There’s one recent movie that all winemakers can easily identify with: “Julie & Julia.” When you watch Meryl Streep playing Julia Child, who coaxes the best out of her ingredients like a virtuoso symphony conductor -- testing this spice and that herb, perpetually tasting and adjusting for balance as the dish takes shape -- you get a familiar feeling that you know this stuff in your bones: It’s the essence of winemaking. Her arena may be food, and yours wine, but the same drive to experiment in pursuit of perfection marks that trial-and-error approach as a common obsession.

For winemakers, the process ideally extends to barrel selection. “Ideally” because where there’s a will there’s not always a winery budget to support broad experimentation, not when French oak tops $1,000 per barrel. Some cooperages recognize these limitations and help small wineries conduct informative component identity sessions and barrel trials before purchasing their oak, allowing winemakers to make educated buying decisions and still remain solvent.

We all like to think we can distinguish smoke from toast or vanilla from caramel by taste and aroma. But these barrel components, produced by primary chemicals such as furfural (sweet toasty aroma), volatile phenols (vanillin), lactones (woody aroma) and carbohydrate degradation are sometimes elusive, as I discovered when I recently participated in a blind-tasting flavor identity and trial of French and American oak barrels at Napa-based Cooperages 1912, which represents World Cooperage and T.W. Boswell. I’ll get to the details later, but here are the takeaways for those of you hurrying off to rack or stir lees:

Know your goal
Cooperages 1912 research director David Llodrá and sales manager Jason Stout make the point emphatically: You’ve got to have a purpose and a goal to get a good result out of your efforts. Don’t waste time and money selecting barrels without a specific objective. They add that if you’re evaluating wine in new barrels, you’re tasting 100% new oak, and your bottled wine will most likely not average anything much above 50% new oak, if that. You’re tasting, then, for characteristics, not final results.

I gave Stout and Llodrá a hypothetical goal: I wanted my Anderson Valley dark-fruit Pinot Noir to develop some Russian River undertones -- cola, mushroom, a hint of dark coffee. They suggested I pass that goal on to my barrel supplier to reduce the variables, then experiment with three or four different barrel “composites.” (Composites are identical versions of the same barrel. Wood being a natural product, every barrel is slightly different from every other.)

If I planned to bottle 500 cases of that Anderson Pinot, I’d need 20 barrels. Expecting to end up with 60% new oak, eight would be neutral as a control. The remaining 12 might break down to three barrels each from four French tonnelleries. The company representative would help me choose barrels from its selection that gave me the singular component I wanted -- forest floor, mocha and so forth. I’d plan to buy three of each, new, after evaluating the barrels in a preview tasting session arranged by the cooperages.

Design your experiment
Barrel evaluation isn’t an exact science, true. As Joseph Phelps Vineyards’ director of winemaking Damian Parker points out, historical experience can play a leading role in the way that winemakers go about figuring out how to pair wood with wine. It’s human nature to go back to cooperages where you’ve been treated well and been pleased with the results.

That said, objectivity is a valuable asset. Stout tries to get winemakers to identify specific variables that they want to explore between barrels -- toasted or untoasted heads, for instance -- and to draw up a barrel-trial plan that maximizes consistency while reducing chance. He recommends sticking to a tight tasting schedule and taking accurate notes along the way. If all this sounds a bit like eating your broccoli -- excruciatingly beneficial -- it is. But if in the end you produce a wine that’s precisely balanced, exquisitely nuanced and deliciously complex, who’ll take the credit? You, not your cooper.

Consistent method of evaluation
Create a sheet that ranks the wines you’re tasting. Include the major flavor and aromatic attributes of oak: vanilla, smoke, spice, toast and fruit. Include mouthfeel and overall preference, and leave a space for general comments about each barrel sample. If your composite consists of three barrels from one cooper, draw one-third from each barrel. Score blind for preferred attributes, and quantify the results. The real challenge in all of this, most veterans agree, is to taste in a consistent manner multiple times and to stay focused as the wine changes from post-fermentation to bottling.

Setting up a trial
Every winemaker takes a different approach to barrel trials. Kimberlee Nicholls at Markham Vineyards in St. Helena samples eight barrels, all the same (i.e., composites) at one time. For Chardonnay -- highly recommended as a variety for checking out barrels -- she experiments with wine from the same vintage and vineyard that has and hasn’t gone through malolactic fermentation.

