138 PRINCIPLES AND CONCLUSIONS. 



The view that the male flowers are more visible than the female in 

 diclinic nectariferous flowers was tested by observations of Salix, Ribes, 

 Asparagus, and Bryonia. The number of visitors was so nearly equal that 

 it was stated that in such flowers the bees do not go first to the staminate 

 and then the pistillate, and that the former do not possess a greater vis- 

 ibility. The development of spots and stripes on the corolla was thought 

 not to be correlated with the nectar or the presence of a large corolla, 

 which is independent of the frequent visits of insects. The development 

 of perfume in the flower and that of nectar are not in agreement. The same 

 flower can be visited in several different ways by the same insect, indicating 

 that there is no reciprocal adaptation between them, as is shown also 

 by the fact that insects continue to visit flowers after the corolla has fallen. 

 Insects are able to secure nectar from the flower without effecting fecund- 

 ation, as is especially true of the large number of flowers that are robbed by 

 bees. The visitors to the same species differ in accordance with the var- 

 iations in the amount of nectar. Furthermore, undesirable visitors are not 

 excluded by odor, form, or by the time or place of blooming. In sum- 

 marizing it was stated that it is impossible to admit that all the features 

 of flowers are designed to attract insects in furnishing them nectar, and 

 thus bringing about cross-fertilization. One can not admit that there may 

 be reciprocal adaptation between flowers and insects, since the observed 

 facts do not accord with the imaginary hypotheses. Finally, the modern 

 theory as to the role of the nectary seems to be inadequate. 



It is obvious that Bonnier's conclusions were much more sweeping 

 than his evidence warranted, and hence it is not strange that he should 

 have been severely criticized by Mueller (1880:219), who declared that 

 Bonnier, in his blind and presumptuous endeavor to destroy one of the 

 most comprehensive and best-grounded of theories by childish weapons, 

 had only succeeded in bringing to it new support. In spite of this, however, 

 he deserves much credit for pointing out some of the serious weaknesses 

 of a theory which attempted to explain the most insignificant features 

 of the flower as both cause and effect of insect pollination. 



Color sense of bees. — Lubbock (1882:291) studied the response of 

 honey-bees to color by means of paper slips and glass slides provided with 

 honey. The latter was placed on blue and on orange paper and a bee 

 brought to the former. After she had returned twice the papers were 

 transposed, but she still went to the blue paper, as she did also when 

 the papers were again transposed after three visits to the blue. Similar 

 results were obtained two days later, in which, as the bee returned to 

 the usual place, now occupied by the orange slip, she started to alight 

 and then darted off to the blue. Experiments of the same sort were made 

 at various times, and agreed in showing that bees return to the color to 

 which they have been accustomed. 



To determine whether there was a preference for one color over another, 

 Lubbock made use of microscope slides on which were pasted slips of paper 

 colored respectively blue, green, orange, red, white, and yellow. These 

 were placed on a lawn about a foot apart and on top of each was put a 

 second slide with a drop of honey; with them was also placed a slide of 

 plain glass with honey. After a marked bee, trained to come for honey, 



