8o 



NATURE /fi 



[Nov. 28, 1889 



Galls. 



In his suggestive paper on Prof. Weismann's theory, Mr. 

 Mivart says, while alluding to the formation of galls, " It would 

 be interesling to learn how natural selection could have caused 

 this plant to perform actions which, if not self-sacrificing (and 

 there must be gome expenditure of energy), are at least so 

 disinterested." 



Mr. Mivart here strikes what has always appeared to me one 

 of the most important facts in organic nature with reference to 

 the theory of natural selection. I have always so considered it, 

 because it seems to me the one and only case in the whole range 

 of organic nature where it can be truly said that we have un- 

 equivocal evidence of a structure occurring in one species for the 

 exclusive benefit of another. 



INIoreover. the structure is here a highly elaborate one, entail- 

 ing not only a drain on the physiological resources of the plant 

 (as Mr. Mivart observes), but also an astonishing amount of 

 morphological specialisation. Indeed, the latter point is so 

 astonishing, that when we study the number and variety of gall- 

 formations in different species of plants — all severally adapted to 

 the needs of as many different species of insects, and all presenting 

 more or less elaborate provisions for ministering to such needs — 

 it becomes idle to doubt that, ifsuch cases had occurred elsewhere 

 and with any frequency in organic nature, the theory of natural 

 selection would have been untenable, at all events as a general 

 theory of adaptations and a consequent theory of species. But 

 seeing that the case of galls is unique in the relation which is now 

 before us, it becomes reasonable to attribute the formation of 

 galls to the agency of natural selection, if there be any con- 

 ceivable manner in which such agency can here be brought to 

 bear. 



Now, although it is obvious that natural selection cannot 

 operate upon the plants directly, so as to cause them to grow 

 galls for the benefit of insects, I think it is quite possible to 

 suppose that natural selection may operate to this end on the 

 plants indirectly throiigk the insects, viz. by always selecting those 

 individual larvae the character of whose excitatory emanations is 

 such as will best cause the plant to grow the kind of morpholo- 

 gical abnormality that is required. 



This explanation encoun'ers difficulties in some special cases of 

 gall-formation, which I will not here occupy space by detailing ; 

 but as it is the explanation given in a course of lectures which 1 

 am at present delivering to the students here, I should like to 

 take the opportunity, which Mr. Mivart's paper affords, of asking 

 whether anybody else has a better explanation to offer. 



George J. Romanes. 



Edinburgh, November 18. 



" Modern Views of Electricity." 



Your reviewer (p. 5) takes rather high ground wherefrom to 

 criticize a confessedly popular and expository book ; and some 

 of the charges of vagueness — as, for instance, that I do not 

 definitely specify the velocity with which electricity travels in a 

 given current — strike me as rather out of place, seeing that the 

 same charge might be made against the. treatise of Clerk- 

 Maxwell. A want of definiteness about the constitution of the 

 ether I must perforce admit ; and I can hardly be surprised at 

 your reviewer's want of sympathy with my struggles to convey 

 to non-mathematicians some idea of the tendencies of modern 

 inquiry, when I find that he thinks it "open to question whether 

 attention has not of late years been too much diverted from the 

 condition of the charged bodies in the electric field to that of 

 the medium separating them." 



But it is not so clear how, holding this view, he can say that 

 the tentative theory attempted to be explained by me "is in its 

 most important features almost identical with the old two-fluid 

 [action at a di-tance] theory published by Symmer in 1759"; 

 nevertheless, by taking a few statements from the earlier and in- 

 troductory portion of my book, and caricaturing them a little, he 

 does manage to make it appear as if the so-called "modern 

 views" were merely a case of reversion to an ancestral type. 



However, it is not on these general topics that I break a 

 wholesome rule and reply to a review : it is because I am 

 charged with four or five definitely misleading statements, and 

 it is these I wish to either withdraw or justify. 



