METHODS OF RESEARCH IN FLOWER POLLINATION 201 



After some experience it is possible without apparatus to catch many insects 

 during their visits to flowers, for these guests are usually so busily employed in 

 consuming the nourishment offered them, that they can be taken from the flowers 

 by means of the fingers. Even the Syrphidae, which even with the help of a net are 

 very difficult to catch when on the wing, may easily be taken on flowers^. Other 

 insects such as the Muscidae as already mentioned are very shy, so that they are 

 secured with difficulty. 



An observer should never omit to draw in their various stages of development 

 the flowers of the inflorescences with which he has occasion to deal. Should his 

 first few attempts fall short of complete success, practice will soon give him a 

 sufficient amount of skill. It has always been very interesting to me to compare the 

 first almost clumsy drawings of Hermann Miiller with the later ones executed by him, 

 which must be regarded as works of art. We may compare, for example, the almost 

 diagrammatic outlines in his first work (' Fertilisation'), on p. loi (Nasturtium sylvestre 

 R, Br.), p. 107 (Teesdalia nudicaulis R. Br.), p. 132 (Cerastium arvense L), p. 256 



Fig. 78. (i) Nasturtium sylvestre, R. Br. (ixova. Hermann Muller, ' Fertilisation," p. loi). (2) Lont- 

 cera nigra, L. (from Hermann Muller, ' Alpenblumen,' p. 394). 



(Lythrum Salicaria L), p. 426 (Lycium barbarum Z.), p. 386 (Hottonia palustris Z.), 

 p. 300 (Galium Mollugo Z.), and so on, tvith the finely executed and beautiful 

 illustrations in his second great work ('Alpenblumen'), e.g. on p. 394 (Lonicera 

 nigra Z.), p. 395 (Lonicera alpigena Z.), p. 406 (Phyteuma), p. 470 (Valeriana 

 montana Z.), and so forth (see Fig. 78). 



I have tried to photograph inflorescences, and this kind of reproduction has the 

 advantage over a drawing that it gives a convincing picture which is true to nature. 

 But an observer cannot always have photographic apparatus with him, nor can 

 he everywhere perform the necessary operations. Besides which there are so many 



Knuth, ' tjber bluteabiologische Beobachtungen,' Heimat, Kiel, iii, 1893, Part 5-6, pp. 8-9. 



