374 NATURAL SCIENCE [June 



ccuclusioiis of Prof. Plateau on the relations between insects and 

 flowers. Under the title " Comment les tieurs attirent les insectes " 

 (see Bull. Acad. BruxeUes, 1895, 1896, 1897), Prof. Plateau 

 described numerous experiments which in his opinion prove that 

 insects are not attracted by the form or colour of flowers, but by 

 their scent. Sir John argues that these experiments were incom- 

 plete, and would, moreover, often admit of conclusions exactly 

 opposite to those drawn. For instance, single dahlias were used in 

 several cases. Squares of coloured paper with a central hole were 

 pinned over the head of flowers so as to hide the ray-florets, which 

 are generally considered to form the attractive portion, while 

 revealing the yellow honey-containing centre. Prof. Plateau 

 •did not find that the bees neglected the partly protected heads. 

 But, as Sir John points out, the paper discs, of red, violet, white, or 

 black, must have been very conspicuous, and the insects were, 

 moreover, used to visiting the bed, and could, therefore, soon find 

 the not inconspicuous yellow hearts of the partly covered 

 flowers. When the whole heart was covered with the paper or 

 with green leaves, and even when all the flowers were masked, the 

 insects still came, which. Prof. Plateau says, proves conclusively that 

 they were attracted by the smell. But against this it is suggested 

 that, first, tlie flowers were only covered above, not completely hidden; 

 secondly, that a bee which has got used to visiting a spot for honey 

 will hunt about for the honey when it is hidden ; and thirdly, that 

 as dahlias have, so far as we can perceive, no scent, " it is some- 

 what illogical to assume that the bees are guided by the sense of 

 smell, when we have no evidence that any scent is emitted." 

 Even if it were proved in his experiments that the insects were 

 attracted by smell, which undoubtedly they are in many cases, and 

 that they did not see the ray-florets, this would not prove that they 

 are not guided by the colour petals when they can see them. In 

 reply. Sir John adduces some experiments of his own to test the 

 respective attractions of scent and colour. A hive-bee was trained 

 to come to a certain spot on a table for honey. The honey was 

 then removed and on one side of the spot was placed a drop of 

 honey on a glass slide, and the great brilliantly blue bracts plucked 

 from a flower-head of Eryngium amethystmum ; at an equal distance 

 •(one foot) on the other side was placed a similar drop of honey, 

 and a far less conspicuous flower-head, which had been deprived of 

 its bracts. During three successive days the visits of the bee were 

 noted, and it was found that of ninety-three visits, it came sixty 

 times to the honey near the bracts, and thirty-three to that near the 

 flower-head. After each visit the head and bracts were transposed 

 to eliminate any possible difference in the two samples of honey. 

 These experiments go far to disprove Prof. Plateau's conclusion 



