Thursday, 31 May 2012

Veronica colostylis

Veronica colostylis from near Arrowtown, New Zealand.
In autumn 2011 I wrote about Lammas flowering, when spring- or summer-flowering plants get fooled by a mild autumn into a second flowering.  This Veronica colostylis is doing just that in a pot at home today, but probably because it was transplanted in February from its home in the Arrow Valley near Queenstown, and it's a bit confused, poor thing.
Veronica colostylis flower (it's about 10 mm diameter).
Veronica colostylis is little more than a small-flowered form of V. linifolia, and in fact when I first described it it was as a subspecies, then under the genus ParahebeParahebe linifolia subsp. brevistylis.  Another botanist, Michael Heads, raised it to species rank later (as Parahebe brevistylis), and David Lloyd and I followed that when we published a monograph on Parahebe (Garnock-Jones & Lloyd 2004), and later (Garnock-Jones et al. 2007) when I transferred it to Veronica (it needed a new name because there was already a Veronica brevistylis).
Veronica colostylis in the Arrow Valley, New Zealand
When compared with V. linifolia, it's clear that the differences are quantitative only.  The flowers are a bit smaller, and the stamen filaments and the style are noticeably shorter.  The flowers have a lot less colour, and the short stamens hold the anthers very close to the stigma.  All these are adaptations to self-pollination, and it's no surprise that in the absence of insects, V. colostylis sets seed, while V. linifolia doesn't.
Veronica linifolia from Arthur's Pass, New Zealand.
So what does this mean about the appropriate rank for this plant?  If the differences are only quantitative (smaller parts with just the same structures), then they don't really provide evidence of any breeding barriers between V. linifolia and V. colostylis.  However, I never had much luck trying to cross them.  The only hybrid I managed to raise looked very like V. linifolia, so much so that I wouldn't have trusted it really was a hybrid except that its female parent was V. colostylis.  That hybrid had 37% of its pollen malformed, suggesting that there are genetic differences that lower fertility in crosses between them.  Additionally, the ranges of the two species overlap slightly in Canterbury, without any suggestion that they hybridize or intergrade there, so I'll stick with species status for now.
I saw V. colostylis again last summer, just downstream from the terminal face of the Franz Josef Glacier, New Zealand.

Garnock-Jones, P.J.; Lloyd, D.G. 2004: A taxonomic revision of Parahebe (Plantaginaceae) in New Zealand.  New Zealand Journal of Botany 42: 181–232.

Garnock-Jones, P.J.; Albach, D.; Briggs, B.G. 2007: Botanical names in Southern Hemisphere Veronica (Plantaginaceae): sect. Detzneria, sect. Hebe, and sect. Labiatoides. Taxon 56: 571–582

Tuesday, 29 May 2012

Wednesday weed: black nightshade

Every so often there's a bit of a fuss over some nightshade berries found in peas, and it's made worse by people confusing their nightshades.  The culprit is black nightshade, not deadly nightshade.  Deadly nightshade, Atropa bella-donna, is very rare in New Zealand.  It's in the same family as black nightshade, but its berries are a bit larger than a pea (12–18 mm) and glossy black (Sykes in Webb et al. 1988).
Black nightshade, Solanum nigrum, Northland, Wellington, New Zealand.
If you do think you've found nightshade berries in among your peas, they're easy to distinguish from peas, even though they're the same size, shape, and, if unripe, colour.  A pea is a seed; it has a thin seed coat and two large hemispherical cotyledons (seed leaves) inside.  It doesn't have a calyx of small leaf-like sepals attached with a stalk.  A nightshade is a berry; it has a thin fruit wall and several small seeds inside it.  It has a stalk and a calyx at one end (see E in the figure below).  If you slice one in half, you'll see its structure is just like a tiny tomato, hardly a surprise because they're in the same genus.  A nightshade berry has two compartments with many seeds attached to the partition that divides it in two.  Cultivated tomatoes often have three or even four compartments; probably they've been bred that way to increase the flesh making a firmer fruit.

