In the three years since the Yamamoto Saikei had its last major pruning, the trees had grown a lot. There were major potential problems as all junipers are being infested with a lichen growing on the rocks, trunks and branches.  The lichen may or may not be harmful.  But it is unsightly and efforts to water-blast to remove it may have done more serious damage as it washed out media between the rocks. The trees were shaking and growing erratically.  This article addresses several major issues.


Part II - Addressing a critical challenge!

                  The Hawaii State Bonsai Repository collection includes several junipers.  One significant one is a saikei tray landscape initially created by Tom Yamamoto,  the heir to Toshio Kawamoto's Tokyo Bonsai-Saikei School.  Tom is a native of Hawaii who was part of the US Army post-World War II occupational forces.  He retired in Japan and assisted Toshio as an English language instructor as many in various foreign consulates wanted to learn facets of Japanese culture. I first met Tom and Toshio Kawamoto when they gave a presentation in the early 1970's in Honolulu.

                 In 1993, Tom visited the Big Island and did a major demonstration at the Kona Fuku-Bonsai Center.  Enroute to Kona, we stopped along the road and he picked up a boxful of ordinary Kona clinker-type lava, stressing that saikei is popular because it uses ordinary rocks and young plants.  With good technique, an arrangement is attractive from the beginning.  The saikei arrangement can be enjoyed while the individual trees are growing larger to one day be individual bonsai.  Or you can take the arrangement apart and use the components to make other arrangements.

                  While generally saikei was not intended to be a long-term tray landscape,  Tom agreed that it could be.  The original arrangement was a group of seashore trees with a low rocky mound.  We've kept the same bonsai oval container and the same horizontal proportions.  But we doubled and tripled the vertical dimensions and the arrangement has evolved into a dramatic presentation.  We substituted taller vertical rocks of the same type but retained the original concept, rock, and tree positions. 


                  THE JULY 2014 CHALLENGE.   The water-blasting efforts to remove the lichen has washed away much of the media between the rocks and the trees were loose.  Limited resources prevented us from pruning more frequently and the top of the highest tree had overgrown and major restoration was necessary. The large first photo and the three above shows the arrangement from four sides.  The two side views show that the top tree especially has wildly grown out and is shading out lower trees.


                  Dwarf Japanese Garden Juniper (Juniperous squamata 'Prostrata') is a creeping type juniper which is superior to the more common Procumbens Juniper or the dwarf version Juniperous procumbens 'Nana."  In California and the continental U.S. these are often allowed to grow long along the ground, staked up with the branches trained horizontally following the typical Japanese single apex-tier branched "pine tree" styling.  Sometimes the trunks are wired into "S-curves,"  the apex is kept thinned and pointed, and individual branches trained into flatten tiers.

                 In Hawaii,  we may wire young trunks upwards, then bend the main apical growth downwards to form a branch.  This type of junipers are "top growth dominant" and the portion on the top of the tree will throw out a major new growth leader.  We encourage it to grow upwards, then bend the replacement apical growth downwards to form a second major branch.  With good growth,  a crown starts developing and another new major growth leader emerges from the upper section to be encouraged to grow upwards, but eventually bent down to form a third major branch.  This juniper has two types of foliage.  The initial "juvenile" foliage are more slender and needle-like.  When trees are growing robustly,  a broader "mature" type of growth appears.  Generally much of the mature foliage should be removed to allow all parts of the tree to grow evenly.  But in the very hectic past two years while I focused on Dwarf Schefflera,  the junipers went out of control and restoration was necessary.

                 In this type of styling, developing three vertical leaders to evolve into the three major branches form the basic tree structure and a rounded crown develops.  New vertical growth leaders continue to develop and the rounded crown becomes wider if allowed.  Because growth is so top dominant,  if the top is allowed to overgrow, lower branches get shaded and begin to thin out.  That's what is happening here. Lichen is especially developing on the sparse thinned lower branches.


