Wisteria

www.seiryu.freeserve.co.uk

'Wisteria' is a genus of flowering climbing plants.

This page contains some stuff distilled from my experiences growing Wisterias, both as bonsai and in the 'normal' manner, and has some hints about that favourite question, "Why doesn't my wisteria flower?".

Introduction

A wisteria - whether grown as a bonsai or as a garden plant - can have spectacular flowers, for a week or so, but will possess little interest for most of the year. Wisterias have a reputation for being more trouble than they're worth, and for taking many years to start flowering. There is a lot of conflicting advice around, to which I add mine.

Everybody's growing conditions vary, so don't take the advice in these pages too seriously!

Genus

The genus Wisteria is part of the pea family (Leguminosae). The plants have typical pea flowers, with the same structure as those of sweet pea, bean, lupin, laburnum, vetch, and so on.

All wisterias are flowering climbing plants, and are woody and deciduous. Nitrogen-fixing root nodules are frequently found.

Species

Wisterias are found in Asia and SE USA. This geographical split is echoed in a behavioural split.

The American species (W. frutescens, W. macrostachya) are of less interest as bonsai. They have small racemes of flowers, carried on the end of summer growth. The closely related genus Milletia follows this pattern, yet is sometimes used by the Japanese for bonsai.

The Asian species (W. sinensis, W. floribunda, W. venusta (which is more properly but less commonly known as W. brachybotrys)) flower in spring, on last year's growth, before or with the appearance of the first leaves. The flowers can therefore be carried on a visible branch structure which we design and develop (rather than on the outside of an unruly straggling shrub, as with the American species). This can be used to produce an emotive contrast between gnarled old wood and the delicacy of the blooms.

Under some conditions, W. sinensis will also produce a few flowers in summer. This appears to be next year's flower buds opening prematurely, and is correlated with good growing conditions and vigorous growth. It can also be produced by summer pruning.

A similar split in location and behaviour also occurs in magnolias, where the American species flower in summer (e.g. M. grandiflora), and the Asian species flower in spring on bare wood (e.g. M. stellata).

Identification

American vs. Asian

Most plants encountered in Britain are of the three Asian species. The axillary buds are covered by bud scales, giving a sleek pointed bullet shape. The buds on the American species, in my limited experience, lack visible bud scales, and the general shape of the bud is approximately spherical. The fine hairs on the bud look silvery, so the overall impression is of a tiny pearl nestling in the angle between stem and leaf petiole.

Direction of twining.

The best advice I heard on discriminating between the species was that the Japanese species 'follow the sun' - the shoot tip grows in a clockwise direction (when seen from above). Think of the Japanese flag. I simply remember that W. floribunda has a left-hand thread, and W. sinensis has a right-hand thread.

The number of leaflets per leaf is often quoted as a discriminator but any minor differences (and they are usually minor) are masked/can be confused by the difference between juvenile & adult growth forms. However, any wisteria with as many as seventeen or nineteen leaflets per leaf is likely to be W. floribunda, and possibly (I'm not entirely convinced about this, but there seems to be some correlation) is a cultivar with long flower racemes.

Raceme length

What's a raceme? It's the long dangly thing with lots of little flowers along it. It's a modified shoot, bearing flowers instead of leaves. The length varies according to the species:

Floret size

venusta: the silky wisteria

The leaves and shoots of W. venusta are covered with short hairs. They look silvery and feel silky, particularly the young growth.

Hybrids

There are at least two wisterias which have hybrid characteristics: Burford and Lavender Lace. Burford has 16"-17" racemes of sinensis-sized florets; twines like a floribunda; has some of the silkiness or downiness of venusta on its young growth; and has greater vigour and sturdiness (big buds, thick shoots) than the norm. I suspect it has floribunda and venusta in its parentage. My Lavender Lace has not yet flowered, and (so far) looks unexceptional. Burford, however, is definitely worth growing. From its downiness, I suspect that W. floribunda 'Royal Purple' possesses some W. venusta genes too.

