, 
PART III. | THE MECHANISMS OF FLOWERS. | 537 
the anthers empty. In the space of two hours we had three more 
opportunities of watching the fertilisation of Orchis mascula ; the 
fertilising agent on two of these occasions was Bombus lapidarius, 
on the third, Psithyrus campestris. B. lapidarius spent a shorter 
time on each flower (about two to three seconds), but otherwise the 
various bees behaved precisely alike. We caught the Psithyrus 
and a B. lapidarius, and found a number of pollinia on the head of 
each. Some of the pollinia had already bent forwards, so that 
they were ready to come in contact with the stigma of the next 
flower; others were still erect. Of ninety-seven humble-bees 
which we caught that day on Stromberg Hill, thirty-two bore 
pollen-masses of Orchis. We frequently observed on our captured 
bees, that when the pollen-masses bent forwards the bee was able 
to tear them off with its mandibles. Some bees which we caught 
with pollinia on their heads had them attached to their forelegs 
when examined shortly afterwards. These frequently successful 
efforts on the part of the bees to free themselves from the pol- 
linia explain why we often find whole pollinia or pairs of pollinia 
attached to the flowers, generally in the neighbourhood of the 
stigma. 
On this one day, which was especially favourable, as it was warm 
and calm, and came after a long spell of cold weather, probably 
more than a third of all the humble-bees on Stromberg Hill had 
been engaged in fertilising Orchids. The following numbers give 
a rough idea of the scale on which the work was performed. At 
seven o'clock on the morning of the same day I gathered ten 
specimens of Orchis morio on a meadow covered with the plant. 
These ten spikes contained 107 expanded flowers; of these, one only. 
had its anther-loculi empty and its stigma smeared with pollen; 
two had their stigmas smeared with pollen, but their pollinia had 
not been removed. At five o’clock in the afternoon I gathered in 
the same place ten spikes containing ninety-seven flowers ; fourteen 
had their stigmas smeared with pollen, and of these two only had 
their pollinia still within the anther-loculi; three other flowers 
had their pollinia removed, but their stigmas showed no trace 
of pollen. Thus at seven o'clock in the morning 2°5 per cent., 
and at five o’clock in the afternoon over 14 per cent. of the flowers 
had been fertilised. 
I need hardly add that this observation fully confirmed Darwin’s 
suggestion. The bees must have found something within the 
spurs, or they would not have continued to visit the flowers. Since 
the spur contains no free honey, but abundant fluid within its 
~ 
