A Northumberland Alpine Gardener's Diary
This entry: 31 January 2013 by John Richards
New Zealand Celmisias. Entry 234.
Celmisia is perhaps the best known of all the alpine genera in New Zealand. The flowers are effectively conventional daisies, but they are borne over perennial foliage which tends to have metallic tones of silver or gold, and include species which are amongst the most attractive of all alpine foliage plants. It is an entirely Antipodean genus, and in New Zealand in particular it has undergone spectacular adaptive radiation over the last few million years so that some 60 species can be found there. This makes it the most diverse of all New Zealand plant genera. Such diversity amongst such attractive plants spawns specialists and enthusiasts, whether in the garden, or amongst those who walk New Zealand mountains.
In fact, nearly all the species are limited to South Island. Several species including C. spectabilis and C. gracilenta also occur on North Island, but the only species I can only think of which is restricted to the North is C. major. Outside New Zealand, a few species are found in the Subantarctic Islands, and about eight species (the number is controversial) are found in Tasmania and the Australian Alps. In fact, some of the best garden plants occur there, and I find that C. sericophylla, C. pugioniformis and C. longifolia grow well in my conditions.
Often the most difficult part of growing celmisias is acquiring authentic material in the first place. Much seed is sterile, lacking embryos, and even fertile seed often has a short viability. It is best stored in a 'fridge as soon as collected. Also, all species are self-incompatible, so that if seed is set in the garden and only one clone of a species is grown, the seedlings will be hybrid (they may be very attractive however!). Many garden plants are thus misnamed. As soon as you acquire fertile seed, it is best sown immediately, even if received in May, as is often the case with wild collected seed. It is best covered with grit and left outside in all weather. It should not be discarded for at least two years.
Vegetative propagation is easier. After flowering, take non-flowering shoots and strip the old and marcescent leaves from the base and insert the cutting into peat (or equivalent) which is never allowed to dry out. It should be left in a cool place out of the sun. This works with the large-leaved species too which form pseudostems and which can root readily.
I think that Celmisia is a genus which has become harder to grow well as the climate has changed. I would place Cassiope, Phyllodoce and Raoulia in this category too. However it may be merely that they don't like my current sheltered garden. Near here, but three hundred feet higher in an exposed site and with much better draining soil, Alan Furness grows the best collection of celmisias anywhere, probably eclipsing any in New Zealand itself. This is an alpine genus which likes good drainage, good cool air-flow and which should never become dry at the root. Nevertheless, in the wild many species are protected from extreme cold under deep snow and are not that hardy. I lost most of my collection, some 30 species, on one night in January 2001 when the thermometer dropped below -20C with no snow on the ground. I rather lost heart after that and now onlty grow about 12 species.
I am starting this account of plants in the field with the large-leaved, pseudostem-forming species. By far the best known of these, mostly because it is widespread (and variable), and grows well in some gardens, is C. semicordata. This familar species has a complex taxonomic past. For many years, plants in the UK were known as C. spectabilis which is a much smaller species with leaves green above and woolly flowering stems. When, finally, they were identified correctly, the species was known as C. coriacea. However, the expert of the genus, David Given, showed that the type of C. coriacea actually represented a rather local species from remote parts of the very wet south-west of South Island which had been called C. lanceolata. Thus, the latter name had to be lost, and a new name found for the familiar widespread garden plant now known as C. semicordata.
C. semicordata is nevertheless a very variable plant, particularly in the colour of the pellicle, which is the name given to the close shiny almost skin-like membrane of hairs found in some of these species. In some eastern forms from Canterbury this is golden. These tend to be variants of the narrow-leaved subspecies stricta, and they have become popular as foliage subjects at Shows, although not always easy to keep in pots (and resenting continuous upheaval!).
Here is a typical example of subspecies stricta from herbfield (the equivalent of alpine meadow) at about 1000m altitude on Arthurs Pass, west Canterbury.
Perhaps the most famous example of C. semicordata was a plant grown in Ireland and distributed by its introducer, the eponymous David Shackleton. This was reknowned for its intensely white, rather broad foliage, and was a very beautiful plant. It became weakened through continued vegetative propagation, and probably no longer exists, although noted Irish growers such as Helen Dillon and Harold McBride certainly grew it until recently.
When we were in the south-west, based on Lake Te Anau we took a helicopter ride onto a remote mountain in one of the wettest zones in Fjordland, Mt Burns. We walked down from the summit to the roadhead, and not long before rejoining the bus I found a lovely C. semicordata which closely resembled 'David Shackleton'. Consequently I have a suspicion that it was originally raised from seed collected in this area, and it would be nice if such seed became available again.
The much rarer and more localised C. coriacea, mentioned earlier, is also found on Mt. Burns. As befits its earlier name, it has long lanceolate leaves and is more a relative of C. armstrongii or C. petriei, although with leaves of a different colour. I thought it very handsome, but it is scarcely in cultivation and probably requires a wet climate.
