FROM time to time, someone still asks me where oil palm breeding should go next.
I should clarify that I was trained as a plant breeder.
Human beings had already been breeding quite successfully for thousands of years without needing people like me to earn a degree for it.
Plant breeding taught us to make crosses, observe variation, select promising individuals and practise that hardest scientific virtue: patience. With oil palm, you cross today, plant tomorrow and wait years to discover whether you were clever or merely optimistic.
I eventually wandered into plantation management and industry work, but the breeder in me never quite disappeared.
These days I return to the old questions with different eyes.
I no longer have a breeding programme to defend, which gives me the luxury to discern rather than prescribe.
That suits my aspiration to inspire not by claiming all the answers, but by asking a few useful questions.
One keeps returning: how high can the oil palm really go?
Not how high a model says it might go, nor how much one extraordinary palm produced in one extraordinary year.
But, how much could a real hectare (ha) produce over time if good genetics, land, agronomy, harvesting and management were brought together to give the crop its best chance?
That question was explored at a global crop-yield-potential workshop in Wuhan, China, last year.
Indonesian agronomist Hendra Sugianto presented the oil-palm case with Malaysian plant breeder and agronomist Christopher Donough serving as adviser, among researchers seeking to understand how crops might be pushed closer to their biological limits.
Much of the discussion concerned annual crops such as rice, maize and wheat. Oil palm, producing over decades rather than months, brought a different personality to the table.
Rice and maize researchers can learn from one crop and try again relatively quickly.
Oil palm looks upon such haste with amusement: it takes years to reach serious production and may remain in the field for around 25 years.
A poor planting decision can accompany a planter for much of his career.
The number that tempts us
Scientists understand much about how oil palm captures sunlight and distributes that biological income among roots, trunk, fronds, flowers, fruits and oil.
From this, estimates of potential yield have been developed.
Additionally, exceptional individual palms have produced around 500kg of fresh fruit bunches in a year during their prime.
Multiply that figure across a normal planting density, and the calculator produces a number exceeding 70 tonnes per ha.
This is perhaps when some plant breeders should occasionally be separated from their calculators.
The qualification is that the same remarkable palm does not necessarily perform like an Olympic champion year after year.
At ha scale, something around 50 to 60 tonnes of fresh fruit bunches per ha in peak years has been suggested as a potential worth investigating.
That is dramatically above ordinary commercial performance.
More importantly, current estimates rest substantially on crop physiology; nobody has really demonstrated what might happen if we deliberately pushed the envelope in a real field situation.
We have estimated the height of the ceiling.
Have we ever properly built the ladder and tried to touch it?
A theoretical yield is like saying a car can travel at 250km an hour because its engine and aerodynamics allow it.
That says little about travelling through potholes, traffic lights and Monday morning congestion.
Our palms live in similar agricultural traffic: weather, variable soils, pests, labour shortages and imperfect harvesting.
To discover the crop’s upper limit, we have to remove as many avoidable obstacles as possible.
The Wuhan concept therefore envisages a long-term experiment on good, reasonably uniform land, using high-performing planting material and plots large enough to generate meaningful field-scale results.
Such work would begin at establishment and continue through peak production, potentially for around 15 years.
That may sound long, but oil palm itself is a perennial crop and a long-haul investment.
If the industry believes in its long-term future, a 15-year experiment should not be excessive; it is research conducted on the timescale of the crop.
In oil palm, patience is not an inconvenience to science.
It is part of the science.
Then comes a phrase guaranteed to make every planter smile: “perfect management”.
It exists somewhere alongside perfect children, perfect golf swings and maintenance-free machinery.
Scientifically, however, it simply means reducing controllable limitations as far as possible.
We would not test an elite athlete on poor nutrition and irregular training and then call the result his limit. Why not give the oil palm the same chance?
Measure enough to be surprised
A serious yield-potential experiment should not merely grow impressive palms and weigh bunches.
It should sit alongside breeding trials because variation itself is information. Somewhere among differences in canopy, roots or environmental response may sit a useful characteristic nobody originally went looking for.
That requires curious people and good records.
The Wuhan thinking, therefore, goes beyond bunch yield to canopy development, roots, plant physiology, soil and plant nutrients, and data that can improve crop models.
If bunches are measured, why stop there?
Oil, kernels and other outputs matter too if we want to understand what the whole ha produces.
This reinforces a principle I increasingly appreciate: what we fail to notice today may be exactly what we wish we had measured tomorrow.
Research cannot measure everything, but it should observe widely enough for the unexpected to reveal itself.
Sometimes science advances when someone simply asks, “Eh, why is this palm doing that?”
Take planting density.
Palms typically march across estates in triangular formation, around 136 to 148 palms per ha, roughly nine metres apart.
Once those points are fixed, there is no easy reshuffling of the parade.
A decision made with pegs and measuring tape can influence tonnes of crops for 20 years or more.
Planting-density research has been done before, but should tomorrow’s planting material automatically inherit yesterday’s spacing?
Materials may differ in canopy size and growth habit; labour is tighter, mechanisation matters more and every ha is under greater scrutiny. Yesterday’s spacing may still be right, but it deserves testing, not sacred status.
