1885.] 
Flowers and Insects. 
545 
served 20 genera, with 77 species. The single genera and 
species are characterised by their distinctive features, and 
especially by those structural features which come into play 
in the collection of pollen and honey, and brought into con- 
nection with their activity in different forms of flowers. In 
connection therewith there is given for each species a list of 
its flower-visits founded on Muller’s arrangement of flowers, 
-as wind-flowers (anemophilous), pollen-flowers, flowers 
with their honey open, partially concealed, or fully concealed, 
flower associations, fly., bee-, and butterfly-flowers. 
First in the series of flower-visitants ranks the hive-bee 
Apis mellifca. Its trunk, neither too long nor too short, 
affords the best conceivable equipment, by the aid of which 
it sui passes all its rivals in the number of flowers which it 
can utilise. It is the most skilful of all flower-visitants, 
being able not merely to distinguish the exceedingly mani- 
fold arrangements of flowers, but to utilise them most 
advantageously by modifying its procedures. Of the flower- 
visits observed by Muller in Northern and Central Germany 
36-3 per thousand were effected by the honey-bee. Low’s 
observations give a proportion three times greater, — a result 
due to the peculiar locality. 
Nevertheless, as regards the seleftion of flowers, there is 
little difference between these two authorities. Muller’s 
observation, that any class of flowers is visited the more 
rarely the opener its store of honey, was fully confirmed. 
These and other facts prove that all doubts as to the trust- 
worthiness of the statistical method must be considered 
unfounded. 
An especially striking fact established by Low is, that of 
the nine kinds of Bombus which he has observed, and which 
may be arranged according to the length of their trunks in 
an ascending series from Bombus terrestris to B. hortorum, 
show a similarly progressive preference for “ bee-flowers.” 
The following facts, advanced by Muller in support of his 
theory of flowers, have been fully confirmed by Low in his 
limited field of observation : — 
1. The more openly a class of flowers presents its honey 
the more it is visited by species with short trunks ; 
the more deeply the honey is concealed the more it is 
frequented by visitors with long trunks. 
2. The more a flower-visitant is furnished with a short or 
a long trunk, the more decidedly it prefers flowers 
with open, or respectively with deeply-concealed, 
honey. In visitants with long trunks, however, in 