Solanum nigrum, from Northland, Wellington. A, leaves, upper surface on left, lower on right; B. leaf hairs; C, flowers, the youngest with a receptive stigma but anthers haven't opened yet, the middle one presenting pollen; D, the smooth stem, E, a cluster of young berries.
Black nightshade and its close kin are common weeds in New Zealand.  Their unripe fruit is just the right size and colour to be very hard to sort from a harvest of fresh peas, although they'd be easy to sort from pea pods before the peas are shelled.

Black nightshade, Northland, Wellington, New Zealand.
The leaves and stems of black nightshade are usually almost hairless, although they have a few scattered minute hairs along the leaf edges.  Sometimes the leaves have a purplish red midrib, and sometimes the whole leaf is reddish.

Black nightshade leaves, underside on left.
So are they poisonous?  There seem to be no documented poisoning cases in New Zealand, and at least some forms of the species are edible.  But the best botanical advice is to treat them with caution anyway, knowing the reputation of the family for some pretty nasty toxins.  However the genus Solanum contains many edible plants—potato, tomato, egg plant, tamarillo, pepino—as well as nasties like Jerusalem cherry, bittersweet, and apple of Sodom.

The native S. americanum is similar to S. nigrum, but it has smaller flowers, shorter anthers, and the sepals are reflexed at the fruiting stage.  I'll watch out for it and post on it when I find it.
Webb, C.J.; Sykes, W.R.; Garnock-Jones, P.J. 1988:  Flora of New Zealand, Vol. 4.  DSIR, Christchurch.

Friday, 25 May 2012

Lobelia flowers

A while ago, I wrote about the spectacular Lobelia physaloides and its relationships to other New Zealand Lobelias. At the time I didn't have good pictures of the flowers, so I illustrated that post with pictures of a fruit.

Last summer I managed to photograph flowers in the Wellington Botanic Gardens (if you're local, you can see this plant beside the path about 20m uphill from the duck pond).
Lobelia physaloides in the Wellington Botanic Gardens.
Also in the gardens I saw a red-flowered plant that I took to be L. cardinalis, the type species of the genus Lobelia.  There are a huge number of ways a big genus like Lobelia could be divided up, but whichever grouping contains L. cardinalis must retain the name Lobelia.

Red Lobelia, male phase presenting pollen.
When these flowers first open, they present their pollen first.  The stigma helps push the pollen out of the joined anthers, and then later it opens out to receive pollen.

Red Lobelia, female phase, stigma expanded to receive pollen.
Apart from their colour difference, you might think these large flowers are quite similar, especially if you compare them to the sort of little white flower that's more common among New Zealand's Lobelias, like this L. arenaria from the Auckland Islands...
Lobelia arenaria, Enderby Island, Auckland Islands.
... or this Lobelia ionantha from the Kettlehole Bog at Cass, North Canterbury.

But molecular phylogenetic research has shown L. physaloides and L. cardinalis are not particularly closely related and in fact their morphology also suggests they're not close.  Some botanists take a wide view of Lobelia and include L. physaloides, and that's the position I took in my earlier post.  However, the earliest branches in the family tree of Lobeliaceae could just as easily be classified as separate genera, and in this case L. physaloides might be called Colensoa physaloides or perhaps placed in another genus.  It seems its nearest relatives are in Australia and Africa, so it's not closely related to our other native species.
Diagram of evolutionary relationships in the family Lobeliaceae (based on Antonelli 2008).  It shows that plants classified as Lobelia are found in many different lineages.  Most New Zealand species are in the C7 clade.
However L. physaloides is classified, one thing is not negotiable for me: it must be classified in such a way that its nearest relatives aren't classified in a separate genus along with species that are more distantly related.


