HOW FLOWERS ATTRACT INSECTS 205 



Kienitz-Gerloff rightly makes the following remarks on this (Bot. Ztg., Ixvi, 

 1896, pp. 123 and 124), 'Both premisses and conclusions are equally open to 

 attack. For of course the covered dahlia heads could still attract animals by their 

 odour though this might not be perceptible to human beings and to infer from 

 this that the colour of uncovered flowers plays no part in attracting insects is the less 

 justifiable as Plateau gives absolutely no comparative figures with regard to such 

 visits, but only makes the very indefinite statement that insects flew to the covered 

 flowers in the same way as to the uncovered, without hesitation and with equal 

 eagerness.' 



In the introductory words to the second part of his memoir, Plateau gives 

 prominence to the fact that his results being so diametrically opposed, to existing 

 views he had continued his experiments on the question how flowers attract insects 

 partly in his own garden, partly in the country, and partly in the Botanic Gardens at 

 Ghent, where he had both repeated the experiments of other investigators and made 

 entirely new ones. 



Of all Plateau's researches those communicated in this second part appear to 

 me to be the most important. In these he made use of flowers which had been 

 rendered very inconspicuous by removal 0/ the petals, or of the coloured part of the 

 corolla, but which nevertheless received a very considerable number of insect visits. 

 But before considering these researches in detail I should like first to deal with the 

 other and less important experiments of this investigator. 



Plateau first repeated the experiments on dahlias with Heracleum Fischerii 

 (Umbelliferae), by covering its umbels with rhubarb leaves. Yet within thirty 

 minutes he observed three visits from Apis mellifica var. ligustica, two from other 

 small bees, one from Calliphora vomitoria, and one from Phyllopertha horticola, 

 followed in a further period of an hour and a half by twenty-five from Odynerus 

 quadratus, ten from Prosopis communis, three from Calliphora vomitoria, and one 

 from Musca domestica. 



In my opinion this only proves that the insects named are also attracted by 

 odour, which has been disputed by none of the recent oecologists. No proof has 

 been aff"orded that the attraction is only by odour, for no control experiments were 

 made with uncovered umbels. From the fact that many insects (Apis, Andrena sp., 

 Bombus sp., Megachile ericetorum, Pieris napi, Vanessa C. album, Eristalis, and the 

 smaller Syrphidae) showed themselves indifferent as regards the various colours of 

 varieties of the , same species, or of the species of the same genus, visiting without 

 displaying preference the blue, white, purple, and rose-red flowers of Centaurea 

 Cyanus, the red, purple, rose-red, orange, and white capitula of Dahlia variabilis, the 

 purple, rose-red, and white capitula of Scabiosa atropurpurea, the red flowers of 

 Linum grandiflorum and the blue ones of L. usitatissimum, Plateau concludes that the 

 colours of flowers cannot play any part in the attraction of insects. He also cites 

 the similar observations of other investigators, e.g. Darwin saw a humble-bee pass 

 from a red Dictamnus Fraxinella to a white-flowered one, and another betake itself from 

 one variety of Delphinium Consolida to another that was differently coloured, while 

 much the same thing was noticed by Gaston Bonnier for the colour varieties of 

 Althaea rosea, Digitalis purpurea, and Brassica oleracea, as well as by Errera and 

 Gevaert for species of Pentstemon. 