She begins pulling wine to taste in March following the harvest, and every three months after that. “By June is when I can first begin to tell,” she says. There’s always a neutral barrel as a control. She tastes blind with her coopers, Markham’s enologist and associate winemakers, scoring for aroma, palate, creaminess and so forth. Eventually she’ll blend together wines from these medium-plus barrels to produce her Chardonnay in Markham’s signature style -- with a dollop of butterscotch and balanced tannins and acids. That’s her goal. She knows she’ll be using 30% new oak, and adjusts ratios with that in mind.

Todd Graff at Frank Family Vineyards in Calistoga has a mantra: Keep it simple. He asks coopers for their house blends, buys a minimum of four barrels from one supplier, keeps detailed trial records but waits two years “to see where the barrel is going” when producing a more opulent red like Frank’s Cabernet Sauvignon. Establishing tight relationships with a few trusted suppliers is better, he adds, than relying on a large group you don’t really know.

However you choose to organize your barrel trials, the first step, logically, is to know what you’re tasting, so that you and your cooper are speaking the same language. Stout typically creates a “standards” flight -- six glasses, in my session, all containing identical Merlot. Two millimeters of smoke, spice, toast or vanilla concentrate have been added to the glasses, one acting as a neutral control, and one a duplicate.

By aroma and taste we tried to identify each oak attribute, and marked the intensity from 1 (lowest) to 7 (highest) on our scoring sheets. Vanilla and spice were the easiest to identify; smoke and toast, I discovered, produced closely aligned results for me. I thought I could quickly distinguish one from the other -- I couldn’t.

I experimented again with both components until I could pick up the difference. More to the point, our dialogue about what a specific barrel might offer in my quest to add mocha and cola components to my Anderson Valley Pinot Noir took a significant turn. It bypassed guesswork; the cooper and I could now target a specific barrel with a common goal and vocabulary in mind.

For the second step -- identifying French versus American, also different toasting levels -- we sampled nine glasses all containing Central Coast Syrah from the same vineyard and same vintage. Each barrel represented a different toasting “profile” -- a slow or fast ramp up to high heat, loose or tight grain, and other variables. Cooperages 1912 creates more than 100 such profiles, each graphed and annotated.

Again, we ranked the wines for intensity of spice and other salient oak characteristics and overall preference. We were continuing to develop a shorthand means of communication. Of our senses, taste and aroma may ultimately be the most elusive and subjective, so any strides toward shared objectivity can only help.

In the end, barrel choices depend on a host of factors -- starting of course with grape variety. These days, many wineries’ downsized production budgets also play a significant role, as oak costs escalate in reverse proportion to descending retail prices for well-crafted varietal wines. The right wine in the wrong barrel isn’t just an aesthetic mistake in this market, it’s the kind of expensive blunder that a small winery can literally no longer afford to make.

“You have to do what’s right for your wine,” says Rodney Strong winemaker Rick Sayre, who has been evaluating barrels for more than 40 years. “I suggest Chard or Pinot to check out barrels: You don’t want complexity or gaminess in the wine. Nobody can see the wood through the juice in mountain-vineyard Cabernet, for instance.

“I soak and process all my barrels identically, fill them on the same day and keep exact records throughout the year. I also replicate four of the same barrel to hedge against individual barrel quirks when I taste. I make sure there aren’t any toasting blisters in the barrels and run down a checklist like a pilot.

“What I’m after is elegance, harmony and above all integration. All the smells and flavors have to meld together. I don’t want tannins over here and oak over there and acid somewhere else. So that’s the goal. And now that we’ve got toasting technologies to dial into, it’s up to me to be able to work with my cooper to translate, say, creaminess, into a barrel profile, so we both know what we’re trying to achieve and what will give it to us.”

Stout sums it all up neatly: “Trials justify purchases. It’s the way to make responsible decisions. You can’t make better wines by cutting back on barrels, even when money’s a major factor. But what you can do is to buy the right barrel to maintain quality, and that’s what smart evaluation practices are all about.”

Stephen Yafa produces limited release Pinot Noir in the Russian River Valley for his winery, Segue Cellars, seguecellars.com. To comment on this article, e-mail edit@winesandvines.com.