veronica hederifolia growing at the foot of an oak tree in the Liffey Domain, Lincoln.
One of the introduced species (there are about 20 of them) that I hadn't yet seen grows right there in Lincoln, so it seemed a good opportunity for a field trip to collect and photograph it.  Veronica hederifolia plants are soft annual herbs that creep along the ground.  Their flowers appear to be solitary in the axils of the upper leaves, but that depends on an interpretation.  Leaves that don't produce flowers are opposite, but there's a shift to alternate leaves, each of which has a flower in its axil.  It's probably reasonable as an alternative interpretation to consider this to be the initiation of a terminal inflorescence. In any case the leaf form doesn't change, whereas in many Veronica the flowers are produced in the axils of much smaller and simpler leaves, which are designated as bracts.
Veronica hederifolia growth form.
V. hederifolia has been growing there in Lincoln along the banks of the L2 river for over 50 years.  The botanist who collected it last—in 1985—was able to tell me exactly where to look, and there it was. There's one other collection in Landcare's herbarium, from St Mary's College grounds in Christchurch, and the Flora refers to other verified locations in Hawke's Bay, Manawatu, and Southland.

Although V. hederifolia looks a bit like V. persica in the way it grows, there are a lot of clear differences.  The leaf shape for one, but also the flowers are smaller, and the anthers are held right against the stigma so it self-pollinates, in spite of producing lots of nectar.  The fruits of V. hederifolia are hairless, circular, and barely notched, whereas fruits of V. persica are hairy along the edges of two widely diverging lobes.
Veronica hederifolia flower.
The calyx lobes are folded length-wise and have long hairs along their edges.
Veronica hederifolia, calyx.
I've taken a few small plants to try to grow it on at home, but annuals can be hard to transplant, so I'm hoping fruits and seeds will be ready when I go back to Lincoln next month.  Then I'll be able to finish my description by describing fruits and seeds and bring home some seeds to grow in the garden.

Thursday, 20 December 2012

Strange violets.

Māhoe (Melicytus ramiflorus) is one of the plants I love to show to visiting botanists from temperate Europe or North America.  They invariably ask, "What family is it?" and the answer always surprises because they're used to Viola as the representative of Violaceae.  Here's a violet that's a tree, has radially symmetrical flowers, and fleshy fruit; the world truly is upside-down in New Zealand.

Māhoe flowers on twigs of a male tree, Karori, Wellington
Well, not so fast.  It's true Viola is the type genus of Violaceae, but who said types have to be typical?  The type simply determines the application of the name according to the rules of nomenclature.  It's the circumscription—the definition of the membership of the group—that outlines its overall collective characteristics.  Violaceae is a largely tropical family, and they're woody except for Viola.  Quite a few have radially symmetrical flowers and fleshy fruits (Mabberley 2008).

A "normal" violet (Viola sp.), France.
Māhoe might be New Zealand's commonest tree.  There's certainly lots of it around Wellington, where it's an important component of the still low second growth forests that, over the last 50 years, have replaced the gorse on hills around the city.

Māhoe flowers in pulses, all the trees in a location flowering in synchrony several times a summer (Powlesland et al. 1985).  Coffee does the same thing.  The flowers are small and borne on the twigs, a form of flower presentation called ramiflory.  They're scented, especially at night, but it's not a very pleasant smell, in my opinion.
Melicytus ramiflorus, male flower.  Note the vestigial ovary in the centre and the swollen connective at the back of each anther that functions as a nectary
Melicytus ramiflorus female flower.  Note the large stigma and the nectar produced from the connective at the back of each staminode.  Female flowers are about 2/3 the size of males.
The flowers are unisexual and produced on separate trees, so a whole tree is either male or female.  The connective of the stamens and (in female flowers) staminodes is also the nectary, and each produces a glistening drop as a reward for the pollinators.
Māhoe leaf skeletons
Their leaf veins are pretty tough, but the rest of the leaf decomposes freely, to leave exquisite lacy leaf skeletons on the forest floor.
Viola cunninghamii, Hooker Valley, Mt Cook National Park.
New Zealand has three native violets (like V. cunninghamii above), plus some introduced species.  We also have quite a few Melicytus; some are small trees and others are twiggy shrubs.  Some used to be classified in the genus Hymenanthera, but that was merged with Melicytus in the 1980s.  Recent research (Mitchell et al. 2009) shows there are two pretty clear-cut groups within the genus, although the authors didn't draw attention to it, and it might make sense to recognise Hymenanthera again.

References

Mabberley, D.J. 2008.  Mabberley's Plant-Book (3rd ed.).  Cambridge.

Mitchell, A.D.; Heenan, P.B.; Murray, B.G.; Molloy, B.P.J., de Lange, P.J. 2009.  Evolution of the south-western Pacific genus Melicytus (Violaceae): evidence from DNA sequence data, cytology and sex expression.  Australian Systematic Botany 22(3) 143–157.