Antonelli, A. 2008. Higher level phylogeny and evolutionary trends in Campanulaceae subfam. Lobelioideae: Molecular signal overshadows morphology.  Molecular phylogenetics and evolution 46: 1–18.

Tuesday, 22 May 2012

Wednesday Weed: Mediterranean mustard

Brassica fruticulosa, Karori, Wellington
Mediterranean mustard (Brassica fruticulosa subsp. mauritanica) is a characteristic Wellington weed.  It's common here, and on the nearby Kapiti coast, with a few collections from Palmerston North.  The first published records, misidentified as B. integrifolia, were from 1957 (Webb et al. 1988).  Plants had been collected a bit earlier than that, from Wellington and Paekakariki.  It's tempting to speculate that this North African subspecies came here with returning soldiers after World War Two, but there's no evidence to support that speculation.
Brassica fruticulosa. A flowers, B ripe fruit, C unripe fruit from valve side, D unripe fruit from replum side, E seeds.
 The uppermost leaves of the common mustards in New Zealand—cabbage (B. oleracea), rape (B. napus), and turnip (B. rapa)—all have lobes that wrap around the stem.  Mediterranean mustard has narrow upper leaves that taper to slender stalks (second from right above).  Its flowers are small and pale yellow, and its fruits are wrinkled around the seeds, no more than 35 mm long, and have a very short gynophore (a stalk above the petal and stamen scars and below the first seed).
Brassica fruticulosa at Roca Grosso, Catalonia
Brassica fruticulosa seems to have stronger resistance to insect pests such as aphids (Ellis & Farrell 1995) and root fly (Jensen et al. 2002) than the cultivated brassicas do.  That suggests this species could be useful in breeding programs as a source of pest resistance in cultivated brassicas, but breeders would need to evaluate the results closely if the resistance is conferred by a toxic compound or a bitter taste, such as the glucosinolates that are found in these plants.
Brassica fruticulosa flowers, Northland, Wellington.
(Footnote:  Just before this post went live, the all-time readership of Theobrominated was 9,993 page views, so one of the readers of this post is likely to be No. 10,000.  No prizes, because I can't tell who you are.)


Ellis, P.R.; Farrell, J.A. 1995: Resistance to cabbage aphid (Brevicoryne brassicae) in six brassica accessions in New Zealand, New Zealand Journal of Crop and Horticultural Science 23: 25–29.

Jensen, E.B.; Felkl, G.; Kristiansen, K.; Andersen, S.B. 2002: Resistance to the cabbage root fly, Delia radicum, within Brassica fruticulosaEuphytica 124: 379–386.

Webb, C.J.; Sykes, W.R.; Garnock-Jones, P.J. 1988: Flora of New Zealand Vol. 4.  Christchurch, DSIR.

