242 



HARD WICKE' S SCIENCE- G OSSIP. 



white grow together, I can distinguish no preference 

 for any of these colours. 



One more example of the habits of bees in connec- 

 tion with different coloured flowers of the same 

 species, and I will spare my readers — who have not 

 already spared themselves. 



It is the transcript of a leaf from my note-book. 



Here is a patch of the Sun cistits, or rock rose 

 {Helianthemum vulgare), with red, yellow, and white 

 flowers. 



Three yellow blossoms are separated from five white 

 ones by one red ; to the right are a large number of 

 red flowers. 



Bee No. I visits all three yellow flowers, the red 

 one and four of the white ; then pays flying visits to 

 one or two of the yellow, and goes off to the 

 numerous red. 



Bee No. 2 visits all five of the white, two of the 

 yellow, goes back to the white and revisits four of 

 them ; finally it goes to the red. 



I have since frequently observed the bees on the 

 same tri-coloured bed, and have seen them pass from 

 colour to colour in ever}' order possible on the theory 

 of permutations. No bee that I have observed for any 

 length of time on these particular flowers kept to one 

 colour. 



Mr. Tansley asks, " Why bees so greatly prefer 

 honey on blue paper, if not because they prefer the 

 colour blue in flowers ? " I think, on the whole, I 

 shall leave Sir John Lubbock to explain the results of 

 his own experiments. 



But, whatever the interpretation may be, they do 

 not prove that bees visit blue and red flowers more 

 frequently than others, and this is the point. If the 

 experiments are inconsistent with observed facts, so 

 much the worse for the experiments —or the facts. 



The statements of Messrs. Grant Allen and 

 Hermann Miiller as to the colours, etc., of the most 

 advanced flowers, touch upon a subject about which 

 it is hardly possible to be otherwise than dogmatic. 

 As Mr. Darwin says, " It is hardly possible to define 

 clearly what is meant by the organism being higher or 

 lower." I will not, therefore, enter into any discussion 

 as to what the characteristics of the most advanced 

 flowers are ; I will simply bring forward a few 

 examples from our native plants, which seem to me 

 to show, that the statements referred to are weighted 

 with too many and important exceptions to rise to the 

 dignity of rules. 



Take the orchids, which Mr. Grant Allen declares 

 to be " by far the highest of the trinary flowers, if not 

 indeed of the entire vegetable world." 



Among these we find some sixteen described as 

 green, greenish- white or yellow ; thirteen as purple ; 

 four as brown ; two as brown and purple. Not a 

 single well-marked blue occurs among them. In the 

 order Boraginaceae, on the other hand, which possesses 

 the lowly mark of symmetry in its flowers, we have 

 a great preponderance of blue. Among the trinary 



tribe we have the intensely blue Scilla bifolia, and the 

 wild hyacinth, both simple as to the form of their 

 flowers. 



The veronicas are perhaps the simplest of the Scro- 

 phulariacece, and they are blue. The iris is a 

 complex flower, and our commonest species is yellow. 



The common flax is a simple flower, and yet blue ? 

 the campanulas are blue, and yet comparatively 

 simple. As a last example, take the Compositae. 

 The highest of the three divisions into which it is 

 usually divided is considered by Mr. Grant Allen to 

 be that in which the corollas are all ligulate. These 

 are, dandelion, hawkweeds, etc. — nearly all yellow : 

 the lowest division, in which the corollas are all 

 tubular, contains many purple flowers. 



But I am not at all satisfied that Miiller intends, in 

 the words quoted by Mr. Tansley, to assert that the 

 most advanced flowers are usually red, violet or blue. 

 He says we find these colours " appearing for the first 

 time in flowers whose honey is quite concealed " ; but 

 there is nothing to indicate that he considers these 

 the most advanced. Now, as a matter of fact, there 

 are some simple flowers — winter aconite and Christ- 

 mas rose, for example — in which the honey is quite 

 concealed. 



Further, it is to be noted, that the second part of 

 Muller's statement is to the effect, that blue appears 

 likewise in simple flowers (Ifcfiatica, IWbascuvi). The 

 accordance with Mr. Grant Allen's views is only 

 apparent if we accept concealment of honey as the 

 mark of advance, and omit this second part. 



And supposing we allow to the combined statements 

 of Grant Allen and Hermann Muller the full force 

 which Mr. Tansley seems inclined to assign to them, 

 they do not in themselves affect the argument. If all 

 highly advanced flowers could be shown to be blue, 

 there would not even be a theoretical possibility that 

 they could have been evolved by the selective action of 

 bees, unless it could also be shown that bees have — 

 or once had — a sufficiently strong preference for blue 

 to cause them to pick out blue flowers for their visits. 



Muller's statement, that bees "are not confined by 

 hereditary instincts to certain flowers, but fly about 

 seeking their food on whatever flowers they can find 

 it," fully confirms my own observations and conclu- 

 sions : it completely annihilates the whole Darwinian 

 theory of the development of flowers by the selective 

 agency of insects. It follows as a necessary and 

 logical consequence from it that, however the form 

 and colour of flowers have been produced, it has not 

 been by the selective" action of bees. The theory 

 absolutely requires that the insects should select 

 certain peculiarities of form and colour, and confer 

 upon them the benefit of cross-fertilisation, so that 

 the flowers so chosen may live down the ancestral 

 form, and the less improved varieties ; and the selec- 

 tion must be continuously carried out for very many 

 generations of bees before any appreciable advance 

 can take place in the flower. 