Powlesland, M.H.; Philipp, M.; Lloyd, D.G. 1985.  Flowering and fruiting patterns of three species of Melicytus (Violaceae) in New Zealand. New Zealand Journal of Botany 23: 581–596

Tuesday, 16 October 2012

Wednesday wildflower: annual mouse-ear chickweed

There's mouse-ear (Myosotis) and there's mouse-ear chickweed (Cerastium).  True chickweeds are Stellaria, which isn't particularly hairy, but the mouse-ear chickweeds are furry like Myosotis, or like a mouse's ear.  They're not related to Myosotis however, in spite of their superficially similar leaves.
Cerastium glomeratum, Karori, Wellington.
This one is very common.  It's the annual mouse-ear chickweed, C. glomeratum.  Its leaves have green edges (some of the other species have scarious—thin, dry, almost translucent—edges to at least the uppermost leaves) and hairs overtop the apex of the sepals.  The flower stalks are shorter than the flowers and fruits, hence the name glomeratum, which refers to the rounded cluster of flowers.  

We have five species naturalised in New Zealand.  C. arvense (below) has larger flowers, and is often established on limestone.
Cerastium arvense, Cass, Canterbury.

Friday, 28 September 2012

Experimental taxonomy at home.

Ernst Mayr pioneered the biological species concept, an idea that brought taxonomy of species into line with population genetics and evolution.  The idea is that a species is defined by the genetic relationships among its members; they’re all part of one big potentially-interbreeding population.  In Linnaeus’s day people sought to classify species based on what they looked like, rather than who they could breed with.
Using appearance is a pretty good proxy for the ability to interbreed, and much of the time it’s what taxonomists still do, simply because doing the breeding experiments or measuring genetic relationships among individuals is just too time-consuming.
But there are two classes of concerns that arise.
On one hand, individuals belonging to the same species can look very different.

Sometimes juveniles are hugely different from adults, like caterpillar and butterfly, elva and eel, or juvenile vs adult lancewood.  In a New Zealand plant example, Jim Le Comte and Colin Webb (Le Comte & Webb 1981) showed experimentally that the speargrass Aciphylla townsonii is actually the juvenile form of A. hookeri.  Different juveniles seem to be a feature of New Zealand plants, but they're common elsewhere too.
Juvenile (left) and adult foliage of mataī (Prumnopitys taxifolia)
 Secondly, small genetic differences can lead to quite big visible differences in plants or animals that belong to the same species.  Some of these are simple polymorphisms, like eye colour in humans.  In Veronica amplexicaulis, a hairy form used to be distinguished as a separate species (Garnock-Jones & Molloy 1983, under the old name Hebe amplexicaulis).  But it turns out this difference is the product of two alleles of a single gene, as are occasional flower colour variants in many plants.
White and blue viper's bugloss, Echium vulgare, growing side by side (Black Birch Range, Marlborough).
Thirdly, local populations might adapt to special conditions.  On mine tailings, where toxic heavy metals pollute the soil, plants may acquire tolerance, and this could involve some differences in form or in underlying physiology, yet they still freely mate with the non-tolerant individuals nearby.  These are classified as ecotypes, but not as separate species.  The differences are maintained by strong selection, even in spite of free gene flow between the tolerant and intolerant plants.
The form of Veronica albicans that grows on the dolomite outcrop at Mt Burnett looks a little different from other populations of this variable species; it might have adapted to its substrate, yet there's no evidence that it can't exchange genes with the rest of its species.
 Fourthly, some plants and animals are able to alter their form to cope with different environments they find themselves in or to escape predators—phenotypic plasticity or polyphenism.  Some inchworm caterpillars develop different appearances to blend in with the foliage of whatever host plants they’re living on (Greene 1989).  The underwater and aerial leaves of aquatic plants can be hugely different.  Plants can have very different leaf shapes, or even leaf anatomy, depending on the amount of sunlight they’re receiving or even what season they’re in.
Eryngium vesiculosum has very different leaves in summer (above) and winter (Webb 1984).
In all four of these situations, the result is two different looking plants growing together side by side, giving the appearance of two distinct species that aren’t interbreeding.  That’s just the sort of thing that gets taxonomists and field botanists excited, because we always like to discover a new species.
On the other hand, the reverse situation can arise.  Two species can look so similar that their existence isn’t even suspected until genetic tests are done.  These are called cryptic species.
It’s important to be aware of these possibilities, and in fact to rule them out as explanations before jumping to the conclusion that the variation we’re observing is due to the existence of more than one species.  The idea that two different-looking plants growing side by side must be different species is simplistic, yet "side by side" has become a bit of a mantra in some circles.
One way to test these potential new species is by growing different-looking plants together in uniform environments—common garden experiments—and also growing genetically identical plants in different environments—reciprocal clone transplants.  These approaches were pioneered in the first half of last century by Swedish botanist Turesson and by American botanists Clausen, Keck, & Hiesey.
Veronica lanceolata in flower, Rimutaka Range.
The speedwell hebe Veronica lanceolata is widespread in the North Island of New Zealand and a few parts of the South Island.  Each region has its own form of the species and, in general, adjacent populations are quite similar.  With some familiarity, it’s possible to tell from its appearance where a plant has come from.  These differences are maintained in common garden experiments, but I don’t regard these forms as different species because they can cross freely, their flowers and fruits are very similar, the differences are quantitative rather than qualitative, they have the same chromosome number, and the changes are mostly gradual and continuous from place to place.
Each leaf is from a different population of Veronica lanceolata.
However, there are places where two very different-looking forms grow together side by side, and this is exactly the sort of situation where a simplistic "side by side" approach might lead a botanist to the view that two species are involved.  In the Ruahine and Kaimanawa Ranges, especially on limestone cliffs, there are low-growing small leaved plants growing together with bushier large-leaved plants.  Although their leaves and stems are different in size and stature, their flowers and fruits are the same, which is a bit of a clue that these plants are responding in a plastic way to their environments, that there are no underlying genetic differences, and no breeding barriers between them.  I’d always assumed so at any rate, even though some field botanists made numerous collections of both forms, mounted them as separate accessions, and labelled them to highlight the differences and the fact they grew together side by side.  The hint was implicit: these might be different species.  Perhaps fortunately, nobody had the confidence in their hunch to give them different names.
South end of the Maungaharuru Range.
A couple of summers ago I was in the Maungaharuru Range in central Hawkes Bay with my colleague Heidi Meudt from Te Papa.  We were looking for forget-me-nots along the tops for a detailed genetic study she's conducting into their taxonomy and evolution.  Along the cliffs at the south end of the range we found Veronica lanceolata growing in mostly shady sites among mosses and algae.  But when we stumbled into some sink-holes things got interesting.  Here were moist shady sites with quite large-leaved plants, very close to sunny outcrops with tiny creeping plants.  It was an ideal opportunity to test my hunch that these were just plastic responses to moisture and shade.