                 It took several hours to clean out this tree --- to remove moss and lichen that had grown on the rocks, trunks and branches.  I used a small brass wire brush that does not seem to harm the bark of the trees.  Once cleaned out,  it's easy to see that the lowest branches are dying out and it is necessary to prune to "balance" the growth. 


                Using the condition of the weakest lowest branches as the standard,  it is necessary to "balance" the tree to dramatically thin out the top growth to the level of the weakest lowest branches.  The arrangement was taken out of the oval pot and the left tree separated from the main arrangement.  All trees were loose and shaking and without major restoration work, it is likely that the trees would die.  As can be seen, some of the rocks on the lower right side were also loose and separating.


                The photos are before and after about 75% of the balancing was completed.  Junipers tend to develop heavy crowns that if allowed to grow, will overwhelm and weaken the growth of the lower portion of the arrangement.  When the arrangement had a major pruning a few years ago,  the arrangement was not properly "balanced,"  the main top crown not sufficiently pruned back,  and this created the critical challenge that is now being addressed.  

                 In general,  a curator of older bonsai MUST especially prune back hard the sections that are growing vigorously.  This is a very time consuming task that was not done and the problem is now magnified.  If not done,  growth shifts to favor rapidly growing sections and the entire concept of the arrangement changes! In addition, even the rocks were loose. The trees were shaking, growing poorly and in danger. With the exception of the top tree which was planted "root-over-rock,"  all trees were planted between rocks.  The top tree had the best growth position as well as the best roots (because it was planted in the superior root-over-rock manner) and that gave it comparative robust growth while the other trees declined.   Emergency work was required and at the end of a long day,  the cleaned and thinned tree was water-blasted to remove any remaining moss and lichen and allowed to dry out over night.


                  The two photos above were taken in mid-morning after a lot of major corrective work was completed.  Using fast curing cement,  new taller rocks were joined together to give stronger support for the trees on the right side. A very heavy cinching wire pulled the major arrangement rocks tightly together both at the bottom (where it would be hidden by the potting media) as well at mid-level with the wire largely hidden. Heavy wire anchored to the bottom cinching wire went up and over the base of the trees and cinched tightly to the bottom circling wire on the other side. 


                 Each time a wire when up and over to anchor down a tree, it pulled the other wires tighter.  Other wires pulled the vertical wires together. Smaller vertical wires were installed and horizontal wires pulled sideways to tighten further.  This formed a wire grid that held sphagnum moss. Media dibbled behind to fill all openings pushed out against the moss and grid.  By early afternoon much of the media had been dibbled in. 


                 It was slow tedious work to dibble coarse media behind the a wire network holding the sphagnum moss and dibbled materials in place.  Using this technique,  we can form firmly positioned vertical walls of media.  In the past we used the Fuku-Bonsai Cornstarch Keto-Tsuchi to hold the surface in place.  But it was not sufficiently strong to withstand the water-blasting efforts to remove the lichen.  Necessity is the mother of invention and I invented what I hope is a stronger more durable but porous "Fuku-Bonsai Concrete Keto-Tsuchi." 

                The new concrete keto-tsuchi was used without the luxury of testing it first.  The basic concept is the standard hypertufa formula of equal parts of cement, fine lightweight aggregate (with fine organic matter mixed in), and sphagnum moss.  I estimate it was 1/2 part screened Quickcrete (which includes fine sand),  1/2 part fine volcanic pumice aggregate,  and about 2 parts organic matter.  The components are in the small containers in the photo above and the first mix is in the orange bowl.  It did not seem to have enough holding power so the amount of Quickcrete was increased.


                 I made two samples of the original mix.  One with a piece of aluminum foil folded in half with some concrete keto-tsuchi pressed between.  A second sample pressed some into the bottom of a saucer to see how well it replicated contours.  More Quickcrete was added to the original mix and two additional similar samples were made.  In addition,  a round clear plastic wrap contained some sand which as encased in the concrete keto-tsuchi that was similar to a sample that I had made as my first hypertufa trial.  Some concrete keto-tsuchi was pressed onto a moss-covered stone.  A plant in a 7" 1:10 Project shallow saucer was surfaced with concrete keto-tsuchi, the surfaced with fine pumice and organic topping.  A clear kitchen plastic wrap lined a cover and left-over concreted keto-tsuchi was pressed into it in about a 1/2" thick piece. 

