FLORIDA STATE HORTICULTURAL SOCIETY. 63 
stimulate the fruit to development even if pollination did not 
take place. I thus in my experimenting opened a number of 
navel flowers last spring and summer (1893) before the pollen 
or pistil had matured, emasculated them (i. e. cut off the 
stamens—castrated) and immediately drew over the flowers 
thus treated firm paper or close cloth bags, and tied them 
around the branch below the flowers, so that all insects were 
excluded. A number of the flowers thus treated, matured 
fruits which were to all appearances perfectly normal, though 
developed we can almost positively say without the access of 
pollen. 
Again this spring (1894) a number of flowers were similarly 
treated and several fruits have set and a r e to all appearances 
developing normally, though all pollen was excluded, and thus 
there was no fecundation. 
The navel fruits developed last year from emasculated 
flowers on examina’ ion were all found to be perfectly seedless 
with the exception of a few small rudimentary seeds. Though 
my experiments have not yet been extensive enough to be 
conclusive it yet seems from the results obtained that the 
navel fruits possess the faculty of developing without the 
action of pollination and fecundation. 
In my experiments, further, a number of navel flowers were 
crossed with pollen from various other varieties of Citrus 
fruits. The most notable feature developed in this series of 
experiments was that in all crossed fruits numerous seeds 
were produced, while in the fruits produced on the trees 
normally very few seeds were formed, and in almost all cases 
none. This commonly seedless character of the nav. 1 fruits 
is, as every one knows, the feature which most recommends 
this variety of orange. 
In every case where navel fruits developed from crosses, I 
early marked, near by on the same tree young fruits of ap¬ 
parently the samm age, and in every way comparable to the 
crossed fruits other than that the}'’ were developed without 
any treatment whatever, as they always naturally do in the 
groves. They were of couise entirely open and may have 
been pollinated by insects < r may have been developed with¬ 
out pollination. However, this may have been, the fact re¬ 
mains the same that in every truit thus early marked as checks 
for comparison, not a single seed was found developed nor¬ 
mally. In two or three instances rudimentary seeds about 
two millimetres long were found. 
Many of the fruits from the same trees, developed normally, 
were examined and resulted in the finding of but few fully 
developed seeds. Occasionally, however, in Iruits developed 
normally, one or two fully matured, plump seeds were found. 
5 
