Cellar scene

Wines & Vines, Feb, 2005 by Tim Patterson

When it comes to yeast, there's broad agreement on one key point: no yeast no wine, no industry Beyond that, knowledgeable people can sound like they come from different planets.

Mention yeast to Shea Comfort, and you'll get an earful. Young, bright, energetic, self-taught, Comfort ran yeast comparison trials for years in the back room of a winemaking supply house, and now he's putting that experience to work at a startup winery. "Yeasts are little factories," he says, "and some of them just don't make certain compounds, some of them make a ton of those compounds." He talks about yeasts that add "notes," others that strike "chords." He talks about using multiple yeasts to "build a wine the way you build a building. You're not really a winemaker," he says, "You're a yeast farmer."

Catch your breath, and give a call to laid-back Rick Longoria, veteran of 30 years of winemaking under his own and several other labels in Santa Barbara. "For most of my career," he says, "I haven't been a yeast man." Confronted by the all the recent buzz, he's made a habit of trying a batch or two every year with this or that new strain, "and they all usually turn out fine. I haven't been knocked out by any huge differences with any strain on any wine I've made. You can get to the point where you're paying attention to so many factors, you drive yourself nuts. For me, yeast is right about at that point."

How much does the choice of a yeast strain--or the choice not to choose one--really matter? Beyond individual testimonials, how hard is the hard evidence?

Dubious Davis

One pole in this debate was set almost a decade ago in Principles and Practices of Winemaking (1996), a collaboration among four prominent UC Davis researchers: Roger Boulton, Vernon Singleton, Linda Bisson and Ralph Kunkee. The chapter on yeast fermentation exuded a good deal of skepticism about claims that different yeast strains produce different organoleptic results. In particular, they argued that observed differences could easily have to do with nutrient levels and juice composition, not yeast strain; that variations in the production of volatile esters tend to dissipate, along with the esters themselves, in a matter of months; and that aromatic differences due to higher alcohols and textural differences due to glycerol levels were likely to be below sensory thresholds.

Their conclusion: "Until published evidence is presented involving verified replicate fermentations and stringent sensory analyses, any effect of wine yeast strain on flavor is disputable." Some winemakers at the time found statements like this annoying, but they were nonetheless accurate: Precious little rigorous research had been published by the mid-'90s.

Things have changed substantially since then. "It's pretty clear that yeasts do make a difference," says Thomas Hennik-Kling of the New York State Agricultural Experiment Station at Geneva, one source of such research. Other interesting work has come out of Oregon State University, Brock University in Ontario, several research centers in France, even the University of Zagreb in Croatia. Differences have been documented in the production of a number of compounds through chemical analysis, and in persistent aroma, flavor and mouthfeel characteristics through formal sensory analysis.

Not surprisingly, a lot of this research has been conducted or orchestrated by companies that sell yeast. Ann Dumont, enology communications coordinator for Lallemand in Montreal, says that today, "there's no denial that yeast has an effect." Lallemand's offerings take from three to five years to move from field selections to a final dried, commercial product, and the testing process is an exacting combination of chemical and sensory analysis. (She notes the company is still running tests on that old warhorse, EC-1118.) At the moment, Lallemand has 150 yeasts on the market. Most winemakers I spoke to felt that the yeast companies aren't selling snake oil; when they claim strain X encourages the production of polysaccharides, it probably does.

Back at Davis, Linda Bisson agrees, up to a point. "There clearly are differences you can measure," she says, "and yeast signatures that can be identified." But the Davis experience with sensory analysis has been much less convincing; trained tasting panels often cannot spot differences. "It's common for us to get different results. There are so many environmental influences. Differences in juice composition and fermentation conditions are critical; California conditions can obscure differences that show up in European fermentations. There's some work from France that suggests that differences disappear once you add DAP (diammonium phosphate, a commonly used yeast nutrient).

"Winemakers might be able to tell the difference," she says, "but trained student panels can't, and the bottom line is, consumers can't. If there's a difference and nobody can smell it, is it a difference?"

Trial, Error, Success

Winemakers, even big-time yeast fans, agree that the only way to know what a strain will really do is to re-hydrate some. "It's essentially trial and error," says Morton Hallgren, chief winemaker for Dr. Konstantin Frank's Vinifera Wine Cellars in Hammondsport, N.Y. Hallgren uses several strains for different batches of Riesling, the flagship wine. For dry styles, VL1 brings out floral and mineral characters; for sweeter styles, QA23 emphasizes apricot and pineapple fruit; along the way, D47, Heiligenstein, Wadenswill 27, Freddo, S6U and a few others are "complexing agents, part of the layering." Hennik-Kling at Geneva has done extensive work with the New York wine industry on appropriate yeast selections, but he points out that "our Chardonnay and Riesling yeasts might not work in California."