Helen Davis was a member of the Triangle Bonsai Society in Raleigh, North Carolina. Her name has appeared several times in this Journal because she was one of the very first people to donate some of her personal bonsai collection to The North Carolina Arboretum, back in the mid-1990s when the Arboretum first got into the little tree game. The focus of this entry is a Satsuki azalea Helen gave us then.

Satsuki is the name given to a popular group of cultivated azalea varieties derived mostly from Rhododendron indicum, an azalea species native to the mountains of Japan. Japanese plant breeders have been working with these azaleas for hundreds of years. More than a thousand different varieties have been developed in that time, mostly for shrubbery in landscapes, although Satsukis have found great favor as bonsai subjects too. The name Satsuki refers to May, the fifth month, which is when this type of azalea usually blooms. Americans pronounce the word as “sat-SOO-kee”, but in Japan they seem to say “SAT-skee”.

Before Helen Davis made her donation, she sent us photographs of the trees she was offering. This was one of them:

The image above shows a relatively young plant, yet it is evident the specimen had been in development for a number of years. The trunk line, although still thin, has been fully described from base to apex. The primary branch pattern is mostly in place.

After the donation was accepted, this azalea was planted in a bonsai pot and work continued on filling out the branching. This next photograph from four years later was the first made of the specimen as part of the Arboretum collection:

Helen told me that Yuji Yoshimura had designed the tree in a workshop. She was a great admirer of Mr. Yoshimura and knew that I was his student, so she passed along that information knowing it would make a difference to me. The design certainly falls in line with classical bonsai styling. The slanting trunk has back-and-forth movement and the alternating branch pattern, with the largest branch in the lowest position and all the others above it gradually reducing in length and heft, forms a canopy in the shape of an asymmetrical triangle:

For many years I followed closely the design of the tree as it was when we received it. I didn’t know much about growing azalea bonsai, so at first I took a conservative approach to pruning this specimen. The end result, as seen in this next image from four years later, was that the branching became leggy:

There was also a noticeable degree of creep taking place in the top of the tree. “Creep” is the term I use for changes naturally occurring in small increments over an extended period of time that aren’t initially recognized. Then one day you see plainly that things have changed and wonder how long it’s been going that way. The apex in the above image can be seen in the act of curving slowly to the right, contrary to the lean of the trunk which is clearly to the left. There is no law against this look. Trees in nature often have similar construction, but compositionally speaking there is a good argument to be made against it. It was Mr. Yoshimura who first introduced me to this premise.

The next photo, made after a spring styling session six years later, shows the result of steps taken to address both the legginess of branching and the rightward drift of the apex:

Here is an image made the same year as the one above, but at the end of the growing season:

I was using wire to reposition branching in order to fill gaps. All the branching had a disagreeable tendency to swing upward at the tips, yet I didn’t want to have to wire the entire tree because I didn’t have time to do that. In the next image, made three years after the prior one, it can be seen that I was using pruning as a means of addressing the upward growth habit of the lower branches. The result was a sort of flattening of the foliage masses in that area. I didn’t do the same with the branching in the upper region because I was starting to have the idea that branches toward the top of the canopy should in fact have a more upward nature:

A month after the preceding photo was made, the Satsuki produced its first substantial floral display:

It will be noted that the above image of the azalea in flower was made in the hoop house and not in the garden. This specimen had not yet been put out on public display because I had never found its appearance satisfactory. In my eyes, the structure of this tree was awkward and produced an uncomfortable feeling every time I looked at it. The flowers made it pretty, but pretty is a surface effect and quite a temporary one at that. The problem, as I was at the time beginning to realize, was that this bonsai had been designed to look like a bonsai. I wanted to figure out how to design bonsai that looked more like trees in nature.

My efforts toward that end can be seen in the next image:

The tree as it was when we first received it had numerous distinct foliage masses spaced apart from each other. Partially, this was an effect of the early stage of training the specimen was in; the young plant needed more time to fill out. There is, however, a desire on the part of many bonsai growers to promote this sort of development, a stylistic device usually described as creating foliage pads or clouds. The resulting appearance is highly artificial, having little to do with the way trees shape themselves in nature. Once I came to this realization I began undoing any pads or clouds I could find, and this azalea proved a prime candidate for a makeover.