Saturday, 19 May 2012


Many people think of botanists simply as people who can rattle off the Latin names of plants.  Charles Dickens seemed to be reflecting this with his character Wackford Squeers in Nicholas Nickelby
  • "... bottiney, noun substantive, a knowledge of plants.  When he has learned that bottiney means a knowledge of plants, he goes and knows 'em.  That's our system Nickelby, what do you think of it?"
Biological identification is the process of assigning a biological sample to membership of a group.  Usually it means giving it a name.  If we're talking about a species level identification, we're making a judgement that the sample belongs to a particular species rather than any other.  Say you see a small four-legged furry animal and you say it's a cat.  What is the process behind that decision?
Rufus, a cat.
First, what is a cat?  In common English language uses, cat could mean one of a number of things.  It could be the biological species Felis catus, the domestic cat.  It could be any one of the other species in the cat family, such as a lion or a tiger.  Or it could be one of a number of other animals that are not strictly cats, such as the meerkat or the civet cat.  In other languages there are other names, le chat in French, el gato in Spanish.  In biology though, the scientific name for the domestic cat, Felis catus, is international and more importantly it's strictly defined by a type specimen, usually a museum specimen that is permanently attached to the scientific name, for reference and stability.  If you try to tell me a meerkat is Felis catus, we can compare it with the type specimen and see if you're right.
But even then, it's not entirely objective, because we're using the process of identification.  Technically we're looking for identity between the type specimen of Felis catus and our meerkat.  And strictly speaking, individuals in the same species are rarely identical (even identical twins have different fingerprints).  So whether two individuals are the same species or not is often a matter of opinion.  Thus we use the term identity in the sense of belonging.
Because of the type system in biology, we can say the type specimen is Felis catus, but every other cat is only identified as Felis catus.  It's a matter of opinion that they belong to the same species, but in the case of cats, and even more of our own species, that opinion is based on a considerable amount of experience and knowledge.  When it comes to organisms that are less like us, it becomes harder to decide if two individuals belong to the same species or not.  Unicellular algae can appear to be identical, yet be unable to interbreed and have quite different DNA sequences.
Branches from this plant near Franz Josef Glacier were collected, pressed, and mounted as a herbarium specimen to be the type specimen of Veronica colostylis (although originally it was called Parahebe linifolia subsp. brevistylis).  Advance of the glacier has wiped out that locality, but plants of the species are still common in the valley, and the type specimen is preserved in the Allan Herbarium at Landcare Research, with duplicates in several other herbaria.
There are many different ways that biologists decide which individuals belong to the same species, sometimes called species concepts.  The best functional definition is the Biological Species Concept (BSC), because this is a genetic definition: a species is a breeding population, or group of interbreeding populations.  Genetically, a species has a gene pool, made up of all the different alleles (gene variants) that are capable of being combined with each other because they're found in individuals that can interbreed.  An allele that's only found in a dog will never combine in nature with one that's only found in a cat, because these two are different species and can't be interbred.
DNA sequencing gel
While the BSC is a useful conceptual and functional definition of a species, it's not used very often to identify individual specimens.  It's simply too bothersome and time-consuming to try all the interbreeding experiments that might be necessary.  Instead, we use a proxy.  Often it's their morphology (the Morphological Species Concept or MSC): if two individuals look the same, or at least very similar, they're reckoned to belong to the same species.  And if we want to give our sample a name, we need to compare it either directly or indirectly (through a description or photograph) with a type specimen.  Recently there's been a lot of interest in using short DNA sequences as identification markers for species, so-called DNA barcodes.
 Morphology works as a proxy for the BSC because we know empirically that members of the same biological species usually look very similar.  But two problems arise with the Morphological Species Concept.  First, species can be very varied.  Sometimes males and females look very different, some barnacles for example.  Other times, juveniles and adults can look so different they're mistakenly classified as different species, as in many kinds of eels.  Sometimes there are polymorphisms like striking colour variants within a breeding population.  In plants and some animals, different forms may be produced in different environments, like the inchworms that grow to be camouflaged on whatever host plant or plant part they're feeding on, or at different times of the year.
Secondly, there are cryptic species, those that look the same but are separate breeding populations.  Sometimes they have physical, timing, or genetic barriers to successful mating, but if we rely on the MSC we'll not discover these.
Veronica odora plants in the North Island and northern South Island (also in the Auckland Islands) have two sets of chromosomes (diploid), but in the central and southern South Island they have four sets (tetraploid).  Probably these two chromosome races can't interbreed and should perhaps be treated as cryptic species, even though I've found no way they can be distinguished based on their morphology.
 So a biological identification is a matter of opinion, but that certainly doesn't mean anyone's opinion is as good as anyone else's.  Taxonomists who study the classification of groups and who understand the nature of their variation and breeding relationships will be more reliable than Wackford Squeers who just "knows 'em".  That's because their identifications are based on a sound understanding of the evidence.  And we can future-proof and error-proof our identifications by keeping a record of what the plant or animal was, preferably as a museum specimen.  That way, if we got it wrong or if the name changes in the future, someone can always check back and bring the identification up to date.  A name in a notebook, publication, or database that hasn't been safeguarded in this way is practically valueless.

