Merlot has had poor press of late, and the producers whose livelihood depends on its reputation have every reason to be concerned. The judges' feedback session at this year's Old Mutual Trophy Wine Show saw a repeat performance of what happened at last year's event, with all of the panellists repeating verbatim the litany of faults which emerged again at this year's tasting.

At one level, there's a lurking injustice in all this. Yes, the average South African merlot is thin, weedy and often a little sweet-sour. However, there are also some pretty respectable examples. A high incidence of gawky, mean submissions at a wine show does not necessarily mean that there are no good wines out there. In fact, there is an argument which says that because good merlots are few and far between, the really successful ones sell out and never make it to the judging benches.

The country's top merlot producers are acutely aware of the problems. They have an association which benchmarks standards. It has been investigating the variety's general underperformance in SA. Their website www.merlotforum.co.za carries the details of their research and reveals an uncommon level of transparency among wine industry associations.

It seems increasingly clear that the cause of merlot's problems in the Cape will not be susceptible to a quick fix. The evidence suggests that improvements in the average quality of Cape merlot will only occur when there is a wider range of clonal material and the vines are planted in places best suited to their particular needs. This nexus of site and variety - self-evident after centuries of viticulture in Europe - has been slow to take root in most of the New World wine-producing countries. DrTony Jordan, a senior Australian show judge and one of the panellists at this year's event in Paarl acknowledged that Australian merlot is equally disappointing.

In the days before the European Union stood by to bail out any misguided attempts at agriculture, farmers were highly sensitive to the concept of crop and clone selection. If what you produced was no good, it didn't sell and you starved. Centuries of this kind of experience determined that merlot worked well in and around Bordeaux - mainly on the Right Bank - just as pinot noir is best suited to Burgundy, and grenache to the Rhone and much of Spain. When New World producers chose to plant merlot they thought of it as an important or necessary ingredient for a fine wine industry - based on its performance in select sites in France. Researching where it should be planted, and what clones to use, was largely incidental.

The variety arrived SA in the 1970s, and initially it was seen as a source of instant wealth. Meerlust was one of the first producers to bottle it as a single variety, and it sold for more than the already super- premium Rubicon. Then, everyone wanted it for Bordeaux blends (which in turn fetched more than straight cabernet sauvignon). With such demand, and so useful a place to bury the defective examples, it took a very long time before its shortcomings were noted. Then denialism - another South African malaise - moved in and it has taken until now for the extent of the problem to be acknowledged.

I cannot pretend to have tasted a large percentage of the more than 100 different merlots bottled in SA. I like the wines of Luca and Ingrid Bein (sold under the Bein label) as well as Meerlust, Thelema, Fleur du Cap and Graham Beck. You could also do worse than drink Leopard's Leap - it may be very commercial but it's pretty easy to drink.

Clearly not all merlot is bad news, so the debate is really related to your risk-aversion index. If you're a duck- or-no-dinner type, go ahead and order an unknown merlot. If however, you seek greater certainty before you risk your cash, there are safer choices.

- Michael Fridjhon has a financial interest in the Trophy WIne Show and chairs the judging panel.

From Business Day, 5 May 2010