6.^,2 



NATURE 



[October 27, 1898 



to cleistogamy, in its ordinary sense it is clear that there is no 

 contradiction between the Knight-Darwin Law and the facts, 

 as Loew has clearly pointed out (" Einfiihrung," p. 144). 



After the passage quoted above, Knuth goes on : "In the 

 place of the one-sided law of the above-named naturalists (of 

 which the general truth remains unproven) Miiller set up a law, 

 proved directly by Darwin's experiments and indirectly by the 

 reproductive arrangements of plants in general, but especially 

 by those of flowering plants. The law, namely, that ' when 

 the offspring of cross-fertilisation come into serious conflict in 

 the struggle for life with the offspring of self-fertilisation, the 

 former (cross-bred) win the day. Only when this contest is 

 absent can self-fertilisation suffice for reproductions for many 

 generations.' " 



I confess that this law is to me unsatisfactory. We ask our- 

 selves " when is the struggle between cross- and self-bred off"- 

 spring 1 absent ? " Clearly when all the offspring are of one 

 kind, i.e. all cross-bred or all self-bred : in a dioecious plant 

 where all offspring are cross-bred, there is no question of self- 

 fertilisation. In a plant with purely cleistogamic flowers, all 

 offspring would be necessarily self-bred, and the law would 

 imply that cleistogamic perpetuation may suffice. The law, 

 therefore, amounts to this : that self- fertilisation will suffice only 

 when it is unavoidable. This is as much as to say that any 

 form of fertilisation is better than none. It is best to neglect 

 this form of Miiller's hypothesis, and to seek his meaning in his 

 simpler and broader statements. In summing up his discussion 

 in the "Historical Introduction," he says ("Fertilisation of 

 Flowers," p. 23) : "There is a good foundation, therefore, for 

 the demand that the explanation of floral mechanisms shall rest 

 only on the sufficient and demonstrable assumption that cross- 

 fertilisation yields more vigorous offspring than self-fertilisation." 

 We have therefore as the chief points in Miiller's theory : 

 (l) Fertilisation at any price, 

 (z) The increased vigour of cross-bred offspring. 

 Let us consider these more fully, and first for the conclusion 

 that self-fertilisation is better than no fertilisation. This is a 

 proposition which Miiller has insisted on in the most interesting 

 and instructive way, but it surely is not very novel in principle.^ 

 In a passage already quoted, Darwin reviewing, in 1876, his 

 work of 1862 ("Cross- and Self- Fertilisation," p. 8), says : 

 " I should have added the self-evident proposition that the pro- 

 pagation of the species, whether by self-fertilisation or by cross- 

 fertilisation . . . is of paramount importance. Hermann Miiller 

 has done excellent service by insisting repeatedly on this latter 

 point." No one had a higher respect than my father for 

 Miiller's work, and he had no disrespectful intention in 

 describing Miiller's contribution to the theory as self-evident. 

 The interesting point is that these views did not strike him as 

 original, because they had already occurred to himself. 



That Miiller based the explanation of floral mechanism on the 

 experimental results of cross- fertilisation cannot be considered 

 as a new departure. I should have imagined it to be notorious 

 that this was Charles Darwin's view, if it were not that we find 

 Knuth and others describing Miiller's theory (in which this is 

 the essential thing) as a great law of nature. 



In a letter ("Life and Letters," iii. p. 291) to the late 

 Asa Gray (September 10, probably 1866), Charles Darwin wrote : 

 " I have seen the young seedlings from the crossed seed exactly 

 twice as tall as the seedlings from the self-fertilised seed. . . . 

 If I can establish this fact . . . in some fifty cases . . . I think 

 it will be very important, for then we shall positively know why 

 the structure of every flower permits, or favours, or necessitates 

 an occasional cross with a distinct individual." 



It seems to me that Charles Darwin's generalisations in regard 

 to flowers may be summed up thus : — 



( 1 ) First comes what he called the self-evident proposition that 

 fertilisation of some sort is of paramount importance. This is 

 of the nature of an axiom. 



(2) Then comes the direct observation that the vast majority 

 of flowers are open. From this fact alone we should be justified 

 in concluding that there is some advantage in cross- as compared 

 to self-fertilisation, which advantage makes it worth while for 

 flowers to run the risks and incur the expenditure necessarily 

 connected with openness, and avoidable by cleistogamy. The 



1 I use the words cross-bred and self-bred to denote the offspring of cross- 

 and self-fertilisation ; we thus avoid the slightly obscure phrases cross- 

 fertilised and self-fertilised seedlings which occur in Darwin's books. 



