28 
Colorado Experiment Station 
It is perhaps proper that I should say a word in regard to these 
determinations. It is a difficult task to manipulate the destruction of 
60 or 70 grams of wood without losing some arsenic, again there is 
some loss in the Marsh-Berzelius method, again it is impossible to weigh 
such small quantities except on a button balance, which is not always 
at hand. In this connection I wish to acknowledge my indebtedness 
to Wm. Ainsworth & Sons who kindly placed one of their balances at 
my disposal. This balance was claimed by these makers to be accurate 
to 0.005 milligrams, and the balances are as represented. The weigh¬ 
ings are accurate but the difficulties in the recovery of the arsenic will 
not justify us in claiming the highest degree of accuracy for them. 
One thing however can be said, i. e., they are all too low, and this with¬ 
out an exception. This I think will be readily acceded by persons 
competent to judge. The pages of a bulletin are not, in my opinion, 
the place to discuss analytical methods, therefore this subject will not 
be mentioned further than to reiterate the statement made in Bulletin 
131 : ‘“That the reader is assured that the arsenic reported was not con¬ 
tained in any or all of the reagents used. The care taken was in all 
ways as circumspect, so far as the analytical work was concerned, as 
though human viscera were in my hands.” Trouble was had in obtain¬ 
ing zinc so nearly free from arsenic that one could use it without mis¬ 
givings. 
The Part Played By Alkalis. 
I was very explicit in my statements regarding this subject in 
Bulletin 131. I see no reason to change any statement made therein, 
namely that lead arsenate yields arsenic quite readily to dilute solutions 
of sodic sulphate or sodic chlorid and these are our common alkali 
salts. The other salts always understood, when we use this term, are 
the sulphates of lime and magnesia. Distilled water dissolved arsenic 
acid from the arsenate of lead which we used though it, the lead salt, 
had been previously washed, but water to which sodic sulfate or sodic 
chlorid, ordinary salt, has been' applied will dissolve it more readily. 
This is true of the neutral lead arsenate and probably in a higher de¬ 
gree of the acid arsenate. Arsenite of lime (lime, sal soda and arsenic) 
is quite soluble in water but more readily so in water to which sodic 
chlorid or Glauber’s salt has been added. I have, however, met witn 
soils in which there is so good as no alkali and yet a very considerable 
amount of arsenic can be dissolved out of this soil by distilled water. 
I have even obtained soils from states in which alkali is not supposed 
to exist and these soils, too, contain arsenic which is soluble in distilled 
water and the quantity is so significant that it raises a doubt in my 
mind as to how big a part our alkalis actually play in the matter. It 
is evident that they may take some part in bringing the arsenic into 
solution in water but they are certainly not the primary cause of its 
solubility. As elsewhere stated, all of the arsenical compounds used 
as spray materials, Paris green, arsenite of lime and lead arsenate, are 
perceptibly soluble in water and consequently any soil containing these 
salts in sufficient quantities ought to yield arsenic to pure water, as 