There wasn’t room in the garden at home for a large randomised trial, so I sampled just a couple of plants of each type from sites only a few metres apart, and brought them home to grow in pots.  I also pressed branches of each, to record and preserve how they had looked in the wild.
Collection 2834, small and large leaved plants just after potting, Feb 2011.

Collection 2836, a small leaved plant just after potting, Feb 2011. The white plastic labels are 13 mm wide.
They’ve been growing now for about 18 months, and some, but not all, of the changes are quite dramatic.  For 2384, the small-leaved plant now has somewhat bigger leaves, but it's still distinctly smaller than the large-leaved plants.  For 2836, leaves are now up to 20 mm long, whereas they were about 5 mm before.
Both surfaces of the largest leaves from each of the three plants (two from 2836 small), Sep 2012.
So the results are a bit mixed, and this shows how important it is to use large samples, not just a couple of plants, and to randomise the trial properly.  The plants have exhibited some phenotypic plasticity but that doesn't account for all the differences.  Note also the two very different leaf shapes from the same plant of 2836: more evidence of plasticity.  Maybe both phenotypic plasticity and ecotypic differentiation are happening in this population.  Next time I'm in a position to collect a bigger sample and repeat this experiment I will do so.  In the meantime, I can try crossing the small- and big-leaved plants this summer.  My expectation is the offspring will be fully fertile.
With these speedwell hebes, the differences in growth form and leaf shape are striking, but they aren’t sufficient to compel rejection of the hypothesis that they’re the same species, because there are two different and simpler explanations—phenotypic plasticity and ecotypic differentiation—for that pattern.  My simple experiment hasn't clearly demonstrated which is happening, because the experimental design and sampling are insufficient.  But it's important to note also that these are quantitative differences—leaf shape and size—just the sorts of things that often vary in natural populations.
I'm sure it’s possible to test species status scientifically and explicitly.  That means starting with a testable hypothesis.  It’s better to start with the hypothesis that the two entities are conspecific, because any differences are evidence to the contrary that would compel us to reject the hypothesis.  If instead we start with the hypothesis that they’re different species, it’s hard to imagine how many similarities between them would compel us to reject that idea.  And if we start with the hypothesis that there are two species, and then seek evidence to support the hypothesis, then we're not doing science, at least not as it was formulated by Karl Popper.
References

Garnock-Jones, P.J.; Molloy, B.P.J. 1983.  Polymorphism and the taxonomic status of the
Hebe amplexicaulis complex (Scrophulariaceae).  New Zealand Journal of Botany 20: 391–399.

Greene, E. 1989. A diet-induced developmental polymorphism in a caterpillar. Science 243: 643-646.

Le Comte, J.R.; Webb, C.J. 1981.   Aciphylla townsonii — a juvenile form of A. hookeri (Umbelliferae).  New Zealand Journal of Botany 19: 187–191.

Webb, C.J. 1984.  Heterophylly in Eryngium vesiculosum (Umbelliferae). New Zealand Journal of Botany 22: 29–33.