Specimens are also important for resolving disagreements about identification.  I remember being called once by a botanist who firmly told me not only had I got the distribution of Sonchus arvensis wrong in a Flora treatment (Webb et al. 1988), but I hadn't described it accurately either.  Fortunately my initial horror and embarrassment gave way pretty quickly to a potential alternative hypothesis, so I asked for a specimen to be sent by mail.  You can guess where this is going, I hope: to my relief, the specimen was Picris echioides.

Tuesday, 15 May 2012

Wednesday Weed: gorse

A couple of my favourite blogs and websites have regular weekly features.  Pharyngula has Mary's Monday Metazoan, Friday Cephalopod, and the wonderful Botanical Wednesday, whereas Why Evolution is True has Caturday Felid.  So I'm going to steal their idea and have a Wednesday Weed, but maybe not every week.  Sometimes I might simply post a picture.  This week: gorse.
Gorse, Ulex europaeus.
When my family arrived in Porirua, where we lived for a few months in 1956, my mum picked some gorse in flower from waste land nearby and put it in a vase to brighten up the house.  Her family had had an ornamental gorse bush in their garden in Stoke-on-Trent, England.  The neighbors thought she was mad, because gorse is probably New Zealand's worst weed.  It was brought here quite early (the first published record is from 1867), probably as a hedge plant, and quickly established and spread.

What makes gorse so weedy?  The young shoots are palatable to livestock, but once the prickles have hardened off they won't be grazed.  The seeds are ejected explosively from the pods, but they only travel a few metres.  Then they can remain dormant for a long time.  We recently took up a concrete path that was laid in the mid-1960s (dated by an old newspaper lodged in the cavity of a concrete block in the adjacent wall) and within months several gorse seedlings had appeared in the newly-exposed soil.  Over a very few years, a few scattered bushes can turn into a dense closed impenetrable canopy.

Gorse is a legume, and legumes have an association with nitrogen-fixing bacteria that enables them to acquire nitrogen easily.  Not a single one of the millions of eukaryotes (organisms that have membrane-bound nuclei and mitochondria in their cells, i.e., including all animals, plants, fungi, and protists) is able to fix nitrogen.  All living things need nitrogen—it's absolutely essential for proteins and nucleic acids—and it's abundant in the atmosphere (about 80%), but it's not available to eukaryotes without being processed first by bacteria.  That's not an intelligent design, but we have to live with it.  Legumes live with it very well by having a permanent association with such bacteria in nodules in their roots, and that can give them a competitive advantage on some sites, particularly in the presence of phosphate (it's worth noting that both superphosphate application and scrub weed clearance were subsidized by the New Zealand government until the mid-1980s).
Gorse at Hinewai reserve, with a fire in the distance.
Burning gorse will kill it (it's highly flammable), but new seeds will germinate to restore the gorse canopy.  But leave it alone and the bushes quickly grow tall and leggy, allowing native plants to establish underneath.  Gorse won't re-establish in this semi-shade, and soon the native forest canopy overtops the gorse one, and a new forest is born.  However, while gorse has been successfully used as a "nurse crop" in this way, such as at Hinewai Reserve on Banks Peninsula, Jon Sullivan, Peter Williams, and Susan Timmins (Sullivan et al. 2007) caution that succession under gorse is different from succession under a native shrub, kānuka (Kunzea ericoides).  They found forests that originated under a gorse canopy had lower species richness, fewer small-leaved shrubs like mingimingi (Leptecophylla juniperina and Leucopogon fasciculatus), and less kāmahi (Weinmannia racemosa) than in forests grown under kānuka.

As alternatives to spraying, burning, and waiting until gorse reverts to forest, Landcare Research Ltd have trialled a range of agents for biological control of gorse, including several insects (moths, thrips, a weevil) and a mite.


