Flowers 
In gardens the plants are unnaturally placed and 
unnaturally visited, so that observations, which can be 
very comfortably made in a garden, are unfortunately 
of no value whatever, 
The close connection between the insect-world and 
the flowers which they visit is not at once manifest. 
The two have varied together, the first stupid sort of 
insect was surely as different from a bumble-bee or a 
butterfly as the clumsy, yellowish and ugly flowers of a 
Cycadofilix differ from those of an orchid or a salvia. 
When one remembers this fact, much that is not at 
once obvious becomes clear and manifest. For in- 
stance, there is a distinct change in the various sorts of 
flower. In spring, the larger proportion are of the open 
type in which honey is easily obtained. The majority 
of our richly coloured, complex flowers with concealed 
honey appear in summer, whilst those with honey half- 
concealed and accessible to short-lipped insects, are 
mostly autumn flowering. 
At least this is what has been recorded as the result 
of two large series of observations taken in quite different 
places, viz., Robertson, Carlinsville (39° 31’ N. lat.), in 
America, and Brandenburg in Germany.” This general 
correspondence suits exactly with what is known of the 
insect-world. The majority of insects with long pro- 
boscids occur in summer, those with medium-sized in 
autumn, and those with a short proboscis in spring.’ 
We know also from geological records that the great 
classes of flower-insects appeared in the Cretaceous 
period just as suddenly and along with the true flower- 
ing plants. 
But how difficult and complex are the modern rela- 
tions between flower and insect, and how impossible to 
catalogue and classify are the types of visitors ! 
In July and August one finds every sort of flower. 
116 
