CHANGES IX COMPOSITION OF CALIFORNIA AVOCADOS. 19 
store the remainder until it has softened before analyzing it. The 
best that could be done was to select a lot containing from two to 
six fruits, as nearly uniform as possible, analyze half this number as 
soon after picking as possible, and hold the other half until they 
had become soft. The data must therefore be considered in the 
light of these facts. Where, however, changes are almost always in 
one general direction, the probability of their being the result of 
individual variation is remote. 
In all, 40 samples were analyzed at once and after storage, the 
number being distributed among immature and mature fruits. Some 
difference between the specific gravity of the fresh and storage fruits 
of many varieties is shown. In some varieties the storage samples 
have a much lower specific gravity than the fresh, but in other 
varieties the reverse is true. Peculiarly enough, also, the position in 
each type when the fruits are thoroughly mature shows a tendency 
to reverse itself. Cases in point are the Taft, six samples of which 
were examined, and the Fuerte, eight samples of which were examined. 
The Taft samples in September had the same specific gravity, but 
in October, November, and December fresh samples had a higher 
gravity than the storage samples picked at the same time, while in 
May and June the reverse was true. The Fuerte samples for Sep- 
tember also had approximately the same specific gravity. The 
October, November. December, January, March, and April samples of 
fresh fruit, however, had a lower specific gravity than the storage sam- 
ples, while in May the reverse was true. Also in the case of the Lyon 
the fresh samples showed a higher specific gravity until June, when 
the storage samples had the higher density. On the other hand, the 
Puebla samples always showed a difference in specific gravity, the 
soft samples having the higher. Unfortunately, the samples were 
exhausted before thorough maturity was reached, so that any change 
late in the season escaped observation. The December samples of 
the Spinks showed a higher density in the fresh sample, after which 
the reverse held true. 
Omitting the data on the Blakeman, only one double sample of 
which was examined, the varieties having the heavier skins (Dickin- 
son, Lyon, and Taft) have higher densities in the fresh samples. The 
thin-skinned fruits (Fuerte, Puebla, and Spinks) show a higher 
density in the case of the storage samples. Sharpless also is rather 
thin-skinned, but acts in this respect like a thick-skinned variety. 
In several cases the pulp or edible matter increased during storage 
of the samples. Two factors may account for this — individua 
variation in the fruit and inability to separate satisfactorily the 
skin and the pulp in the very green samples. In this connection 
it will be noticed that the proportion of skin in the storage samples 
is almost always lower than that in the fresh samples. Of the 40 
