THE CONSTRUCTION OF PROTEINS 171 
observers to be shared by certain lowly Algw, but the 
evidence as to their activity in this direction is conflicting. 
It may be that they are capable of a similar symbiotic 
relationship with certain of the nitrogen-fixing bacteria of 
the soil already mentioned, but it is more probable that it 
is carried out by bacteria living simultaneously, but not 
symbiotically, in the soil with them. Several of these 
organisms appear to be associated symbiotically with each 
other, though not with any green plant. 
When we turn to the method of construction of protein 
by a green plant we find ourselves in possession of very 
little accurate information as to the stages which are 
involved. We find that nitrates especially are absorbed 
by the root-hairs from the soil, and that a continuous 
stream of them passes into the plants. This naturally 
is associated with a transportation of the nitrates through 
the root and stem. They can be detected in varying 
quantities in these regions, but the amount seems to diminish 
as the termination of the stem is approached, and little 
can be found to be present in the leaves. It may be inferred 
that a gradual decomposition takes place as they pass 
along the axis, and that this is completed in the leaves. 
A theory has been advanced to explain this disappear- 
ance, which may be mentioned here. It is that the nitrates 
are decomposed by the organic acids of the plant, and in 
particular by oxalic acid. Simultaneously the sulphates 
which are absorbed undergo a similar fate. The resulting 
bodies, the nitric and sulphuric acids, unite with some form 
of non-nitrogenous organic substance, possibly formal- 
dehyde, or a fairly simple carbohydrate, to form protein. 
From what has already been advanced, however, it is evident 
that this scheme of construction is purely hypothetical. 
When we search for a form of nitrogen compound that 
is nearer protein in its composition than these simple salts, 
it is natural to look at the products of the decomposition of 
protein material to see if these furnish any clue to a possible 
constructive process. When proteins are digested in the 