                These test pieces will  help determine the porosity of the concrete keto-tsuchi to develop working procedures for using it as well as insights into how best to alter the formula for better results.  In addition to the repository trees being restored,  two additional juniper bonsai await the results of the initial trials as immediate emergence remedial work is also necessary with the most challenging one still to be addressed.


                 The two above photos show before and after completing the installation of granular media followed by a layer of concrete keto tsuchi followed by a surfacing of fine pumice dust with organic matter.  I evaluated the results the next morning.   When the concrete keto-tsuchi was installed about 1/4" thick and compressed to about 1/8" thick,  it was still slightly flexible but porous.  When installed thicker and compressed, it dried hard and was far less porous. 

                 I'm guessing that like the hypertufa, that the mixture will harden over time, but will still be porous due to the generous amount of sphagnum moss in it.  It was decided to create some openings filled with sphagnum moss at the top areas to allow water to enter  and travel down to moisten the area behind that outer concrete keto-tsuchi.  I was also decided that after photos were taken, that a temporary foil apron would protect the keto-tsuchi for the two weeks or so needed to cure the cement.  The arrangement was trimmed to better define each tree,  the largest top tree was cut back a little harder, and the photos were taken. 



                    It took three full days to complete restoration of the Yamamoto Saikei as extensive work was necessary to prune back and balance the growth of the trees and to repair and tighten up the rock work that had been loosened by water-blasting to try to rid the lichen that had gotten established.  If you compare the above four photos to the first four photos in this article,  you'll see that while the lowest branch of the small low tree had limited trimming,  a lot of the top of that small tree was pruned. 

                    But in comparison, a HUGE amount of the crown of the top tree was removed to try to balance the weaker growth of the bottom branches.  This tree is an example of what not to do.  Under optimum conditions,  it is best if there are sufficient resources to properly maintain the trees throughout the year. We try to create the best possible growing situation, but with limited resources,  we've had to learn to improvise when necessary.  I believe the concrete keto-tsuchi will provide suitable support and the trees will recover fully.  Once the surface hardens,  the danger will have passed.  It was decided to use an aluminum foil partial protection for two to three weeks as shown below.



                     This first pair of photos show the tallest vertical rock on which the largest top tree was planted root-over-rock with about half of the root system on each side of the rock.  This type of rock planting is a superior and preferred technique.  When the revised saikei was done,  we could only find one very thin and tall rock and had not yet developed the skill to create such rocks by joining one atop another.  In complex landscapes, we try to find similar rocks all from the same area to create a harmonious scene.  Generally the main lines and planes of the various rocks are parallel as this produces a comfortable effect. 


                     Two additional photos of the left back and the right back. The aluminum foil only protects the exposed media between the fitted rocks. The rocks were fairly well selected and much of the landscape is made up of vertical rocks with relatively narrow sections of exposed media. The above right photo shows the rocks that were cemented together to provide solid support for the right side of the saikei landscape.  Generally in saikei,  only keto-tsuchi is used to hold the rocks into position. With smaller rocks and younger plants,  the challenge is a lot simpler. 


                      This is also a Dwarf Japanese Garden Juniper that is a bit older --- about 30 years in training on a 25" tall rock that has a cement base to keep it upright.  This is a two trunked tree planted at about the mid-point with the top of the rock showing above the crown of the upper tree.  The lower tree grows away from the rock and this arrangement is amongst the more admired trees at the center.  A large amount of potting media had also been water-blasted out  --- but the tree has suffered less damage and was more easily restored with cornstarch keto-tsuchi.


                          The main root of the top section is still against the rock, but a major root that was once part of a built-out media area shows daylight as the entire media area is gone due to the water-blasting. 


