Botantcal Society of Edinburgh. 323 
100 and 200 flowers, yet every capsule aborted. I dissected the column 
of many of these flowers as they dropped off, and invariably observed an 
abundance of pollen-tubes, which in most cases I traced into the ovary. 
Thus if the function of the stigma be simply to excite the emission of 
pollen-tubes, and that of the style their conduction to the ovary, we have 
evidence of its accomplishment, and might then attribute abortion to 
some inappreciable change in the sexual elements preventing normal 
conjunction. 
“‘ Having thus failed in fertilising O. sphacelatum with its own pollen. 
I determined to try crosses with other species. With this end in view, I 
first crossed it reciprocally with the neighbouring species, O. altessimum, 
under the impression that the probabilities for successful results would 
be inversely proportionate with the more or less immediate systematic 
affinities of the plants. In this my experiments have somewhat disap- 
pointed me; and the results of the above were the unexceptional abortion 
of every capsule. I next tried O. graminifolium with the pollen of 
O. sphacelatum, and succeeded in producing a capsule, which contained 
about one-fourth of embryonated seeds. I did not succeed vice versa. All 
the flowers thus treated dropped early. I may state that O. gramini- 
foliwm does not appear from my experiments to be very susceptible of 
fertilisation with its own pollen. 
‘From four flowers of O. ornithorynchum, impregnated with the 
pollen of O. sphacelatum, I got one fine seed-capsule, though I had never 
before succeeded in fertilising this species with its own pollen; I have 
now a single capsule thus produced. On dissection of the crossed capsule, 
I was disappointed to find that it contained few seeds, and of these a 
great majority presented only a loose transparent testa, entirely destitute 
of the embryo. The capsule, on the other hand, fertilised with its own 
pollen, though not quite so large as the above, was quite filled with seeds, 
of which about three-fourths presented an embryo. I did not succeed in 
impregnating O. sphacelatum with pollen of O. ornithorynchum, though 
the capsules thus treated at first showed symptoms of swelling. 
*‘'The most successful experiments which 1 have made were with 
O. divaricatum var. cupreum, and O. sphacelatum. I impregnated 
six flowers upon the latter with pollen from the former; and I have 
now four fine capsules, nearly mature, as the result. I failed, however, 
in my attempts at crossing them reciprocally. I made also numerous 
attempts to impregnate O. divaricatum with its own pollen, and this as 
well upon those plants growing in baskets suspended in the hot-houses as 
those growing in pots. On one of the plants in the latter condition I 
have now, after many failures, succeeded in fertilising four flowers; the 
capsules, however, are very abnormal-iooking productions. In one of 
them which I cut off for dissection, I did not find a single perfect seed. 
On the other hand, the results of experiments on plants growing in 
baskets, though the more natural mode of cultivating them, was, singu- 
larly enough, the abortion of every capsule thus impregnated. ‘The only 
indications of pollinic influence were an earlier fading of the flowers, and 
the closing of their stigmatic orifices, which I may remark is in this case 
effected by a gradual depression of the clinandrium, instead of a simple 
incurvation of the wings of the orifice, as occurs in O. sphacelatum, men- 
tioned above. 
‘‘ Thus, in the above experiments, I have failed in crossing recipro- 
cally any two species; nevertheless, I think I have satisfactorily shown 
an individual impotence of the sexual organs, in their mutual action, con- 
joined with a capability of normally performing their functions by the 
action of other species—and this in comparative disregard of recognised 
