HARDWICKE'S SCIENCE-GOSSIP. 



25 



COLOUR DEVELOPMENT IN LEAVES AND FLOWERS. 



By a. G. TANSLEY. 



AM very glad to 

 see (Science- 

 Gossip, October 

 and November) 

 that Mr. Bulman 

 has argued in 

 some detail in sup- 

 port of his rather 

 sweeping charges 

 against the theory 

 of colour develop- 

 ment in flowers 

 by means of insect 

 selection. In my 

 former paper 

 (Science-Gossip, 

 June, 1888), I 

 briefly pointed out 

 a line of defence 

 of the points at- 

 tacked by Mr. Bul- 

 man ; I will now proceed to consider his " Reply." 



In comparing the colours of leaves and flowers, 

 with a view to finding out the causes of their exist- 

 ence, we are, I think,; at once struck by the fact 

 that the bright colours of leaves, which appear 

 in the autumn, are the attributes of a period of 

 decay, depend in point of time on that period, and 

 are therefore presumably most intimately connected 

 with it in causation : this we know on good authority 

 to be the case. When, on the other hand, we turn to 

 the colours of entomophilous perianths, we see that 

 they last during the period from expansion to fertilisa- 

 tion ; that is to say, the period when the organs of the 

 flower are preparing for the exercise of their function. 

 The inference here also is clear ; either the colours are 

 connected in causation with this functional activity, or, 

 whatever the cause of their original appearance, they 

 at present exist in virtue of the exercise of some 

 function of their own. Now it is not h priori likely, 

 from the great variety and complex distribution of 

 these colours, that they can be accounted for as the 

 mere result of the chemical processes connected with 

 No. 290. — February 1889. 



pollen and ovule development ; but, however this 

 may be, it has not, so far as I know, been done or 

 attempted ; we are, therefore, driven to the other 

 alternative, that they have a function of their own. 

 The question now becomes, what is their function ? 

 From observation of the habits of insects in visiting 

 flowers, and from the knowledge of the benefits con- 

 ferred by cross-fertilisation, which we further know 

 these insects are constantly effecting, we are led to 

 believe that the function of the colours of flowers is to 

 attract insects. This leads at once, by the help of 

 the theory of natural selection, to what we may call 

 the general theory of colour development through 

 insect selection. 



It is thus plain that the difference between the 

 colouring pigments of leaves and flowers, which, Dr. 

 Sorby tells us, are often chemically identical, is a 

 difference in functional significance. Mr. Bulman is 

 quite right in saying that "the fact, that the colour 

 is in both cases due to the same chemical changes, 

 does not make the suggestion (that colour in flowers 

 is not developed by insect selection) less probable." 

 I ought not to have said that Dr. Sorby's statement 

 "fully meets" Mr. Bulman's objection, but rather 

 that it is answered by this fact of the difference o 

 functional significance, which was of course the as- 

 sumption upon which my argument was based. Dr. 

 Sorby's statement, however, in one sense, does meet 

 Mr. Bulman's objection: — "We are just as much 

 bound to account for the colours of these (leaves) 

 as of the varied hues of the blossoms," Mr. Bulman 

 says; and Dr. Sorby's statement does account for 

 them — chemically — in the same way as for those of 

 the flowers, but it is no part of his province to point 

 out, what seems fairly obvious, the great difference 

 between them in respect of functional significance. 



Here, in fact, hes the fallacy of Mr. Bulman's 

 original argument, the assumption that, because the 

 colours of leaver have not been developed by means 

 of insect selection, therefore there is no need to sup- 

 pose that those of flowers, which appear, as I have 

 shown, under totally different conditions, and there- 

 fore require, from a physiological point of view, a 



c 



