RELATED STUDIES AND CRITIQUES. 169 



was removed, he emphasized the fact that the mutilated flowers were always 

 near normal ones and the insects attracted by the latter were guided by 

 odor to the former. Moreover, control numbers were lacking in most cases 

 and in that of Lobelia were mostly against the assumption. The fact that 

 a honey-bee flew to faded flowers and buds, and even to fallen petals, gave 

 evidence of the effect of color, the fact that it took nectar only from the 

 mutilated flowers being explained perhaps by the greater odor and accessi- 

 bility. It was further remarked that the results with flowers of different 

 colors of the same species indicated just the opposite conclusion to that 

 drawn, as shown especially in the case of Centaurea. The small number of 

 experiments were regarded as having little weight in comparison with the 

 observations of Mueller on 5,674 visits by 841 species of insects to 422 

 alpine flowers. Whether insects see the various colors as we do or not is 

 immaterial, if in general they distinguish them (1897:84, 108). 



Plateau was further criticized for not being familiar with Mueller's 

 results and for opposing a single observation on Listera to the many made by 

 the latter, as well as for not distinguishing between insects of different 

 orders and diverse intelligence, although Mueller had frequently empha- 

 sized the importance of this. Moreover, he should have repeated the experi- 

 ments of Lubbock and of Mueller, which showed that insects possessed a 

 clear perception of color, instead of trusting too much to the evidence 

 drawn from artificial flowers. Likewise, he contradicted himself when he 

 explained the visit of a cabbage butterfly to an artificial flower on a lady's 

 hat by saying that it sought the flower for its protective white color and not 

 for nectar. The critic also found unconvincing the use of artificial flowers 

 made from leaves and provided with honey, owing to the superior attraction 

 of honey, irrespective of the object on which it may be placed. The final 

 conclusion was reached that Plateau had proved nothing against the color 

 sense of insects, although it was admitted that odor played an important 

 r61e, as Mueller had long before insisted. 



Knuth's critique. — The results of Plateau's first series were reviewed 

 by Knuth (1898, 1906:204), who pointed out that they permitted another 

 interpretation than that given. In the case of umbels of Heracleum covered 

 with green leaves, the visits of a variety of insects were regarded as indi- 

 cating that insects are attracted also by odor, but that this was alone 

 effective was not proved, since there was no comparison with the visitors 

 of the umbels not masked. The evidence that color is not attractive, 

 afforded by behavior to the differently colored varieties of the same species, 

 is inconclusive and it can only be concluded that color is immaterial in 

 flowers of the same form. The experiments in which honey was added to 

 bright flowers that are normally little visited merely prove the well-known 

 fact that the odor of honey is a powerful lure. Since both honey-bees and 

 bumble-bees distinguish readily in many cases between flowers visited and 

 unvisited, it is not strange that they distinguished the artificial from the 

 natural flowers in Plateau's tests and left them alone, even when supplied 

 with honey. Neither the paper flowers nor those made from green leaves 

 and provided with honey prove that color is unattractive to insects, but 

 both merely emphasize the powerful attraction exerted by the odor of honey. 