Named cultivars

Many cultivars have a Japanese name and one or more English names. The generic Japanese name for Wisteria is 'fuji'. Some Japanese names:

Plants which flower young: Issai; Domino (are these the same?); Royal Purple; Peaches & Cream/Kuchibeni/Lipstick; W. sinensis; W. venusta (?).

Plants to avoid (remember, the author is biased): Rosea, Black Dragon (both take ages to flower, and aren't particularly special), anything labelled 'chinensis' (there's no such thing), anything labelled with the wrong species. If the label is partly wrong, is there any guarantee that it has any accuracy, or even relevance? Unless you recognise the cultivar of a wrongly-labelled plant, or it is particularly desirable for some reason, I suggest that you avoid it.

Propagation & sources

Seedlings

Seedlings show a characteristic bottle shape (rather than a flare, which would be more desirable for bonsai) at the trunk base around the hypocotyl area, for the first few years. The roots are energy storage organs, which is why they are so thick.

Seedlings have a reputation for taking a long time to start flowering.

Cuttings

Cuttings (hard or soft) generally produce the best trunk-base/root flare.

softwood:

Softwood cuttings need lots of heat. In this cool English climate, my best results were from cuttings taken a couple of weeks before a heatwave. First predict your heatwave.

If the normal methods are used to keep them humid (e.g. plastic-bag tent), softwood cuttings will go mouldy (botrytis or similar) and start to rot within 36 hours. Immersing the cutting in fungicide does not prevent this. In order to prevent mould, good ventilation is essential. To provide enough water in the absence of a humid atmosphere, the cutting can stand in water. Reducing the leaf area reduces the water demand and allows the cutting to survive long enough to root. I found that two leaves, each reduced to two pairs of leaflets, was about right, for my conditions. Then the cutting can be planted in compost once it has decent roots.

hardwood:

These should be taken soon after the autumnal leaf drop. Midwinter is too late. I had reasonable success with cuttings about 9-12" long, in groups in tall pots. I kept the tops cool and provided bottom heat, in order to get callus (and root?) formation long before bud-burst brought its water demands. In order to keep the roots from suffocating, immersion watering was used to provide a change of air and keep the soil oxygen levels adequately high. This requires a compost which retains its air spaces under such conditions. Some (most) composts will collapse into a waterlogged mass. I found Seramis - fired clay chippings - to be adequate for this. Care (or mesh) over pot drainage holes is needed, or it'll all wash out.

The leaf shape just after bud burst is an indicator of the presence or absence of roots (i.e. whether the cutting has 'taken' or not). If the first leaves curl outwards and downwards in a tight arch, the cutting is probably a failure. Leaves on rooted cuttings expand without significant distortion in their rachis (the spine of the leaf). Remember, the first few leaves are already formed in the bud. The process is mostly one of cell expansion, not cell division. The necessary water is only provided if there are enough roots. Leaf deformity is correlated with root problems in many types of plants.

The first roots on cuttings are very brittle, and the callus not firmly attached. Be careful if you pot up early; it may be better to leave it awhile.

Root cuttings

Root cuttings work well but need to be at least pencil thickness. Half an inch diameter is better. Set them at an angle and cut the top at an angle to reduce the number of shoots produced and create a plant more useful for bonsai. This method generally produces a very asymmetrical plant.

Air layering

There are various options. Either remove bark, either in a complete ring or leaving a lifeline, or use a wire loop tourniquet. Leaving a lifeline will lead to an asymmetrical root structure. Compost should be as airy as possible: sphagnum moss is probably the best. Flood-and-drain watering should keep the oxygen content high, although beware of compost compaction. Warmth will help to start roots growing: covering the layer in black plastic over the clear plastic holding the compost should provide this. Because of the species' thirst, it is essential to check the layer every few days. Once root growth starts, it can dry out the layer surprisingly rapidly. This can kill the new roots, and also the layer if a complete bark ring was removed.

