— 12 
Study is not sufficiently extended to justify fuller discussion, 
but a study of my notes indicates that the samples cut in 
early bloom have suffered the least change ; also that the 
change is less in one year than in two years. The last three 
samples were only one year old at the time the analyses 
w^ere made. 
The increase in the percentage of water in the sample 
and the increased percentage of nitrogen in the dry matter 
seem to me to point to a transformation in the constituents 
included under the term nitrogen-free extract. It is cer- 
tainly beyond question that the nitrogen cannot be in- 
creased by absorption, and it is improbable that the material 
which we class as crude fiber will break down rapidly 
enough to account for the formation of so much water, in 
some instances four per cent., especially when we remember 
its ability to resist the action of dilute acids and alkalies. 
Our experiments indicate that there is no loss of the 
proteids, but that chemical changes take place to a con- 
siderable extent in some of the other constituents, probably 
in those complex and less stable compounds grouped under 
the head of nitrogen-free extract and frequently spoken of as 
carbohydrates. The apparent increase of nitrogen is easily 
accounted for by the elimination of water and probably of 
other compounds also, as oxides or hydrides of carbon. 
We have in the above explanation of our facts, I think, 
a full and satisfactory explanation of such facts as have 
been observed relative to the saving of hay, for instance, by 
putting it in a mow, in which case the loss of weight is much 
less than when the hay is kept in the stack. Our applica- 
tion of the observed facts would be, that, other things be- 
ing equal, hay preserved in the stack is more freely exposed 
to influences which promote these changes in the less stable 
constituents of the hay ; and the loss due to these changes 
becomes noticeable. 
I understand that it is found almost impossible, in prac- 
tice, to feed out anywhere near the amount of hay for a ton 
put in a stack, that is usual to feed out when the same 
weight is put into a mow. It is evident that some hays may 
age very much faster than others, due to kind of hay — al- 
falfa, clover, timothy, etc.^ — also due to development at time 
of cutting, manner of storing, and other conditions. 
I'here is, perhaps, a suggestion in the figures represent- 
ing the percentages of water found in i8q6, that the sam- 
ples in i8q 4 did not really represent air dried hay and that 
in the course of the intervening year and a half, or more, 
they had really become such. 'Fhis receives support from 