2 I am far from wishing to suggest that H. Miiller's work does not 

 contain much that is new and valuable ; I am here considering only its 

 fundamental bases. 



innumerable adaptations for pollen-transport suggest and 

 strengthen the same conclusion. But this is, properly speaking, 

 only an elaboration of the fact that flowers are open. 



(3) Direct experiment demonstrates the nature of this 

 surmised advantage of cross-fertilisation over self-fertilisation. 



As already pointed out, the Knight- Darwin Law in its usual 

 form, i.e. no plant is self-fertilised ad infinitum, or in its 

 improved form — " Nature abhors perpetual self-fertilisation " — is 

 a generalisation drawn from observations on structure and 

 experiments on crossing, the value of which in Darwin's opinion 

 was rather its applicability to the problem of sex in a wide 

 sense, than its use as a basis for understanding the mechanisms 

 of flowers. 



The point which seems to me important in the history of the 

 subject, is that the above generalisations, which are in substance 

 to be found in Darwin's works, are still the foundation-stones 

 of floral biology, and would stand as firmlyif the Knight- Darwin 

 Law had never been formulated. For the naturalist who takes 

 a wider field, and studies the origin of sex and the action o. 

 changed conditions, the existence or non-existence of perpetual 

 self-fertilisation must always be an important question ; but the 

 law in which its non-existence is formulated, is not a fundamental 

 canon of floral biology. Francis Darwin. 



NO. 15 13, VOL. 58] 



BOTANY AT THE BRITISH ASSOCIATION. 



'X'HE subject of alternation of generations in plants played a 

 prominent part in the work of the Botanical Section. The 

 President (Prof Bower) devoted a considerable portion of his 

 address to the controversial questions connected with " the 

 great enigma of the alternation of generations " in green plants. 

 Mr. Lang, of Glasgow University, and Prof Klebs, of Halle, 

 contributed important papers on this subject, and these were 

 followed by a general discussion on the problems of alternation. 

 Mr. Lang gave an excellent summary and critical review of our 

 present knowledge concerning alternations of generations in the 

 Archegoniatse. The recent work of this investigator on some 

 striking cases of deviation from the normal life-history of ferns, 

 must be ranked among the most important contributions germane 

 to this subject which have appeared in recent years. In con- 

 cluding his account of some of the main factors in alternation, 

 the author suggested three subsidiary questions as worthy of 

 attention — the probable line of descent in archegoniate plants, 

 the bearing of the cytological facts on the question, and the 

 significance to be attached to apospory and apogamy. 



Prof. Klebs' paper dealt with the alternation of generations 

 in the Thallophyta, a subject which he was particularly well 

 fitted to discuss from a critical standpoint. After taking a 

 general survey of the various divisions of the Thallophyta, 

 Prof. Klebs referred more especially to certain cases which have 

 a more direct bearing on the question of the first appearance 

 of a regular alternation of generations. The majority of the 

 Algae and Fungi have two or more kinds of propagation, each of 

 which necessarily depends upon definite external conditions. 

 According to the conditions the different kinds of propagation 

 may appear on the same or on different individuals, inde- 

 pendently or in any succession. The fertilised ovum in sexual 

 forms does not differ essentially on germination from another 

 propagative cell. In none of these cases is there any reason 

 for speaking of an alternation of generations. In conclusion, 

 the author briefly referred to the possible connecting links 

 between the Algns and Archegoniatse. Sir Edward Fry, Dr. 

 Scott, Profs. Marshall Ward and Marcus Hartog took part in 

 the discussion which followed the two contributions by Mr, Lang 

 and Prof Klebs. 



Another important item in the programme of Section K was 

 a semi-popular lecture by Dr. F. F. Blackman, on the breathing 

 mechanism of plants. The lecturer gave a clear and interesting 

 summary of the progress of experimental work on the phenomena 

 of gaseous exchange between a green plant and the medium 

 in which it grows, concluding with an account of some recent 

 investigations which have not yet been published. 



Algce and Fttngi.— The Committee on Fertilisation in the 

 PhiKophycese reported very satisfactory progress in the re- 

 searches on the Fucacese and Dictyotaceae. Mr. Lloyd 

 Williams, of Bangor, whose researches have been carried out 

 under the auspices of the Committee, gave an account of his 

 important work on the reproduction of Dictyota dichototna. 

 Dic/yota, an annual brown seaweed, germinates during the 



