Colour in Flowers. 
53 
by which the visitors are attracted. 1 His earlier papers were 
criticised at the time by Kienitz-Gerloff and also by the well-known 
Swiss entomologist Forel, who repeated some of Plateau’s experiments 
and arrived at quite other conclusions. In the last number of the 
Beihefte zum Botanischen Centralblatt, Bd. XV., Heft. 3, Eugen 
Andreae publishes an able paper in which he not only shows very 
clearly that Plateau’s conclusions are ill-founded, but also records 
the results of a number of experiments of his own, which distinctly 
advance the whole subject. Andreae’s observations and experiments 
were made in the Botanic Gardens at Jena, under the direction of 
Professor Stahl, at Cadenabbia on the Lake of Como, and in 
Corsica, both on wild flowers and in gardens. He found, however, 
that the garden work was much the more satisfactory of the two, 
since wild flowers are seldom sufficiently closely massed to enable 
the observer to get decisive results. 
The main result which emerges from Andreae’s numerous and 
varied experiments is precisely the enormously greater importance 
of colour compared with scent, in attracting the higher anthophilous 
bees, such as Bombus, Anthophora , Osmia (especially the hive- 
bee Apis mellifica ; and also butterflies at a distance. The author 
used various methods of contrasting the effect of these two factors. 
Thus he covered a bunch of flowers belonging to a species he was 
experimenting upon, with an inverted bell-jar, so that the flowers 
could be seen while their scent could not escape, while a second 
bunch he covered with dark paper, open at the top, thus hiding the 
flowers from the eyes of insects but allowing their odour to rise 
into the air. In other cases he worked with artificial flowers, and 
often obtained the most striking results in inducing insects to 
visit them in preference to hidden, but strongly scented, natural 
flowers, which latter, when exposed, they were visiting freely. 
On the other hand some of the bees less highly adapted to the 
anthophilous habit, such as Anthrena and Prosopis, were often 
clearly attracted by scent at short distances, particularly in an 
experiment with Reseda —where the flowers were hidden in a gauze 
net—and did not appear to be guided by sight. In the case of 
night-flying insects, such as the crepuscular Convolvulus Hawk 
Moth, visiting a species of Crinum , and of the flies visiting a species 
of Arum, it was also shown that the odour of the flowers visited 
was the dominating factor, though these insects were also guided 
to some extent by the sense of sight. 
1 In one of liis last papers, Plateau rather modifies this con¬ 
clusion and attributes a “secondary action ” to vision. 