The Wuhan proposal therefore includes systematic density experiments comparing different spacings.
Imagine a dining table: every palm needs sunlight, water, nutrients and space.
Put too few around it, and resources may be underused; put too many, and everyone begins elbowing one another away from dinner.
The challenge is to find the seating arrangement that feeds the ha best.
The thousand-ha R&D “playground”
The Wuhan thinking eventually moves beyond individual experiments to a more provocative proposition: what if the industry had an estate specifically for research and development (R&D) complete with its own mill?
Not several plots tucked inside an otherwise commercial estate, but perhaps 1,000 to 2,000 ha where breeding, agronomy, harvesting, mechanisation and processing could be studied together at meaningful scale.
This caught my attention because the idea has some history.
Before leaving IJM Plantations, Christopher had imagined Sijas estate becoming roughly a 1,000-ha R&D “playground”, perhaps large enough to support a small conventional mill of around three tonnes per hour.
IJM then had roughly 26,000 ha.
The dream never materialised, but perhaps the question deserves another hearing.
Small research plots remain indispensable, but oil palm does not finish its journey when someone weighs a bunch.
Fruit must be harvested, transported and processed.
A change that looks excellent in a trial may create another problem downstream.
The field and mill are parts of one system, and an R&D estate could study that system rather than only its pieces.
Yet the most important ingredient may not be ha.
It may be people who are allowed to think differently.
Ordinary plantation executives are judged on crop, field standards, manpower, costs and budgets.
Ask someone whose appraisal depends on this year’s numbers to try an idea whose answer may arrive six years from now, and enthusiasm may understandably be limited.
Commercial estates rightly reward consistency.
Research needs curiosity.
The idea, therefore, needs progressively minded field executives with room to test better ways of working alongside researchers.
An R&D estate should be neither an ordinary estate where scientists arrive, collect data and disappear nor a research station insulated from moving crop to a mill.
The two cultures have to meet.
The researcher asks, “What if?” The planter asks, “Will it work?”
Progress begins when both are standing beside the same palm.
Such a place must also allow experiments to fail intelligently.
A spacing may disappoint, a mechanisation idea may prove impractical or promising material may behave differently at scale.
Finding that out across 30 ha can be far cheaper than discovering the same truth across several thousand.
In research, a small failure can be a very good investment.
That is why I like the word playground. It does not imply frivolity.
It describes a protected place for disciplined curiosity, somewhere people can try, measure, adjust and try again.
Then comes the awkward question: who pays?
A consortium of stakeholders could share the cost and learning, while the government might reinvest a modest portion of industry taxes or levies into long-term R&D.
If palm oil can generate more than RM100bil a year for Malaysia, setting aside a modest space to test better ways of working is hardly an extravagance.
No single company may wish to carry a 15-year experiment whose benefits could ultimately spill across the whole sector, which is precisely why collective funding makes sense.
After all, oil palm is a perennial crop, and this is a long-haul business.
It seems only logical that its R&D horizon should be willing to think long too.
Such a playground needs the right leadership and a capable, curious team, people able to bridge science and operations, challenge assumptions, measure properly and admit when an idea does not work.
Get the leadership right, and the playground becomes more than a research cost; it becomes an investment in making sure tomorrow’s billions are earned a little more intelligently.
Many roads to Rome
Agriculture rarely gives one universal answer.
One density may suit one material but not another; a harvesting system may work with young palms and struggle later.
Research need not discover the road; it should map the roads.
Some may become highways, some useful shortcuts, and some may end embarrassingly in a swamp.
But unless somebody explores, we keep travelling the familiar route because an earlier generation built it.
That may be safe.
It is not necessarily progress.
When people ask where oil-palm breeding should go next, I am less inclined to offer another shopping list of desirable traits.
Shorter palms, more oil, higher yields, easier harvesting, climate resilience and machine-friendly architecture all matter.
But perhaps the deeper question is no longer only what should we breed?
It may also be: how do we create the conditions for breeders, agronomists, planters, engineers and millers to discover the next answers together? Malaysia already has generations of oil-palm knowledge, experienced people and increasingly sophisticated technology.
What may still be worth creating is a place where these capabilities meet at a meaningful scale, with enough scientific discipline to learn and operational reality to remain useful.
A thousand ha sounds extravagant as research land.
But if lessons from it help many thousands produce more crop, lose less fruit or recover more oil, the arithmetic changes quickly.
Perhaps this is where my role now sits.
I can still aspire to inspire by asking questions, connecting a few dots and prompting the industry to reconsider what it does, assumes and might still dare to try.
We have spent decades estimating the oil palm’s ceiling.
Perhaps the next step is to give curious people good land, good planting material, enough scale, a mill, plenty to measure and permission to experiment.
Then put the breeder, agronomist, planter, engineer and miller in the same field, let them dirty their boots together, and ask one deceptively simple question: How high can you really go?
Joseph Tek Choon Yee has over 30 years of experience in the plantation industry, with a strong background in oil palm research and development, C-suite leadership and industry advocacy. The views expressed here are the writer’s own.
