HARDWICKKS SCIENCE-GOSSIP. 



243 



Further, the instructive preference must be 

 hereditary ; for what could a single generation of 

 bees do? The life-time of a bee is short — in 

 many cases only a single season — it has scarcely time 

 even to get educated in the matter of colour, much 

 less to effect any alteration in the race of flowers. In 

 fact, if the preference were not hereditary the tastes 

 of the bee would vary from generation to generation, 

 and one would undo the work of the other. 



I think Hermann Midler's conclusions on this 

 point would have been more in place as an appendage 

 to my own paper, than in a professed criticism of the 

 same. 



And it is strange after such a statement, and the 

 previous admissions that coloured flowers are not 

 developed, but only "stereotyped and perpetuated" 

 by insect selection, and that the bee's taste is simply 

 a matter of education, to read Mr. Tansley's words : 



"I think, in fact, we may safely conclude that 

 .... red and blue, appearing as they do in flowers 

 highly developed in other respects, were evolved 

 through the selective action of long-tongued insects 

 such as bees and diurnal Lepidoptera." 



I cannot make out from what premises this remark- 

 able conclusion is drawn : it remains an inscrutable 

 mystery how the facts and arguments of the paper 

 itself are supposed to oppose my views. 



I am glad to find that Mr. Tansley agrees with me 

 in thinking that the red hawthorn ought, according 

 to the bee-selection theory, to attract bees more than 

 the white one. As far as I can make out, it does not, 

 and therefore goes against the theory. 



The word " probably " used in my former paper, I 

 admit makes the argument from the red hawthorn a 

 poor one : if it can be shown that bees visit it more 

 frequently than the white, the particular argument 

 fails altogether : as far as I know, this has not been 

 done. 



With regard to the concluding paragraph of Mr. 

 Tansley's criticism, it is doubtless unwise to argue 

 from the existence in flowers of certain colours which 

 cannot have been influenced in their development by 

 insect selection, that the colours of flowers in general 

 cannot have been so developed : it is, however, a 

 legitimate and necessary conclusion, that they may 

 have been developed without the selective action. 



Carefully considered in connection with the general 

 principles of natural selection, I think a very strong 

 argument can be deduced from the facts in question. 



These facts are, that many white umbelliferous 

 flowers are pink before opening, and that other 

 flowers — hawthorn, Christmas rose, white evening 

 primrose — are so when beginning to fade. 



Suppose for a moment that chance variations such 

 as these occur in certain races of white flowers, 

 when as yet red and blue are not. What advantage 

 will the flowers obtain in the struggle for existence ? 

 Suppose that bees are attracted by the red, and prefer 

 it to the white. In the former case, they visit flowers 



not sufficiently advanced to receive any benefit in the 

 form of cross-fertilisation, and in the latter, those 

 which are too advanced for it — which are already 

 fertilised, and probably rifled of their honey, 

 altogether or in part. The reddish flowers obtain no 

 advantage from their colour : the bee obtains no 

 honey from the unopened flower, and very little — if 

 any — from the faded one. 



I am assuming here — I know not with what 

 authority, but the reddish tints in question do seem 

 akin to the bluish shade which began to appear in 

 Mr. Grant Allen's developing monk's-hood — that such 

 colours in white flowers are the chance variations out 

 of which red flowers are supposed to have been 

 evolved by selection. As I have shown, neither bee 

 nor flower obtains any benefit, and no development 

 of colour can be supposed to take place. 



This is a strong argument, but it becomes stronger 

 when taken in connection with the educational theory 

 brought forward by Mr. Tansley. 



The bee, according to this theory, has been 

 educated by experience to go to red flowers : it is not 

 attracted by red as a colour. Why, then, should it 

 pick out pink buds, and fading flowers, yielding no 

 honey, or very little ? And if it did, consider the 

 educational effect. Will not the bee learn to distrust, 

 and avoid red flowers ? 



But as I understand Mr. Tansley, he does not 

 believe that the bee carefully selected those flowers 

 showing a slight tendency to redness, and by 

 accumulating such variations evolved the fully- 

 coloured blossom : he rather seems to insinuate, that 

 the flower may have been as perfectly coloured as the 

 red leaf before the bee's action came into play — 

 the bee "stereotyped" the flower, but not the 

 leaf. 



It is to be remarked that this theory, equally with 

 that of development, requires a fixed taste, and a 

 constant exercise of the same, on the part of the bee : 

 it must be carefully examined in connection with the 

 colours of the flowers receiving the greatest number 

 of bee-visits. 



NOTES ON FLYING-FISH. 

 By Surgeon G. D. Trevor-Roper, R.N. 



TRAVELLERS along the paths of Nature ever 

 find fresh food for wonderment, combined 

 with a feverish anxiety to know the why and where- 

 fore. Whether it be the history of the past, the 

 levelling of mountains, and the upheaval of continents, 

 with all their accompanying phenomena, or the study 

 of things existing both animate and inanimate, they 

 all give ample scope for reflection, thought, specula- 

 tion, and the building up of golden theories too often 

 to be dashed to the ground by a few moments' sober 

 thought. All of us have heard of, and perhaps 



M 2 



