Successful tree improvement programs rely on the ability to screen, select, and breed for specific traits. The speed at which a program achieves its goals is often increased with the help of asexual, vegetative propagation, usually in the form of rooting or grafting. In the case of Acacia koa (koa), the speed and success of koa breeding efforts to select for koa wilt (Fusarium oxysporum) resistance, frost tolerance, tree form, and wood qualities can be improved with the development of reliable techniques for both experts and novices to graft koa. Nevertheless, past efforts to develop these techniques have either proven unsuccessful or suited to only the smallest root and scion stocks. Developing advanced techniques of propagation of koa (Acacia koa Gray) is one of the prerequisites for successfully moving the koa tree improvement program forward.
In summer of 2018, a grafting study collaboration with Kyle Rose, Owen Burney, and Simon Landhäusser was conducted at the New Mexico State University’s JTH Forestry Research Center to examine how different grafting methods (whip-tongue, mechanical V, and non-grafted control) and contrasting seed sources (Umikoa, wet; Kona Hema, dry) influence grafting union success, plant growth, and plant hydraulics. Grafting methods were based on those developed in a pilot study by Earnshaw and McKenna (2018).
Overall, seed source did not play a significant role on the survival and growth of grafted seedling which was evident from the lack of a significant interaction between seed source and grafting method. This lack of an interaction suggests that seed source did not have an influence on grafting success regardless of the method use. When examining the grafting methods, survival of grafted koa seedlings averaged approximately 50% across all methods compared to 95% survival for the non-grafted control seedlings. Within the two grafting methods, the whip-tongue method had slightly higher survival rates (60% survival) compared to the mechanical method (43% survival), although not significantly different. Three months after grafting, those seedlings that survived were measured for biomass (new leaf, new stem, old stem, root, and total) and non-structural carbohydrate concentrations (sugar and starch). No significant differences were detected between any of the treatments for these response variables. This lack of response may be a function of the high variability observed for individual seedling responses. Regardless of the variability, this study is proving that our methods can result in successfully grafted koa.
One interesting result from this study was seen with the difference in the hydraulically active xylem vessels between pre- and post-grafted koa seedlings (Figure 1). In the pre-grafted stained material, water is being transported throughout the entire cross-section of the xylem from the pith to the cambium layer. However, in all post-grafted seedlings, including the control, water is only be transported in the section of the xylem that was newly constructed after grafting. Our grafting methods remove the top of the scion material and all leaves from both the scion and root stock. This may be eliminating the water transport mechanism in the existing xylem at the time of grafting. Thus, it is hypothesized that all growth and newly formed hydraulically active xylem are relying on stored carbohydrates. Research on grafting koa will continue in an effort to improve both survival and growth. One area of research interest will be in how we can maintain usage of the existing xylem structure to transport water and nutrients at the time of grafting.