When a rooted layer is severed from its parent, its water requirements can be made more commensurate with its suddenly reduced supply (no more sap is available from the parent) by reducing each compound leaf to two pairs of leaflets. Removing tips of growing shoots can help the new plant to establish by redirecting growth to the roots.

Grafts

Grafts in the horticultural trade are 'nurse grafts', purely for propagation purposes. The rootstocks are seedlings, and have not been selected to provide any properties beyond sustaining life. Delayed graft incompatibility is, sadly, quite common.

Treatment: Plant deeply with the graft union below soil level, and hope for scion rooting; layer. Whip-and-tongue grafts (which are fairly common) tend to swell lopsidedly. This is fine for slanting styles where it can be used as a part of the trunk flare and the defects can be concealed. Otherwise, the plant can be layered, either in the normal way or more crudely with a loop of wire around the trunk a couple of inches below soil level. Select the level at which roots are desired; apply the wire loop, or remove bark; and replant deeply enough that the treated area will not dry out.

Transplanting a garden plant

Don't be afraid of cutting back hard, either roots or top. Ideally transplant in autumn or just before bud burst in spring. If the plant is in leaf, keep the leaf area within the ability of the roots to supply it with water (see section on severing air layers, above). Many Japanese wisteria bonsai started this way, or from stumps collected from the wild.

After propagation

After propagation by any method the plant will tend to produce juvenile growth. Expect a delay of at least four years (though this depends strongly on the size and vigour of the plant, the cultivar, and its treatment) before flowering takes place. There are a few exceptions: for example, an air-layer of a flowering branch will flower with the buds laid down before it was severed. It will generally fail to flower for a few years afterwards, however, because it has insufficient ability to store energy. Another possible exception to this is a graft of flowering wood onto a large and mature stock. This is apparently a Japanese technique, which I have never seen used in England.

Growth pattern

Many woody plants have a spring growth spurt (e.g. maple). After bud burst, there is perhaps a month of growth, followed by a period when the plant's energies go into fruit, wood, and next year's buds. Terminal buds form at the ends of the twigs. Thus there is a point beyond which the twig will not extend until some stimulus (usually temperature or day length the following spring) causes the bud to grow.

Levels of growth hormones in a typical woody plant are higher during the growth period, and reduce afterwards, which is partly why the growth after summer defoliation has smaller leaves and shorter internodes. Wisterias, however, grow continuously, and do not form terminal buds. It might be reasonable to expect from this that hormone levels stay high, and that defoliation would therefore not have its usual dwarfing effect on the replacement leaves.

In autumn, when growth slows, the shoots die back to the point where the wood and buds are mature ('ripe'). Next year's growth is from axillary buds, since there are no terminal buds.

Climbing vs climbing-prevented

If a wisteria finds something to twine around, it will climb it. Until it runs out of things to climb, or reaches the highest point it is able to sustain (6-10m above the roots, at least!), or reaches a substantial size, it will not begin to flower vigorously. Only when it runs out of things to climb, will it stop climbing - and only then will its energies go into reproduction.

When it comes to the top of its support, the long shoots cannot support themselves as they grow further, and will unavoidably curve over and descend under their own weight. These will tend to die back to near the highest point, or at least to lose vigour. The upper or remaining part will be more likely to bear flowers in the following year than if the shoot had found something to climb.

We can promote flowering by artificially using this mechanism.

Training patterns for garden use

From the above, it should be apparent that garden training should mimic the effect of the plant's reaching the top of its support: T, F, or upside-down L shapes will all work much better than V or Y shapes. And M, N, and W shapes (I've seen these tried) don't often work at all! Gentle flattened W shapes can be formed from a mature plant, for example to train it along the railings of an ornamental bridge.

Bonsai Styles and Training Techniques

Geometry and flowers

The shape should be designed to show off the flowers. Avoid having them trailing all over the surface of the compost. Either have plenty of height or let them dangle clear of the side of the pot.

