36 6 
I . 
11 . 
III . 
0 . 
•22 
/ 
•330 
1 . 
•55 
•120 
2 . 
. -67 ' 
•050 
3 . 
. -72 ' 
•022 
4 . 
•742 
•017 
5 . 
. -76 ' 
•012 
6 . 
•772 
•010 
7 . 
•782 
•007 
8 . 
•79 ' 
•005 
9 . 
•795 
•005 
10 . 
•80 ' 
It will thus be seen that the rate at which the B colour 
changes into C is far more rapid in the case of new wine, 
when much of it is present, than after the wine has become 
older ; being, in fact, about ten times as rapid in the first 
year as in the third, and about 100 times as rapid as in the 
twentieth. On this account the difference for each year is at 
first so considerable that wines of different vintages could 
easily be distinguished ; but after about six years the differ- 
ence is so small that it would be difficult, or impossible, to 
determine the age to within a single year. After twenty 
years a difference of even ten years does not show any striking 
contrast, and the age could not, therefore, be determined to 
nearer than ten years by this process. However, up to six 
years I think it quite possible to determine the age to within 
a single year. I took specimens of various ports from the 
casks, of different ages up to six or seven years, and labelled them 
in such a manner that I did not know the age of any, but 
could ascertain it afterwards by reference. I then made the 
experiments with great care, and found that by proper atten- 
tion to the details described above I could correctly deter- 
mine the year of vintage of each particular specimen. 
As already mentioned, the change of the B colour into the 
C is most probably the effect of the oxygen of the air, which 
appears to act slowly through the wood of the cask ; but the 
relative amount of these two colours is also modified by the 
deposit of crust. When kept in well-corked bottles the 
change does not appear to be the same, and goes on far more 
