Growth traits should be given key importance in the interpretation of trial results, as pulpwood is the major product from P. kesiya.
The present results confirm the observation of Morris (1984) from the first series of P. kesiya trials that the ranking of provenances depends on the observed growth trait. In these trials, this is mainly because of differences in survival.
If height growth is considered, the results of the two trials show a quite consistent picture. The highest ranking sources are the John Meikle CSO (Zimbabwe), the Zambia CSO, the P. patula control, the Philippine sources Mt. Province and Benquet, and the Vietnamese sources Thac Prenn and Lang Hanh.
A rather different picture is displayed if total volume production per ha is considered. This trait combines the height, diameter and survival into one figure. The low survival percentage of some of the sources influences the total volume per hectare results. The best overall growth performers based on total volume production per hectare are the Mt. Province source from the Philippines. Also the Benquet source from Philippines and the Vietnamese sources (esp. Thac Prenn and Lang Hanh) are doing well.
The Zambia CSO source, which was top-ranking in height growth, is not among the top ranking sources in regard to volume production. This is because of a low survival percentage in both trials. The low survival raises questions about the growth potential of this source, and the issue should be investigated further (see below).
The sources from the John Meikle CSO express a rather confusing picture. They are – as mentioned above – ranked high in height growth, but the performance differs in volume production, again because of differences in survival. In the Block G trial, the 1779 seed lot is ranked high, and 1776 low. In the trial in Block J the picture is opposite with 1776 at the top and 1779 low. In any case, the seedlots from John Meikle CSO should be interpreted with care as they are single mother tree progenies, and hence not a fair representation of the John Meikle CSO.
The results of the statistical analysis show that the best P. kesiya sources are equal to, or better than the controls. Best growth performer among the controls is the P. patula source. It is ranked at the top in Block G, but intermediate in Block J, mainly because of a relatively low survival rate in this trial.
The two other controls, P. taeda and P. elliottii, express above average growth in Block G, but below average growth in Block J. Here especially P. taeda is slow growing. These results confirm the unsuitability of P. taeda and P. elliottii at higher altitude sites.
The Chinese sources (Jinghong, Jingdung and Lancang) are intermediate in regard to growth together with the Thai sources, with Thai sources having slightly slower growth than the Chinese. Doi Inthanon is the best of the Thai sources, especially in Block G.
Poorest growth performers are in both trials the P. yunnanensis sources (Baoshan, Shangsi and Ceheng), and the Burmese sources (Zok Hua and Aungban). Volume production of the P. yunnanensis sources is less than one third of the best performing P. kesiya sources.
Production potential of the best performing P. kesiya sources, based on growth in the first 8 years, is 22-23 m3/ha/year.
Analysed adaptive traits include survival percentage, foxtail frequency and flowering.
The low survival rate expressed in the trials of especially the Zambia CSO source, but also the John Meikle CSO source, raises concerns about the suitability of these sources. The reasons for the poor survival should be investigated further. It would be of particular importance to establish if the low survival of these sources – which in not evident from other trials in the international series – can be confirmed in other trials, and if so, if the low survival is caused by dead of seedlings in the early stage of establishment or later on in the life of the plantation.
Apart from this, there are no clear patterns of provenance variation in regard to survival. The results of the two trials present a rather confusing picture where sources with a good survival in one trial have a poorer survival in the other, and vice versa.
The level of foxtails is relatively low in the two trials, in Block J so low that no analysis is possible. The results from Block G indicate that the P. kesiya/P. yunnanensis sources set more foxtails than the controls, and that P. yunnanensis sources have higher frequencies of foxtails than the P. kesiya ones.
Flowering as an indicator of adaptation should be interpreted with care because of the early stage of trial and the different reproductive cycles of the species (i.e. species may flower and set cones earlier or later). There are significant differences among seedlots in both trials, but no clear patterns of provenance variation.
Analyzed quality parameters are stemform, relative wood density (pilodyn) and branch diameter.
The P. kesiya sources in general have poorer stemform than the controls, but on the other hand have higher wood densities. An exception is the Zambia CSO which compares favourable to the controls in regard to stemform, especially in Block J. This result illustrates that selection and tree improvement can significantly improve the stemform of P. kesiya.
The John Meikle CSO is not among the top-ranking sources in stemform. This is somewhat surprising, as a similar effect of selection and tree improvement as observed for the Zambia CSO would be expected. The reason may be that the two tested John Meikle sources are not fair representations of this source. The issue needs further investigation.
In general, the Vietnamese sources have better stemform than the Philippine and Thai source. P. yunnanensis and Burmese sources have poor stemform.
The P. kesiya sources have higher relative wood densities than the controls. The Thai sources, esp. Doi Inthanon, have high densities, but also the Vietnamese sources are doing well, and better than the Philippine sources.
In regard to branching, especially P. elliottii has thin branches. Among the P. kesiya sources, John Meikle CSO and Mt. Province are best. The Philippine sources in general have thinner branches than the Vietnamese and Thai sources.