It might be considered artistic to have the very tip of one raceme brushing the moss.

Heavy trunks

Wisterias fatten up fairly rapidly. This means that a heavy trunk is a common feature. It would be a waste to hide this when the plant is in flower, so arrange the flowering branches accordingly.

Tendency for wood to rot

The wood rots quickly in the cool damp English climate. Really old plants are often reduced to a few strips of living material, with gaps between where the majority of the wood has rotted and disappeared. This is rather less apparent in hotter drier climates!

Contrasts and artistic waffle

My personal preference is to have all the elements as close as possible, increasing the visual impact. This also results in the flowers appearing disproportionately large, which adds to the impact. In order to avoid a confused image, the elements of the composition may need to be reduced and kept separate.

Pot size & shape

Pots should be fairly deep. A height to width ratio of 2:3 works well for me. Shallow pots seem to lead to weak growth and root problems. Very deep pots (cascade pots) must not be too large for the plant.

With some other plants, fast growth can be achieved by using a large pot. With wisterias, this treatment seems to lead to stagnant compost and root problems. Better growth is achieved in more modest pots, or (given good soil) in the ground. This is probably a function of the compost which I use.

Care regime: repotting

The best time to repot a wisteria in training is just before bud burst. This leaves a full uninterrupted growing season. It will, however, generally cause the tips of the new shoots to absciss (die and drop off) after about an inch of growth. Because of this, it is not a good thing for a wisteria which is about to flower, because it will make the flower buds abort. In this case, repotting immediately after flowering is more sensible. A possible alternative is to repot in autumn, at or around leaf drop time, but I have no experience of this. Spring repotting does have some use for encouraging ramification.

It is not essential to prune the roots. Some of the feeder roots will have died and rotted, and will come away with the old compost. As much of this material should be removed as possible. Heavier storage roots may need to be cut in order to fit the rootball into a smaller pot. The cuts should be as clean as possible, in order to avoid rot. I use a scalpel or a sharp knife for this. The cut ends will not heal over. These trimmings from fat roots are a useful source of root cuttings. I also tend to remove excessibely long feeder roots, any circling storage roots, and any roots showing signs of rot.

Bottom heat?

Don't apply bottom heat to anything except cuttings. I tried this once after a severe pruning of both top and roots. The result was that the roots left dormancy before the top did, and the plant bled to death.

Care regime: compost

The compost should be heavy, moist, water retentive, not airy. This is altogether unlike a typical bonsai compost. I use compost made up from pre-fertilised bricks of compressed coir. I used to add perlite or vermiculite but it doesn't seem to help noticeably and makes repotting more difficult.

I have seen peat used, and have been advised to use lime. Miracid (an acidifying fertiliser) seems to do more harm than good, but wisterias grow happily in areas with acid soil as well as on limestone.

Care regime: feeding

Feeding: lots! The coir which I use contains fertiliser, which seems to last for 1-2 years. Sufficient (meaning plenty) feeding is needed for flowers.

Care regime: water

The summer growth of a healthy wisteria is very rapid. At this stage, it is very thirsty. I keep the pot in a dish of water, up to 2" deep, and let it almost run dry before refilling. A Japanese article suggested using a dish with a slow leak, emptying over say 24 hours. This enforcies water level changes even when the plant is less thirsty. Such an engineered leak might be very useful if an automatic watering system is used.

Don't keep the pot in water when the plant is not in rapid growth, or you risk killing it. Water level changes draw fresh air (meaning oxygen) into the compost. No change leads to stagnation and suffocation.

If the pot is allowed to remain very wet - waterlogged - over winter, the plant will suffer root damage, and is quite likely to be killed. A typical example is a plastic training pot (flower-pot) sitting on concrete: it just won't drain.

Care regime: light levels

Wisterias benefit from being in full sun, or as near as is practically possible. In deep shade twigs will die and the plants will stop flowering.