The next image shows the azalea on display in the bonsai garden for the first time, more than two decades after being received in donation. I put it out because I was a little more satisfied with the tree’s overall appearance and because it had its best-ever flower display. People love flowers:

I was a little more satisfied with the tree’s appearance, although not enough to leave the specimen out on display once the flowers faded. My acceptance of the Satsuki azalea’s look was predicated on my not looking at it too closely. If I spent any time examining the little tree and thinking about it, the outcome was always dissatisfaction with its credibility as a representation of a tree. As a point of fact, azaleas are not trees — they are shrubs. In bonsai, azaleas are cultivated almost entirely for their floral effect. When an azalea bonsai is in flower you put it out on display and collect compliments, and when the flowers fall off you take the bonsai off display. That means in a good year your azalea bonsai might be shown for a week or two, tops. That seems a waste to me because azaleas have so much more going for them as bonsai candidates than just their flowers. The floral display is so striking, however, that many people can’t seem to see past it.

Azaleas can be made into highly credible little trees and that’s my primary interest in bonsai. Make a good little tree and then if it happens to flower for a week or two, so much the better. Flowers, when thought of this way, are like icing on a cake. Icing is very nice, but the cake is what matters. Good cake is still good cake, whether it has icing or not.

What happened next was impulsive. That is to say, the action was impulsive, although years of consideration preceded it. One day in either 2020 or 2021 I was working out in the hoop house and came across the Satsuki azalea and decided I just couldn’t live with it the way it was any longer. I remember Mr. Yoshimura saying “If something bothers you, remove it!” With those words in mind, I took a saw and cut off large portions of the little tree. The result wasn’t pretty. It’s telling that I didn’t take a photograph to document this deconstruction, probably out of sheepishness or some pessimistic suspicion that I had ruined a perfectly good bonsai.

Once one decides to take such a severe step, the next step is easy — leave the tree alone for a while. Let it grow for a year, or two years, or three, or however long it takes for the tree to push out lots of new stuff. Here is what the Satsuki looked like last year, when I finally had the nerve to photograph it again:

I was ready to photograph the tree again, but not yet ready to work on it. That waited until just this past week. Here is how the azalea looked at the start of the session, shown in one-quarter turns beginning with what has always been thought of as the primary view:

The extended period of uninterrupted growth resulted in both vigor and vitality, as seen in this closeup of the azalea’s foliage:

The first step in the restyling process was to go through the canopy removing all growth that seemed least useful to my purpose. Nothing was shortened at this time and no major decisions were made other than identifying those shoots I felt confident were unnecessary and taking them out in their entirety. Here is the result:

When the azalea was radically pruned several years ago, the top had been substantially lowered. Azaleas, being shrubs, are basally-dominant. This means they spend more energy on peripheral growth than top growth, which is the opposite of how most tree species grow. Said another way, azaleas tend to be weaker in their top growth and this must be taken into consideration when pruning them. Here is what this specimen’s top looked like after the initial pruning pass in this first stage of the session:

Azaleas produce new shoots in whirls at the end of existing stems. The whirls can consist of three, four or five shoots in a cluster, and for bonsai purposes these points are typically reduced to just two shoots:

In the next phase of the operation, I concentrated specifically on the uppermost portion of the crown. This is the weakest area where I needed to be most restrained in my pruning. I went about structuring it in the way I thought looked best, always hedging my bet in favor of maintaining as much fullness as I could while achieving the desired reorganization. This was the result:

I focused on pruning the weakest part first in order to establish a baseline for branching density. Once that area was pruned I moved on to the other more robustly growing areas and pruned them so that their density was in harmony with the more weakly growing area. This was the result:

I’m afraid this specimen is stuck being a one-sided bonsai, due to its origin as a classically designed tree. So many important early design decisions were made on the assumption of this azalea being seen only from one perspective, that looking at it from any other creates a strange appearance. If I had wanted to completely overcome that limitation it would have required reducing the azalea even more drastically than I did, which would have set it back even further. As it is, I think it will take another year or two for this specimen to once again be showable.

Here is a closeup of the new, more naturalistic apex. It is plainly still in development, but the fundamental structure has been established. Note the area of die-back where the trunk was cut in order to lower the top. Such die-back is not surprising, although it is not necessarily desirable:

The lowest branch on the tree is the most developed at this point and exhibits an agreeable naturalistic structure:

The base of this azalea is excellent, perhaps the specimen’s best feature:

The old structure and the new, side by side for comparison:

Redesigning a Satsuki