C. armstrongii itself is more a plant of the west coast, although it too is found in wet areas. Nevertheless, it can be found in such accessible areas as Arthurs Pass. Like the Natterjack Toad, it has an orange stripe down its back! It is still in cultivation, although very hard to come by.
Further north still, in the mountains of the north coast of South Island in Nelson and eastern Marlborough is C. monroi. This is essentally a smaller, silver-pellicle, rather clump-forming relative of C. semicordata with leaves of a subtly different shape. As befits its less extreme climate it is a rather better grower in most gardens and I suspect that many of the silver-clumping celmisias, such as 'Egglestone Hall' found in our gardens today are either this, or hybrids between C. monroi and C. semicordata. Here it is on Mt Arthur in Nelson, not to be confused with Arthur's Pass!
The last of the large-leaved species I am showing has a remarkable distribution, for it is found bpoth on Mt Arthur, in the very north of South Island, and on Mt Burns in the very south, but scarely at all in between. This is C. traversii, reknowned for its beautiful rusty-orange hairs beneath the leaf. For a time it was thought in the UK that individuals in which the rusty hairs were confined to the leaf edge were hybrids, perhaps with C. verbascifolia, but if you visit the wild populations it becomes clear that only a small percentage of plants have the desirable orange backs to the leaves. Here is one of the better forms from Mt Arthur.
In my experience, plants from the deep south are not as attractive as those from the north, and it could be argued that they are not quite the same. It might be possible to separate the geographical races as subspecies?
The other two main growth habits in this varied genus are subshrubs, and mats/cushions. Some species lie somewhat in between these two forms, including two of the most popular garden species, C. allanii and C. incana. Like many of the more amenable garden plants, these species are typical of the drier east, mostly north of Christchurch, and we saw both species growing on Mt St Patrick, a rather dry mountain in Marlborough. On the year of our visit (2003), both species had decided not to flower at all, a common phenonemon amongst the species from the drier side of the island. There is some controversy in this country about the correct naming of plants in cultivation, but in my view C. incana has the narrower leaves with a silky indumentum, whereas C. allanii has broader, more oblong leaves with a more furry covering. Both can make huge mats. Here is C. incana first, followed by C. allanii.
I mentioned Mt Arthur, and the ubiquitous C. spectabilis. Mt Arthur is also home to a very localised small endemic which is effectively a very dwarf C. spectabilis, but to some extent tending towards a mat-forming habit. This is C. dubia. I grow this rather rare plant in a trough, where it has settled down and flowered for the first time last year. Here it is on Mt Arthur.
Two more mat-formers from Mt Arthur follow (you will have gathered that this is an ace locality for the northern endemics). First is the slightly dowdy and well-named C. discolor, a slightly woody mat with rounded grey leaves which grows well with me at home.
Much more exciting to my way of thinking is C. dallii. This is a mat-former with rather thin leathry leaves related to the lovely C. hieracifolia (also a northerner) and the southern, more silvery C. holosericea. In good forms the leaves are tinged orange below. I was really excited to see this lovely and scarce plant.
There are a number of exciting woody subshrubs on the great Whalebacks of Otago, far to the south. Perhaps the richest is the Old Man, here species such as C. haastii cover the ground.
On our visit in 2003, many of the species had declined to flower, but my first trip there in 1979 was remarkable in that all the plants had formed enormous mats of flower. This poor photo of the shubby C. ramulosa is worth showing merely to demonstrate the hundreds of flowers on each mat.
The sticky, shrubby C. viscosa was not flowering on the Old Man in 203, but it did flower on the Pisa Range near Queenstown, and here I found that almost legendary sport, a pink Celmisia!
One of the shubbiest species comes from the south-west. C. hectori has wonderful silver-plated spiky foliage, and so is popular in cultivation, although not always easy to manage.
I am finishing with four of the smaller species. First, one of the most widespread, C. gracilenta. This is quite common in gardens and sometimes is found in foliage classes in Shows. In the best forms the leaves have a curious, attractive pattern rather like a snakeskin. More often, as in this plant from Arthurs Pass, it is more plain.
In many ways, C. alpina is like a reduced form of C. gracilenta, but the leaves are erect, and it links with C. laricifolia. However, C. alpina has a distinctive habitat which rather belies its name as it is typical of sopping set bogs at moderate altitudes. Here it is at Arthurs Pass, another ace site for celmisias with at least 10 species present.
The last two species to be discussed here are the two most cushion-forming species, and these used to be cultivated quite frequently, although they are rarely seen today. These, in particular, seem to be victims of climate change. C. sessiliflora is by far the most widespread and can be found forming mats on summit plateaux throughout South Island. Unfortunately it is rather untidy in flower but makes a lovely foliage plant.
Finally the lovely C. argentea, effectively a very reduced C. sessilifolia, and occurring with it on the summit plateau of the Old Man, but always in much wetter places. Again, very much a bog plant, and this may give a clue to its successful cultivation. Certainly, when I grew it is seemed to be happiest in a peaty compost in a plastic pot.