This thesis reports on the factors relevant to the allocation of assimilates to oil palm bunch dry matter (DM) production, based on an extensive set of growth and yield records from experiments in Papua New Guinea and Malaysia.
Basically, assimilates from gross assimilation of the canopy are first used for maintenance of existing biomass (maintenance respiration). The remaining assimilates are converted into structural DM production. Carbohydrate requirements for components of DM production are derived from conversion factors based on the biochemical composition of the DM. Maintenance respiration is estimated as the difference between gross photosynthetic production, derived from the assimilation light response curve of individual leaves, and the amount of carbohydrate required for DM production. About 50% of total gross assimilation is lost in maintenance respiration; losses appear to be more related to growth rate than to the amount of existing biomass.
The allocation of assimilates to bunch DM production has a low priority in the carbon budget. Thus bunch production starts when certain minimum requirements of assimilates for maintenance respiration and vegetative growth have been met; above this critical value additional assimilates are virtually all utilized for economic yield. This implies that yield will benefit greatly from measures which increase photosynthetic production. Further, that yield will respond strongly to a reduction in internal competition, in terms of assimilates, from vegetative growth and maintenance respiration.
This thesis shows that the efficiency of converting intercepted radiation into carbohydrates for dry matter production decreased with the expansion of the crown leaves, which might be linked to the resulting increase in maintenance respiration losses. These losses were apparently not compensated by an increase in photosynthetic production. The efficiency increased again when crown expansion was complete, which appears to be due to improved light distribution consequent on an observed increment in light penetration. These effects of changing levels of interception and distribution of sunlight on efficiency were more pronounced as the planting density increased, and resulted in marked differences in yield trends with age between densities. The end result was a decrease in optimum density for current yield until 12 to 13 years from planting, followed by a strong increase. It is concluded that optimal density for cumulative yield might be increased by measures which optimize the balance between light interception by and distribution of light over the foliage. The response of improved light distribution would be enhanced by selection for net assimilation rate of the leaves.
Palms differ considerably in rate of leaf expansion. As expected, this is reflected in total DM production per palm, which to a great extent benefits bunch yield.
DM production per palm was also increased through reducing the incidence of deficiency of magnesium, an essential component of the chlorophyll molecule; again, the benefit is mainly obtained in oil yield. Selection for high magnesium status is feasible and appears to be an efficient way to increase oil yield.
increasing bunch yield, at a certain level of DM production, implies reducing requirements for vegetative growth (increase of the Bunch index). ideally, a reduction in vegetative DM production should be restricted to nonphotosynthesizing tissue. This aim can be achieved by selection for high leaf area ratio (LAR), defined for this crop as the ratio of new leaf area produced to new dry matter used for vegetative growth. Another method might be to reduce the period of crown expansion subsequent to canopy closure; from this stage onwards, no further increase in gross assimilation occurs but additional crown expansion continues to increase requirements for vegetative growth and maintenance respiration. Therefore, speed of crown expansion should be combined with a reduction in crown expansion time. These two objectives can be achieved by selection for so-called rapid expansion (RER) ideotypes, i.e. a high ratio of leaf area at maximum expansion rate and a relatively low final size.
The observation that Palms selected on the basis of increased light interception and reduced vegetative requirements do indeed have higher BI, indicates that selection for ideotypes (defined as biological models which are expected to perform or behave in a predictable manner within a defined environment), is feasible. This finding justifies exploring other methods which, through manipulation of the carbon budget, would increase assimilates available for bunch dry matter production. Selection for increased photosynthetic production at light saturation, and also for reduction of leaf maintenance requirements, may be rewarding. Screening of progenies in the nursery on the basis of increased photosynthetic production of the leaf surface and for reduced maintenance respiration, as well as for leaf-Mg level and LAR, may be possible.
To permit large scale testing of the response of yield to ideotype selection , a method to improve the conventional time- consuming method of growth recording is proposed. The main finding is that rate of leaf production, which is the key for an instant method of growth recording, can be obtained by counting the leaf bases on the trunk. The proposed, so-called one-shot" method of growth recording, in particular, resulted in an efficient selection of progenies on the basis of BI.
In addition to vegetative growth and maintenance respiration, the development of bunches competes with inflorescence primordia for carbohydrates. Fruiting activity also, to a certain extent, competes with vegetative growth. Thus if the potential sink size for fruit bunch production is not fully uti1ized, palms tend to produce excessive vegetative growth. A direct implication is that adequate pollination is essential, as could be demonstrated by comparing vegetative growth of similar planting material under poor and adequate assisted pollination.
The supply of carbohydrate affects all the components which determine the number of bunches (sex ratio, abortion, and leaf production) and those which determine their mean weight. The latter can be divided into the frame (stalk and empty. spikelets) and the fruit (the product of spikelet number, ,flower number per spikelet, fruit set and the mean weight of individual fruits). The number of flowers per spikelet and sex are both determined just prior to spikelet initiation. For floral abortion the critical stage is at the onset of rapid expansion of the inflorescences, about 10 leaves prior to anthesis. It was also at this stage that the first response of frame weight to carbohydrate supply was observed. The weight of the frame is further affected at two earlier developmental stages: first, when the central axis ceases to be meristematic and, second, just prior to spikelet initiation. The latter stage, which is about two years before harvest, is thus the most important stage in inflorescence development in that here key components of bunch yield are affected by the supply of carbohydrate.
An analysis of components of oil yield at different spacings (varying levels of carbohydrate supply) showed that components which are determined at an advanced stage in inflorescence development (floral abortion and stalk weight) are more sensitive to changes in carbohydrate supply than those determined at an earlier developmental stage (sex ratio and number of flowers per spikelet). These finding are supported by an analysis of response of yield components to different levels of fruit bunch removal (disbudding). Moreover, in respect of abortion, it appears that initial response to shortage of carbohydrate mainly decreases the number of female if this shortage is prolonged both sexes abort.
Insect pollination, which replaced the method of assisted pollination in the course of this study, appears to have increased optimal planting density by at least 5 palms per ha. This finding is derived from a measured increase in extraction of oil and kernels in response to planting density, and from a decrease in vegetative growth due to, as mentioned earlier, an increase in sink strength of the fruits.