Gloriosa superba

Gloriosa superba blooming in my garden after over-wintering in the ground.

When a plant has a name like Gloriosa superba, you expect something special, and this plant does not disappoint.  It is surely one of the most beautiful of all flowers, with its upturned petals in various shades of red, orange, and yellow that give it the common name of flame lily.  G. superba is native to much of southern and tropical Africa and tropical Asia, and I have been surprised to discover that it is reliably hardy in my piedmont garden. My batch of seed-grown plants are currently blooming, both in a large pot and (as seen above) in a sunny, fairly dry flower bed.

G. superba is a deciduous vine which grows from elongated underground tubers.  The leaves have twining tendrils at their tips that allow the vine to scramble up and over other plants or to cling to a trellis.

Gloriosa leaf tip

The vines grow rapidly during the spring and put on a wonderful show over a fairly long period during the summer.

Gloriosa superba climbing our deer fence

In the autumn, the vines die back, and the plant goes dormant for the winter.  In the garden, the dormant tubers seem to be able to tolerate some moisture, although the soil should probably be well drained.  In pots, they can be stored completely dry.  Sometimes I remove the tubers from their pots and store them all winter long in a paper bag.  I have also left some in a large pot of dry soil, unwatered for five months.  The tubers don’t seem to care either way.

Although its common name is flame lily, G. superba is a member of the Colchicaceae, not the Liliaceae.  However, the form of its flower is very similar to that of many true lilies.  For instance, this Lilium leichtlinii flower also has upswept petals with a ring of anthers below:

Lilium leichtlinii, a Japanese lily species that is also blooming in my garden this week

I suspect that the similar morphology reflects adaptation of both plants to pollination by butterflies.  An odd aspect of the floral anatomy of G. superba is the angle at which the style attaches to the bottom of the flower:

Note the style bent up at an acute angle from the ovary at the bottom of this G. superba flower.  As the blossoms age, the yellow markings on the petals turn red.

Some lilies have curving styles, but I’m not aware of any that bend to this extent.  Consequently, the flower can only be pollinated by a butterfly approaching from one side.  The flower is basically radially symmetric and has anthers pointing in all directions, but there doesn’t seem to be anything to guide a pollinator to the stigma.  The style does generally point away from the stem, though, and perhaps that is sufficient to get the stigma into the flight path of the pollinators.

Kew suggests that G. superba may also be pollinated by sunbirds, but I’m not sure how plausible that is.  Most sunbirds perch to feed on nectar, and a sunbird clinging to the stem of the Gloriosa vine or its support plant would be on the wrong side of the flower to effect pollination.  An alternative explanation is that sunbirds visit G. superba to seek insect prey [1].

G. superba tubers are readily available in the spring from bulb vendors, but when I grew plants from purchased tubers, color break in the flowers suggested that they were virused.  I subsequently obtained some virus-free seed and found it fairly easy to germinate.  The first seedlings bloomed two years from germination.  The plants have the potential to become invasive weeds in mild climates, so be careful about planting them out in the garden.

One final word of warning:  G. superba produces an array of potentially lethal alkaloids, including colchicine, and all parts of the plant are highly toxic.  Avoid contact with the sap, don’t allow children to pick the flowers, and keep pets that might chew on leaves away from the plants.  The tubers are particularly poisonous, so don’t store them with your sweet potatoes, OK?


  1.  Cheke, RA, Mann, CF, Allen, R (2001)  Sunbirds:  A Guide to the Sunbirds, Flowerpeckers, Spiderhunters, and Sugarbirds of the World, Christopher Helm Publishers, London, page 314.

Bottlebrush buckeye

Flowers of bottlebrush buckeye. Judging by the length of the inflorescence, this is probably Aesculus parviflora var. serotina.

I first became aware of Aesculus parviflora, the bottlebrush buckeye, while reading Rick Darke’s gorgeous book The American Woodland Garden (Timber Press, 2002).  Darke shows large, mounding shrubs covered with white wands that look fantastic at the edge of a lawn or among tall tree trunks.  I envisioned an arc of shrubs, eventually growing into a single mass, at the southeastern edge of my garden.  When viewed from the living room windows, they would create a visual frame at the bottom of the lawn, separating it from the vertical trunks of the pine trees beyond.