Growing in the ground

Using the natural growth patterns

If a growing shoot finds itself pointing downwards, it will have a tendency to die back to near the highest point. The remaining part will be more likely to bear flowers in the following year than if the shoot had found something to climb. Thus we can tuck long shoots under the pot to promote flowering.

Mature material will be more likely to survive being forced downwards than a young shoot, but will tend to sprout from some high point, which may be followed by death of the cascade section. A cascade design must be approached with considerable thought and some caution.

Alternatively, if some part of the plant needs to be encouraged, its shoots can be given something to climb. This will mean plenty of shoot growth, bearing more leaf area, and the shoots will fatten. This may cause other areas of the plant to become weaker and perhaps even to die back (more likely in shade).

Pruning

Winter pruning is followed in spring by many buds bursting and growing short shoots. Only a few (the number depends on the size and vigour of the plant) of these will dominate and continue to extend. The rest will lose their growing tip after a few leaves, forming a short spur.

Pruning during summer - removal of a dominant shoot - simply leads to a different dominant shoot growing and fattening.

Flower buds

In late winter and early spring, it will become possible to distinguish between flower buds and shoot buds. Shoot buds retain their streamlined shape, while flower buds swell. The contents will become visible, eventually looking rather like a Georgian stone pineapple (sorry, I don't have a better description!) as the tightly-packed floret buds are revealed in their spiral pattern. The time at which the flower buds begin to swell depends on their situation and the temperatures. Against a sunny south-facing wall, flower buds may be swelling by early January. In a cold area, this may not be apparent until late April. If pruning is left until this stage, last year's growth can be trimmed back to the flower buds. This applies equally to bonsai and garden plants.

Ramification

Few shoots are active at once; aim for branches, not twigs; compound leaves will give an impression of ramification and conceal its absence. If fine twiggy growth is achieved, I would not expect it to survive intact for more than a year or two. Furthermore, thin growth tends to be weak growth; strength is required to bear flowers. This section needs rephrasing.

Budding back normally only happens from sites which once had a bud (e.g. leaf axils, branch collars). These sites will spread (increase in area) as the branch or trunk swells.

Wounds

Wounds don't heal. The wood around a wound will slowly grow out past the dead area, leaving it as a sunken hollow (uro). Eventually the wound may close, depending on its geometry, but only after the surroundings have increased in thickness by rather more than the size of the original wound. This behaviour is important when considered in combination with the way the wood rots.

Water shoots

Sometimes a vigorous plant will send straight fast-growing shoots out from low on its trunk. These will tend to go horizontally, have vestigial or no leaves, and (particularly in W. sinensis) are often strongly pigmented. They show none of the usual tendency to twine and climb, and may extend along the ground for several metres. Their apparent purpose seems to be to extend the size of the plant. Climbing shoots grow from them in subsequent years. Despite growing in contact with the ground, they do not appear to root themselves, and are not good candidates for layering. If such shoots appear on a bonsai, they should be removed as soon as they are seen.

It may be possible to reduce the frequency of reccurrence of this form of shoot by removing the areas of burr (and the underlying cambium) which form at their base.

Suckers

Sometimes a vigorous plant will produce a shoot from the roots. This is more likely to happen when the top growth is being pruned, or when long shoots are being forced downwards. These shoots are useful for propagation, if they can be separated during repotting with some attached root. Whether used for new plants or not, removal should include any 'collar' at their junction with the root. This material would otherwise be a source of more suckers.

Suckers may also appear if root damage (rot, vine weevil larvae, etc.) has separated or partly separated a section of root. In effect, the damage has created a root cutting.

Pests

Bonsai summary

Things you don't do to a wisteria bonsai: pinching, defoliation, put it in a shallow pot.

Bibliography

For growing wisteria as bonsai, some good sources are:

Finally...

Remember, any idiot can put up a web site. This one is proof...