180 
The flowers of Crescent, not consciously fertilized at all, 
produced and matured some fruit that bore apparently 
normal seeds. While open, these flowers were several 
times examined, and not a single anther was found. The 
plants bearing them were kept within the same frame as 
the others. Vv hether the polien was disseminated into the 
air from the brushes used in fertilizing the other plants, and 
so fell upon these flowers in sufficient quantity to effect fer- 
tilization, or whether the flowers really did contain a few 
fertile anthers that escaped observation, I do not know. It 
is clear that in an experiment of this kind, the plants to be 
differently treated should not be inclosed within the same 
frame, and if possible, not within the same room. 
9nd. The Grape. 
If foreign pollen exerts an influence upon the character 
of the fruit in the grape, it would be expected that the 
blossoms of black varieties fertilized with pollen from white 
ones, would produce berries of lighter color than would 
self-fertilized blooms. 
The flowers of several bunches of the Burnet, Monroe and 
Senasque, all black grapes, were castrated before the caps 
had fallen, and inclosed in paper bags. In due time they 
were fertilized with pollen from the Lady Washington, a 
white grape, and again inclosed in bags, in company with 
several other bunches on the same vines that had not yet 
opened their flowers. Thus some of the bunches on each 
vine must have been self-fertilized and others  cross- 
fertilized with the Lady Washington. When the 
grapes had matured, it was quite impossible to de- 
tect any difference either in color or flavor between 
the self-fertilized and the cross-fertilized berries of the same 
variety. 
From these experiments and the one made in 1885, it 
would appear that the cases, if any, in which the pollen of 
one variety of strawberry or grape has an influence upon 
the fruit of another, must be regarded as exceptional. 
AN ATTEMPT TO SECURE A RECORD OF THE SOIL MOISTURE. 
We know of no reliable method for securing an accurate 
record of the degree of moisture contained by the soil. The 
- operations of the laws that govern the movements of water 
in the soil, though of vast importance to agriculture, are 
not well understood. Hence an investigation has been 
undertaken during the past two seasons to discover a way 
whereby the moisture of the soil can be measured and 
recorded. As an aid to the discussion, experiments that 
gave only negative results are briefly alluded to, with the 
reasons for their failure. 
