Issue link: http://winesandvines.uberflip.com/i/629061
February 2016 WINES&VINES 49 PRACTICAL WINERY & VINEYARD BARRELS for 90 minutes. (See the wines and experimental setup in the table "Nature and Origin of Wines.") For red wines, an international panel of trained tasters (17 for the 2013 Cabernet Sauvignon and 10 for the 2014 Merlot) from the National School of Agricultural Sciences of Bordeaux (Bordeaux Sciences Agro) participated in the blind sensory analysis. For white wines, a panel of 10 trained tast- ers participated. For each series, there were two stages of sensory analysis: a tri- angle test followed by a descrip- tive analysis in pairs (Chardonnay and Cabernet Sauvignon estate D) and trio (Merlot). Tasters were asked to rate the wines from zero to five on criteria related to the perception of oak in wine and wood-wine harmony. The results were subjected to statistical analy- ses to highlight the significant differences between wines: X 2 test for the triangular test and two- factor Anova without repetition for descriptive analysis. The sig- nificant differences are shown in "Sensorial Analysis of Cabernet Sauvignon-Estate D" on page 52. The impact on tannin reactiv- ity was evaluated by measuring an index of salivary protein (SPI). 9,10 The assay of wood aromatic compounds was performed on wines by Stir Bar Sorptive Extrac- tion/gas chromatogram/mass spectrophotometry and liquid chromatography/mass spectro- photometry, which are internal methods at the Sarco laboratory in Bordeaux. Grape-based tannin composi- tion and concentration were de- termined using analytical methods capable of differentiating grape tannins from oak tannins. 11 Results and discussion Measuring tannin potential on a sample of more than 2,000 staves of various origins and grain width confirmed the variability of wood ellagitannin concentration. The relationship between tan- nin potential and grain was inves- tigated for two grain widths: growth ring width between 1.5 and 2 mm (FI) and more than 3 mm (GR). In accordance with data literature, 12 ellagitannin con- tent increased with the grain width. However, this correlation was weak; therefore, grain width was determined not to be a strong predictor of stave ellagitannin content. Triangle tests: For a 2014 Chardonnay tasted after 12 months of élevage, the triangle test clearly discriminated the two samples with 95% confidence. For a 2014 Merlot tasted after 12 months of élevage, two triangle tests have been performed: MTP compared to LTP, and MTP com- pared to HTP. The panel clearly discriminated the samples, the differences being more significant for the set MTP/HTP (99% confi- dence) compared to LTP/MTP (99.9% confidence). Mouthfeel perception: For the Chardonnay, the wine aged in HTP barrels exhibited signifi- cantly more pronounced round- n e s s a n d l e n g t h , b u t t h e descriptive analysis could not dis- criminate the two samples on the balance criteria: LTP gives more freshness and HTP more round- ness, the two modalities being complementary. For the Merlot, the descriptive analysis clearly discriminated the wines on length and fruity retro- nasal criteria: Length increased along with TP, while fruitiness decreased. No significant differ- ences were found between the three samples on the criteria of tannin intensity and quality. Tan- nin reactivity measured by an index of salivary protein was very close (respectively 45%, 49% and 51% for LTP, MTP and HTP). During élevage, the ellagitan- nin concentration is known to increase during the first months and then decrease slowly due to ellagitannin oxidation and/or condensation with the wine con- stituents. 5 It is also established The references for this article are available online at winesandvines.com