BULLETIN OF THE BUSSEY INSTITUTION. 295 
them was discarded as the work went on and experience showed 
that it was not absolutely needed. ‘This plan of employing the 
chemicals in solution was found to work much better in practice 
than the old one of mixing dry chemicals with the soil before 
planting. I have found that the plants are much more apt to be 
distressed by the chemicals when the latter are mixed with the 
earth in the beginning, than when they are applied day by day in 
the form of highly dilute solutions. Indeed I have rarely seen 
my plants suffer from the chemicals with which they were watered 
excepting in some extreme instances, where instead of a true soil 
of fit capillary character the plants were made to grow in mate- 
rials, such for example as coarse fragments of bone-black, anthra- 
cite cinders or clippings of mica, that were lumpy as well as poor 
and inert in the chemical sense. Naturally enough, it sometimes 
happened that plants were seen to be distressed by chemicals in pre- 
liminary trials that were made to test the availability of some par- 
ticular solution rather than to study the character of the soil in 
which the plants were standing. It would. be easy in any event, in 
case the growing crops began to be distressed by the chemicals, to 
substitute rain-water for the chemical solutions, for as many days as 
might seem fit. Or if danger of this sort were anticipated, a small 
measured quantity of the chemical solution might be given the 
plants once a day or every other day, while rain-water was used 
for the watering, properly so called. Practically I have seldom 
found any difficulty in using the solutions of chemicals. The assay 
would be a capital method of research if all its requirements could 
be fulfilled as easily as this one. Of course by measuring the solu- 
tions it is easy to determine just how much of each chemical sub- 
stance has been used in each of the jars. Definite quantities of 
food might readily have been given, in this way, to the plants on 
the soils under examination, if there had been the least encourage- 
ment to do so. 
I have found that the pure white sand from Berkshire County, 
Mass., which is obtainable at the glass works of this vicinity, is 
tolerably well adapted for my experiments; and have satisfied 
myself by repeated trials that rain-water collected from the slated 
roof of a building, which like the school building of the Bussey 
Institution stands by itself in the country, is abundantly pure 
enough for the purposes of the assay. The water in question, 
which is stored in a large covered cemented brick cistern from 
which only insignificant quantities are taken for use, has proved 
