HOW FLOWERS ATTRACT INSECTS 209 
that were uninjured, it follows that the brightly coloured corolla exerts an attrac- 
tive influence. 
It was only in the case of Antirrhinum majus that the mutilated flowers received 
no visitors. After Plateau had cut down a number of such flowers till they were 
only 1 cm. long, the humble-bees (Bombus terrester Z.) which were sucking the 
uninjured blossoms showed no inclination to visit them. Plateau tries to explain 
this by saying that in order to reach the nectar of the mutilated flowers—which are 
open above—by flying up from below, the insects could get no foothold except by 
suddenly converting their ascending movements into descending ones. For this 
reason they preferred to forgo the use of the nectar in the mutilated flowers, and to 
seek out uninjured blossoms in the neighbourhood. 
This explanation does not seem to me to be a happy one, because these insects 
settle on many other flowers from above, without suffering any inconvenience from 
this mode of alighting. It appears to me far more probable that the odour of Anti- 
rrhinum majus ts not sufficient for purposes of attraction, but that here the form and 
colour of the corolla play a leading part in this matter, so that the insects do not visit 
mutilated flowers in which these means of allurement are wanting. 
Plateau conducted a most interesting experiment with Centaurea Cyanus. He 
removed the blue ray-florets—which according to our view merely serve to attract 
and are therefore neuter—leaving only 
the rather inconspicuous disk-florets 
(see Fig. 81). In spite of this many 
bees (Megachile ericetorum) visited the 
cornflowers so treated. Plateau sup- 
poses that the insects were attracted 
by the odour. I believe that this 
occurrence is quite as easily explained 
by assuming that the visitors of intact 
cornflower capitula have become ac- 
quainted with the form of the nectar- 
yielding disk-florets, and are able—from 
a short distance—to recognize them 
after the removal of the ray-florets just 
Fic. 81. Centaurea Cyanus, L. (after 
Plateau). a, Capitulum with ray-florets ; 
as well as before. 6, capitulum without ray-florets. 
Friedrich Dahl, among others, has 
proved that insects are able to distinguish forms accurately at a short distance 
(cf. Zool. Anz., Leipzig, xxii, 1889). Professor Dahl told me of the following 
observation, which can only be explained on the assumption that insects are 
guided by the sense of sight—-The young honey-bees that appear in spring 
settle upon flowers with nectar beyond their reach owing to the depth at which 
it lies, and make vain attempts to suck it. Older bees (that have already 
swarmed), on the other hand, come near such flowers, but turn away without 
settling on them. This must be due to the fact that the older bees having learnt 
that the nectar of these flowers is not within their reach, look at them on getting 
near and realize that there is nothing to be had. Were it the specific odour 
of the flowers that deterred them from visiting they would not need to go so 
DAVIS Pp 