Sullivan, J.J.; Williams, P.A.; Timmins, S.M. 2007.  Secondary forest succession differs through naturalised gorse and native kānuka near Wellington and Nelson.  New Zealand Journal of Ecology 31: 22–38.

Tuesday, 8 May 2012

How tall do annual plants grow?

Read any Flora, and you'll find each plant species' description starts with a statement about what kind of plant it is (tree, shrub, climber, or herb) and usually how tall it grows.  You might notice that a lot of the herbs (non-woody plants) are said to grow to about 80–100 cm tall.  Why is that, you might ask.

Arthur Healy (1917-2011), was a weed scientist at Botany Division of the Department of Scientific and Industrial Research, which was the Government's science research arm from 1926 to 1992.  Arthur's extensive herbarium collections of New Zealand weeds are the foundation for our knowledge of introduced plant biodiversity, but he wasn't just a collector, he experimented as well, and importantly he interpreted the significance of his findings for New Zealand.  One of his simple experiments was to plant the fluff that accumulated in his trouser cuffs on overseas trips, to see what seeds had lodged there.  Others were to see what grew out of packing straw and birds' nests.  These experiments demonstrated the need for more effective biosecurity at our borders and showed how weeds spread and got established.  They remind me very much of Charles Darwin's experiments.
Arthur Healy (DSIR)
But back to the weed collections.  Maybe it was a quirky interest, or maybe something that followed a plan, but Arthur collected the weeds that grew in odd urban sites like the cracks in footpaths and walls, or roof guttering.  And some of these were tiny, especially the annual herbs.

Annual herbs are ones that complete their life cycle within a year.  They get only one shot at flowering, so plants that die without managing to produce any seeds don't leave any offspring for the next generation.  Over the centuries, the annuals that succeeded were those that were genetically predisposed to flower and make seeds, however small and under-resourced they might have been.  Some of the weeds from footpath cracks and the shallow dusty soil of roadside gutters were much smaller than anything described in the Floras.  I remember Arthur's specimens of opium poppies just 1-2 cm tall, with tiny flowers a few millimetres across.  One had a seed pod that had room for just a single seed, all that starved plant could manage, but perhaps enough to pass its genes on for another generation.
Cirsium vulgare, called Scotch thistle in New Zealand.
When we were writing Flora of New Zealand Volume 4, Colin Webb and I noted Arthur Healy's collections of tiny footpath crack plants, and wondered about the other end of the range; what were the biggest plants we could find?  It became a common event to come in to work on a Monday and find a huge example of whatever genus one was working on lying on the office floor: Colin would leave giant brassicas or monster thistles that had to bend over at the ceiling of my office, and I'd leave 3m hemlocks or massive nettles covering the floor of his.
Hemlock, Conium maculatum.
When most botanists collect, they like to choose a sample that has flowers and fruits, and is healthy and representative.  Importantly, it also has to fit conveniently on a herbarium sheet, a standard sized piece of card, usually about 420 x 270 mm.  Tall herbs can be folded twice to be mounted on one sheet, but folded three or more times they clutter the sheet too much and make the specimen bulky.  Two folds cause the upper limit for a specimen to be rather less than three times as tall as the herbarium sheet; leaving some space for the label, that is about 80–100 cm tall.  So very tall plants are not well represented in herbaria, so long as there are conveniently-sized ones to collect.  Tiny plants are also not easy to collect.  Just a few on a sheet look silly, while collecting a few dozen and gluing them down in rows is a lot of work when a single larger plant can fill the space with ease.

Thus the two tails of the normal distribution of plant height might often get neglected for a herbarium collection, and this could be reflected in the plant sizes that are published in scientific descriptions.  By selecting the most aesthetically pleasing plants for the collections, are we biasing the information they hold?  (However, I checked a few plant sizes in Flora Europaea and Flora of New Zealand, and I think they're pretty good overall.  They do give the tall extremes at any rate, but maybe don't always include the smallest sizes.)