Well, we’re not there yet.



Part of the problem is that I seem to have obtained Aesulus parviflora var. serotina which has very long, spectacular inflorescences (up to 60 cm long on my plants) but which tends to grow more upright and doesn’t always have leafy branches down to the ground.  However, the main difficulty is that the place where I wanted them is not the best spot for growing them.  The ground is some of the driest, hardest, poorest soil in the garden, and it seems to be where gravel was dumped during construction of the septic system.  The southeastern edge of the lawn is shaded in the morning but bakes in the sun during the late afternoon, the hottest part of the day.  Grass has pretty much given up, and the “lawn” here consists largely of moss and pine needles.

Despite these difficulties, the A. parviflora that I planted eight years ago now bloom reliably and slowly grow larger.  What I really need to do is edit out some of the tree saplings and brambles that have sprouted among and in front of them.  I’ll dig up the young willow oaks and tulip poplars, but I think the sourwood at the front of the buckeyes can stay.  I hate to destroy such a beautiful and bee-friendly native tree, and once mature, I think its white flowers will complement the buckeyes quite nicely.  I’ll try limbing it up as it grows, until all of its branches are well above the buckeyes.  Then, regular application of a good thick layer of hardwood mulch should help the soil and inhibit sprouting of new saplings.  Maybe in five or ten more years, I’ll have something similar to the pictures in Darke’s book.

Aesculus parviflora is another native-but-not-really plant.  It has a limited range, primarily in Alabama, but is cultivated more widely.  USDA shows it as having a disjunct native range that also includes Pennsylvania and New Jersey, but BONAP’s North American Plant Atlas suggests that the northern records are of naturalized plants.  The flowers attract bees, butterflies, hummingbird clearwing moths (Hemaris thysbe), and the occasional hummingbird, making them ideal for gardeners who care about pollinators and local wildlife.  The plants seem to be generally pest-free, although Japanese beetles do like to munch on the flowers.  Unfortunately, flowering in North Carolina coincides with the peak of beetle season.

Six on Saturday #2

With just under an hour left of Saturday,  here are six pictures from the garden today.  Lots of yellow this week.

1. Alstroemeria ‘Konkajoli’

Alstroemeria ‘Koncajoli.’

I have tended to avoid Alstroemeria hybrids, because many are reported to be invasive. This was advertised as a civilized cultivar that doesn’t take over the flowerbed. After two years, I’m starting to wonder if it is too civilized. It seems to produce just one stem at a time and shows no inclination to form a nice clump. Pretty flowers, though.

2. Hydrangea quercifolia

Hydrangea quercifolia

When I blogged about oakleaf hydrangea a couple of weeks ago, I mentioned that the flowers would soon fade to pink.  Well, they have.

3.  Verbascum chaixii

Verbascum chaixii, nettle-leaved mullein

Unlike many mulleins, V. chaixii is a perennial rather than a short-lived biennial.  I have both the yellow- and white-flowered forms growing in the sunnier areas of the garden. Hoping for volunteer seedlings but haven’t seen any yet.

4.  Ligularia japonica ‘Chinese Dragon’

Ligularia japonica ‘Chinese Dragon’

This species isn’t thrilled with our hot summer, but it seems to do reasonably well in the shade of a dogwood tree. The deeply cut leaves and big yellow flowers are interesting, but I really like the flower buds.  They are ribbed longitudinally and look like miniature green pumpkins.

5. Canna ‘Tenerife’

Canna ‘Tenerife’

Canna season has started, which means lots of bright flowers, lush foliage, and constant checking to make sure that caterpillars of the lesser canna leafroller moth (Geshna cannalis) aren’t feasting on leaves that they seal  with a loop of silk before the young foliage can unroll.  The leafrollers are ugly, maggoty-looking things that skeletonize canna leaves and make a mess with their frass.  I hate to use pesticides on plants that attract so many pollinators, so I have to squish the caterpillars by hand.  Yuck.

6.  Canna ‘Pacific Beauty’

Canna ‘Pacific Beauty’

The flowers of this clone look almost fluorescent against the dark brown/purple foliage.  In this climate, cannas can be left in the ground year round, as long as the rhizome is planted six or eight inches deep and mulched well in the autumn.