First, concerning the relation between the Peltier effect and 

 the E. M.F. at a junction. I have argued this matter out fully 

 in the Philosophical I\Iagazi nciox March 1886, p. 269, and have 



shown that the only " further assumption" needed is this: — 

 The nteasitre of the E.M.F. at any section of a circuit is the work 

 done per unit electricity conveyed past that section, or, dlV = 

 QdE. Until this is disproved I regard it as axiomatic : and, so 

 regarding it, I hold that what I have said about contact E.M.F. 

 is true. My position in the matter is, at all events, perfectly 

 clear and definite, and is fully explained in the Philosophical 

 Magazine article referred to, as well as in several others of older 

 date. 



Second, as regards tourmaline. I certainly did not intend to 

 explain pyro-electricity as due to unilateral conductivity solely, 

 but perhaps my brief statements concerning it on p. 122 might 

 be more cautiously worded so as to avoid any possible mis- 

 conception. 



Third, the " dead-water " argument against electric momentum 

 (p. 103) is not left as a valid proof of its non-existence, though it 

 is introduced as at first sight so tending ; and all that my critic 

 says against it resolves itself into a question of degree. 



The same is true of what he says on the fourth point, concern- 

 ing Fitzgerald and the Kerr effect ; and his assertion that Fitz- 

 gerald's deductions do not coincide with the observations of Kerr 

 and Kundt seems to me to convey a much falser impression than 

 my nine-year-old statement (p. 323) to which he objects : " Mr. 

 Fitzgerald, of Dublin, has examined the question mathematically, 

 and has shown that Maxwell's theory would have enabled Dr. 

 Kerr's result to be predicted." 



Lastly, my suggested possible account of the Thomson effect 

 (po. 117, 120, 295), though it does not indeed altogether hold 

 water (as both Prof. Everett and Prof. J. J. Thomson have 

 kindly pointed out to me), breaks down for a reason entirely 

 different from that supposed by your reviewer, who is estimating it 

 only from his own caricature of an ether theory. The real weak 

 point lies in forgetting that the condition required is unequal 

 impulse, not simply unequal force. 



In thus replying to objections raised, I by no means suppose 

 that my critic has made them in any unfriendly spirit. I only 

 feel that he has read the book rather un sympathetically, and 

 (possibly on account of faults in the preface) has regarded it as 

 more scientifically pretentious than its style and object at all 

 warrant. Misleading statements as to matters of fact I have 

 indeed strenuously endeavoured to eschew, and I trust that to 

 very few of them shall I have, in a second edition, to plead 

 guilty. Oliver J. Lodge. 



November 16. 



Geometrical Teaching. 



Mr. Woodall has called attention to an evil which, even at 

 the present day, is more extensive and persistent than is 

 generally supposed to be the case by those who imagine that 

 "improved methods of geometrical teaching" are making 

 themselves felt. 



It is surprising that such a subject as Euclid, which of all 

 subjects perhaps is best calculated to produce in the minds 

 of young persons an exact method of reasoning, should be so 

 badly taught. There can be, I should imagine, only one 

 opinion as to the method of teaching described by Mr. Woodall, 

 viz. that it is decidedly bad ; and even worse, that it is perfectly 

 useless. 



It is often objected by this class of teachers that young people 

 cannot be brought to appreciate the intricacies and subtleties of 

 Euclid's propositions, and that, in consequence, if they be learnt 

 at all they must be learnt by heart. But is not this a great 

 mistake? My own experience has shown me that young persons 

 can be induced to appreciate and take an intelligent interest in 

 Euclid if it be taught intelligently. This demands some little 

 trouble on the part of a teacher, and I suspect that a large 

 proportion of our bad geometrical teaching is due to the 

 disinclination of the teacher to take overmuch trouble in his 

 work, coupled with the fact that it is often very difficult for 

 him to get over the superstition of his own school-days, that a 

 proposition, if it be learnt at all, must be learnt by heart, 

 without any display of intelligent interest. 



It does not seem to me to be nece>sary, at the outset at 

 any rate, in order to improve the teaching, that the ordinary 

 well-known edition of Euclid should be taken to pieces and a 

 new and elaborate arrangement of the propositions made out of 

 the fragments. The effective teaching of Euclid may be con- 

 ducted upon the old lines, so well known to us in Potts and 



