18 BULLETIN 1073, U. S. DEPARTMENT OF AGRICULTURE. 
When sampling could begin soon enough, the fruits showed a con- 
sistent and more or less uniform increase in fat up to a certain point, 
after which the increase was much less (fig. 1). Often, late in the 
season, apparent decreases are indicated, showing at least that the 
increases were not sufficient to overcome the variability of the 
samples. In the limited experience of the authors, it would seem 
that the point where the uniform increase in fat ceases is about the 
point where a satisfactory maturity is found. With the Fuerte this 
occurred in December, with the Puebla in February, with the Lyon in 
May, with the Blakeman in May, with the Spinks in March, with the 
Taf t in May, and with the Sharpless in April or earlier. The Dickin- 
son samples afford no data on this point, as the sampling ceased before 
the fat content had become constant. 
The reducing substances or sugars in the pulp of the avocado 
decrease markedly as the fruit ripens, but the range is hardly suf- 
ficient to be of use in estimating maturity. Little or no sucrose is 
present in avocado pulp. 
The crude fiber changes but little during the growth of the fruit, 
such changes as take place probably being due to the variability of 
the other constituents. From the data given in Table 3, showing 
the actual content of fat on a water-free basis, however, it is evident 
that at the time the rapid increase in fat ceased, the percentages on a 
water-free basis in the different varieties were, with one exception, 
close to 70. The following are the figures: Blakeman, 70 per cent; 
Dickinson, 60 per cent; Fuerte, 75 per cent; Lyon, 70 per cent; 
Puebla, 73 per cent; Sharpless, 72 per cent; Spinks, 68 per cent; 
Taft, 72 per cent. It might be supposed from these data that 
Dickinson was not yet mature, inasmuch as sampling stopped at this 
time, but miscellaneous samples harvested even later the following 
year showed a lower content of fat than was shown by these samples. 
Experience with fully matured fruit of other varieties indicates that 
avocados rich in fat usually contain at least 70 per cent on a water- 
free basis, but the rule does not hold when the fat content at maturity 
is low. 
One other source of information concerning changes taking place 
while the fruit is ripening is afforded by the comparison of the datf, 
resulting from the analysis of the fresh and storage samples. When 
these data are confined to the edible portion of the fruit, the differ- 
ences are more striking after the data have been reduced to a water- 
free basis (Table 3). Man}- interesting phenomena are revealed by 
a close study of these results. Some of them may be the result of 
errors in the analyses or of the natural variability in samples con- 
sisting of but two or three fruits, but most of them are undoubtedly 
the result of changes in composition of the fruit after it is removed 
from the tree. It is hardly practicable to analyze part of one fruit and 
