100 



An E EC C 



H L 



METHODS OF RESEARCH IN FLOWER POLLINATION 199 



periods of time, by still further dividing the insect groups, and by increasing the 

 number of observations. 



The first of MacLeod's graphic tables is reproduced, where each ordinate 

 represents one of the seven flower classes {An, E, EC, C, S, H, L), while abcdefg 

 is the general insect line for the month of June in the Alps— below the limit of trees— 

 and a^yhfin is the Diptera line. 



For complete understanding of the law regulating the visits of insects to flowers 

 the statistical observations that have hitherto been made in flower pollination are not 

 suflScient, and full knowledge of this 

 law will only become possible when 

 numerous investigators take part in 

 the work. Besides which many other 

 related matters require more thorough 

 study, e.g. the determination of the 

 distribution of the sexes in different 

 districts, the mechanisms of many 

 flowers, and so forth. These pro- 

 blems, however, cannot be solved till 

 such investigations have been sys- 

 tematically made in as many small, 

 ■well-defined areas as possible. 



When only occasional observa- 

 tions are made on flower pollination, 

 it will appear as if insect visits were 

 often very rare, even in the case of 

 the larger and more conspicuous 

 flowers. As a matter of fact we cannot 

 calculate on seeing numerous insects 

 visiting flowers in rainy or even dull 

 weather, or in strong wind, and under 

 such circumstances only occasional 

 stragglers will be noticed. Even in 

 warm, quiet weather there is frequently 

 no abundance of flower guests to be 



seen, and it then becomes necessary to patiently stay in one place and not wander 

 aimlessly about. ' One must not be annoyed,' says the great Sprengel (' Entd. Geh.,' 

 p. 23), 'at having to spend a long time near a flowering plant, and at having often 

 to repeat the same observations on any species of flower, for it is not always visited 

 forthwith by the particular insect which is designed to fertilize it.' As a rule patience 

 will be rewarded, the proper insects putting in an appearance at last, even when the 

 plant under observation stands alone. In the study of flower pollination it is neces- 

 sary always and everywhere to be ready at a moment's notice to make observations. 

 It should therefore be a rule never to go out without apparatus for capturing insects, 

 and always to carry about cases for preserving captured specimens. Immediately 

 after the sun has dried up the dews of night from the flowers the insects come forth 

 from their night's quarters to make their visits, and the student of flower pollination 



90 



80 



70 



60 



SO 



'rO 



30 



20 



10 



r 



?"...- 



Fig. 77. Graphic representation of the visits of Diptera 

 to various classes of flowers in the montli of June (after 

 J. MacLeod), reduced one-fourth, abcd^fg^ general insect 

 line for the month of June in the Alps below the limit of 

 trees; a^ySf^r}, Diptera line. [The symbols for the flower 

 classes used in this translation have been substituted for 

 P, A, AB, B, B, H^ .F employed by MacLeo<l.— Tr.) 