                      At one time the concave area between the rock and the large root was filled with media and while some smaller roots have died, the tree is well established in the pot and restoration was to build out to create a larger area for new roots to develop .  The center photo shows that a 1/4" thick cornstarch keto-tsuchi wall about 2"-3" high is outlining the area on the surface of the media in the pot to be built-up.  Additional keto-tsuki with some sphagnum moss outlines other areas to be built out.   An inch of coarse bottom slopes up to the rock and firmed in place with a back and forth motion.  Then another inch of coarse medium with Nutrient Granules was firmed and sloped with that back and forth motion.  This method allows building up slopes with only decreasing size potting granules and if covered with fine granules, then a layer of fine dust with organic material and protected, there will be excellent drainage and vigorous root hairs will quickly colonize and hold the slope in place. 

                      The keto-tsuchi edges are then folded over to partially protect the edges and more keto-tsuchi is used to form the base of the column that will hold media that will fill that concave rock cavity. With the column formed, it is quickly filled with an open top funnel and dibble. For our collection,  we like to find rocks that have a concave area on the lower half of the rock as it can hold a lot of media to keep the trees very healthy without a visually heavy appearance.  


                        As the column is filled, the sides are built up.  To close off the top of the front,  I found a matching rock that fitted nicely and keto-tsuchi was used to hold it in place.  There was still access from the back to continue filling.  It ended with a ball of sphagnum moss that will allow water to quickly enter to water the media in the column.  Fine dust with organic matter was rubbed onto the keto-tsuchi to take away the stickiness and allow firmly pressing to compact the media within the column.  Take your time and compress it firmly as the column will collapse when softened with water if the media is loose.  To finish off and to protect the surface of the keto-tsuchi,  Java Moss was planted.  Generally Java Moss is effective when the area will be shaded  and finer Kyoto Moss favored in sunny areas.  Sometimes both types are planted and the one that is best suited for the situation takes over. 

                       If you compare the larger before and after photos,  it may not seem to be a big change.  But this will greatly benefit the health of this bonsai.  The trees now have access to an additional two to three quarts of potting media!  The additional bulk of the area built out makes the bonsai more attractive and this tree will likely continue to be amongst the favorites of the visitors to the center. 


                      I came up with this "keto-tsuchi variation" based upon my experience with "cornstarch keto-tsuchi" and initial "hypertufa" trials.  Both of these are similar in that the base formulas are 1/3 each of an aggregate, organic matter, and a binder.  Each of these three basic components can be in many different forms.

                      For the cornstarch keto-tsuchi, we were trying to come up with a substitute for the Japanese keto-tsuchi which is a really fine muck used in their rock plantings or forest plantings on large artificial or natural flat stone slabs. It is sticky and dries hard and can be used like clay to build a wall on the slab which is then planted like a pot with coarse bottom, medium and finer aggregate (with some organic matter).  The keto-tsuchi can be shaped and sheet moss adheres to it easily. 

                     I was told that "the Japanese keto-tsuchi was dug three feet below a rice patty that had been in cultivation for over 100 years."  That's probably an interesting cultural fable and I'm sure each master has his own secret source.  But such conditions would produce a very fine organic silt as the rice fields are burned annually.  Our cornstarch keto-tsuchi is quickly produced with readily available components. But after a week or two, it smells a bit as it decomposes.  Concrete will not likely stink, but would result in a rigid material like hypertufa which really is a lightweight cement used to make porous planters and garden sculptures. 

                     In lieu of Portland cement, I used QuickCrete with the quartz-like aggregate screened out.  It is likely about 50% Portland cement with about 50% very fine silica-like sand.  In my first hypertufa trials,  I used about 1/3 each colored QuickCrete,  fine 1/8" minus black volcanic pumice ejecta, and fine 1/8" minus ground coconut husk (coco-peat).  It can be molded and workable for about a week and after drying hard (and porous), it   can be drilled and shaped with grinder heads to create attractive artificial rocks. 

*** Return to the August 2014 issue of Journal of Tropical & True Indoor Bonsai
*** Go to the Mid-Pacific Bonsai Foundation website
*** Go to the Fuku-Bonsai website
Mid-Pacific Bonsai Foundation and Fuku-Bonsai, 2014