74 §S. W. Johnson on some points of Agricultural Science. 
on its right. Way remarks, ts ‘of yong the reverse of this 
action cannot occur.” Prof. Li g (Ann. de Chem. u. Phar. 
xciv, 380) has drawn attention “the fact that Way *denently 
contradicts himself in describing the preparation of the potash- 
alumina-silicate, which may be obtained by digesting either the 
lime-alumina- or soda-alumina-silicate in nitrate or moephate of 
potash, when the soda or lime is dissolved out and replaced by 
tash. 
Way was doubtless led into the error of assuming ‘a fixed 
order of replacements by considering these exchanges of bases 
_ as regulated after the ordinary manifestations of chemical affinity. 
His own experiments abundantly show that among these silicates 
there is no inflexible order of decomposition, nor any complete 
replacements. 
Liebig, in the paper just cited, was led from this contradic- 
tion and from other considerations, to reject the conclusions of 
Way, capouialiy as there was no direct proof that these double 
silicates exist in soils. 
he er ev of Eichhorn, “ Ueber die Einwirkung 
verdiinnter Salzlésungen auf Ackererde,” _ Cancers 
liches Centralblatt, 1858, ii, 169, and Po gg. A ., No. 9, 1858,) 
have cleared ~ the discrepancies of Way’s heeuion (which 
is itself one of “(pce interest), and have confirmed and ex- 
plained his fa 
As Way’s ‘artificial silicates contained about 12 per cent of 
water, the happy thought occurred to Eichhorn to test the action 
of saline solutions on native hydrous silicates. He accordingly 
instituted some trials on chabazite and natrolite, an ieteons of 
which is here given. 
On digesting finely pulverized chabazite with dilute solutions 
of chlorids of potassium, sodium, ammonium, lithium, barium, 
strontium, calcium, magnesium, and zinc, sulphate te of magnesia, 
carbonates of soda and ammonia, and nitrate of cadmium, he 
a part of the silicate, while lime passed into the solution, The 
rapidity of the repla acement varied exceedingly. The alkali- 
chlorids reacted evidently i in two or three days,  Chlorid of ba- 
rids of zine and strontium at first, appeared not to react; but 
after twelve days, lime was found in the solution. Chlorid of 
magnesium was still tardier in replacing lime. 
our pear! es powdered chabazite were digested with 4 grams 
chlorid of m and 400 cubic centimeters water for 10 days. 
The iron rae the original mineral (1), and of the same 
ie, the action of chlorid of sodium (II), were as follows: 
