9 o 
FLOWERS AND INSECTS 
Parnassia, Galium, most Umbelliferae, Hedera, &c. Obser- 
vation shows that the insect visitors to flowers of this 
class are chiefly short-tongued ; bees and Lepidoptera rarely 
visit them 1 . They are exposed to the risk of having the 
honey spoiled by rain. The upward evolution of flower- 
visiting insects seems to have proceeded in the direction of 
longer-and-longer-tongued forms. Their cleverness, too, in 
finding concealed honey, or in dealing with hanging or 
irregular flowers, is largely correlated with the length of their 
tongues. The concealment and protection of the honey, 
by the formation of a tubular structure in the flower, seems 
to have kept pace with the lengthening of the insects’ tongues. 
The next stage is represented in existing flowers by those 
which form the class AB, ‘‘flowers with partially concealed 
honey.” Such are Ranunculus, Sedum, Cruciferae, Fragaria, 
Potentilla, &c. The visitors require tongues of 2 to 6 mm. 
or more in length. Observation shows that the actual 
visitors are mainly those whose tongues are long enough to 
fit the flowers. Bees and Lepidoptera visit flowers of this 
class more often than those of class A, but the bulk of the 
visitors are the short-tongued Hymenoptera, and the longer- 
tongued flies. From a morphological point of view the 
chief interest is to notice the ways in which concealment of 
the honey is effected. In Ranunculus, &c. it is merely by the 
position of the stamens, &c., a method which does not admit 
of much further elaboration. In Cruciferae the sepals stand 
stiffly erect, thus making the lower part of the flower a 
narrow tube, as is well seen in wallflower. In Fragaria, 
Potentilla, &c. we see the beginnings of perigyny of the 
flower. 
Following any of the evolutionary lines we have thus 
indicated, many flowers probably arrived at the further 
condition represented in the existing flowers of class B, 
“with fully concealed honey.” Such are Anemone § Pulsatilla, 
1 The student should spend a few fine afternoons in observing the 
insect visitors to flowers of the various classes. Even if unable to 
recognize the insects he will be able to observe their length of tongue, 
size of body, &c., and compare these observations with those made on 
the depth of tube in the flower and so on. He should also notice the 
numbers of visitors of each kind that visit the flowers and compare the 
proportions with the class of flower ; see below and refer to Annals of 
Bot ., June 1895, and books mentioned above. 
