eMail Us . Facebook . Twitter

Oct 21, 2017

Search our Site


Advanced Search

From Our Archives...


Wine X World Headquarters

© Copyright 1997 - 2015
X Publishing, Inc.

home  |   archives   |  about us  |  events  |  media kit  |  

Lab Work
by Brendan Eliason
Magazine Issue: U.S. Vol. 4.3

So, Ya Wanna Be A Winemaker?

Most of the "real" work's been done on the current vintage. We've moved the wine into barrels and can let it sit there for a while. It's now time to focus on previous vintages. For example, this year we started getting some unexpected aromas/flavors in a couple of our '96 wines. Although these have been sold out for almost three years, we're still interested in how they develop and change over time. The aromas/flavors we started noticing in our '96 estate cuvee and 1996 petite sirah were very subtly earthy, minerally and fungal in nature. Although these new smells add a very nice level of complexity to the soft fruit and spice elements in the wine, they're not characteristics we expected. We're intrigued.

In The Winery

Our best guess about the source of these new aromas/flavors is that they come from a wild yeast strain called Brettanomyces, or simply Brett. Brett often lives in used wine barrels and can cause in wine very distinctive characteristics, like anything ranging from pleasant earthy smells to overpowering aromas/tastes of wet horses and barnyards. Our curiosity piqued, we sent wine samples to Vinquiry, a local wine lab, for analysis. The basic results looked like this:


Culture for Brettanomyces: 0.2 ml spread plate: 5 cfu/ml 50 ml membrane-filtered: approximately 150 colonies

Culture for Bacteria: 0.2 ml spread plate: lawn of Pediococcus 50 ml membrane-filtered: confluent growth

Now, my degree's in viticulture, not bacteriology, so when we received these results, I was forced to humbly quiz the head of microbiology at Vinquiry about what the results actually mean. Evidently, there are four different levels of measuring growth. The first level simply involves counting the number of bacteria colonies in a given petri dish that's been inoculated with the wine. The second level refers to "Colonies too numerous to count" -- usually somewhere around 150 or more. (There's no set cutoff between being able to count or not count colonies.) The next level's "Lawn," a consistent and solid layer of colonies with no empty space. Finally there's "Confluent Growth," which is wall-to-wall growth, with no differentiation of colonies.
Now you're probably sitting and thinking the same thing that I was: Yeah, so what does this mean? Is five cfu/ml of Brettanomyces good? Is it bad? And most importantly, how does this affect my wine? The short answer is that there's no short answer. The longer answer is that the answer depends. On what? Lots of things. Take, for instance, the reading for the Brett on the petite sirah, which is five cfu/ml using the 0.2 ml spread plate. Evidently, five cfu/ml isn't that bad. But it's also not good. The bad part's that Brett can reproduce in bottled wine, so what's five cfu/ml today could end up as much more later. The good news... Brett readings can reach into the thousands, so five's pretty low.

None of this really matters, however, because of the largest confounding element -- multiple species. There are many different strains of Brett. Each strain has unique characteristics. For some strains, 40 cfu/ml is a lot and can affect a wine's flavor. For other strains, 40,000 cfu/ml can leave the wine untainted. Which strains do we have? We don't know. We have the option of sending the samples to a lab on the East Coast that'll separate the strains, but that'll set us back $105 per sample. We probably won't do this unless it becomes an actual problem.

A simpler answer is to run a four-ethyl phenol test. "Of course!" you're thinking, "The four-ethyl phenol test. Why didn't I think of that?" Four-ethyl phenol's a by-product of Brett that's used to impartially quantify the level of actual sensory contamination (i.e., how much you smell/taste). In general, if you get a four-ethyl phenol number less than 500, you're fine; from 500 to 1000, well, you have a definite impact; and around 3,000 you get Chateau Lynch-Bages (a Bordeaux Grand Cru known for its pronounced Brett). Okay. Now you're thinking, "Aha! I understand. I know what the numbers mean. This all makes sense now!" Unfortunately, my response is still a resounding, unequivocal "sort of."
This brings us to the final variable for consideration. Is Brett bad? Let's assume that we get to the point of knowing that a certain strain of Brett's in our wine, and that it has a certain quantifiable effect on its smell/taste. Should we be unhappy? According to some people (and the Vinquiry scientists that run our tests) the answer's a definite yes. To others (Dave and me included), it really varies from wine to wine, and in some cases (e.g., the Chateau Lynch-Bages and many Rhone wines) Brett can actually add a nice level of complexity. So far our opinions of the 1996 estate cuvee and petite sirah fall into the latter category. We didn't have the wines tested because they were bad, we simply had them tested because they tasted slightly different than expected. Whatever the impact of Brett on our wines, they still taste wonderful.
In summary yes, no, maybe, sometimes, sort of and"only if you think so."

Check out Brendan's "Harvest Diary -- A week in the life of Crush at David Coffaro Winery" at

E-Mail a Friend

Add Your Comment





Remember my personal information

Notify me of follow-up comments?

Please enter the word you see in the image below:

Back to top

home  |   archives   |  about us  |  events  |  media kit  |  

Sister Sites