NATURE 



February i8, 1892 



tenon arrangement which distinguishes Lumbricus from AIlolo- 

 bophora. 



- -pros, 

 -peris. 



The male pore is situated normally on segment 15, but as the 

 papillae which carry the pores are large, they extend over the 

 adjoining segments on either side. Earthworms vary greatly in 

 this respect. Rosa says that spermaiheca; are absent in this 

 species, a peculiarity which has been noted in worms belonging 

 to several other genera. I have not sufficient material to enable 

 me to confirm or dispute this statement at present. I have 

 counted the segments of three specimens, and found them to be 

 in each instance 106. As the year advances I hope to.be able 

 to obtain mature adults for dissection, when it will be possible 

 to give a detailed account of the internal anatomy. Meanwhile 

 the external characters are amply sufficient for distinguishing the 

 worm if the girdle is properly developed, as its nearest British 

 ally {Lumbricus purpureus, Eisen) has the clitellum on segments 

 28 to 33. HiLDERic Friend. 



Idle, Bradford. 



The Implications of Science. 



Will you allow me to say something in answer to Mr. 

 Dixon's letter on this subject in Nature of January 21 (p. 272) ? 



(i) I admit that there is a verbal or symbolic "convention " 

 if two (or more) persons agree to understand any given words or 

 symbols in a way arbitrarily chosen by themselves. But the 

 scope of such convention is exceedingly limited : if people wish 

 to be understood, or even to understand themselves, they must 

 use the same words as others use, and use them in the same 

 sense (except in an infinitesimal proportion of case-). If it is 

 said that the common application and use of current words is a 

 mere convention, the word convention is taken in an extremely 

 strained and metaphorical sense, since nothing like an explicit 

 agreement has ever been made. The " convention " as to the 

 use of language is as fictitious as the social contract of Locke 

 and Rousseau. But in the one case, as in the other, there is a 

 solid basis of facts, to suit which the hypothesis has been pro- 

 duced. Language has been moulded by thought and feeling, 

 which, in their turn, have been impressed by facts ; and it is 

 facts and relations of facts that language seeks to express. As 

 Mill says (in the first chapter of his " Logic ") names are a clue to 

 thmgs, and bring before us "all the distinctions which have 

 been recognized not by a single inquirer but by all inquirers 

 taken together." No one, I imagine, would say that a 

 pcirticular case of the impossibility of affirming and denying a 

 given statement, depends "solely on the law of contradiction " ; 

 but in the case of any particular assertion, the impossibility, in 

 that case, is seen, and to a mind that has reached the generalizing 

 stage, the universal is discernible in the particular. As regards 

 the question of "real propositions," I will not occupy space 

 with quotations, but will only refer to Mr. Dixon's letter of Dec- 

 ember 10, in which the passages occur which led me to think 

 that he regarded assertions (or denials) of the existence of 

 particular objects as the only " real " propositions. 



(2) As regards induction, I agree with Mr. Dixon that the 

 startmg point in induction is hypothesis or discovery. But with 

 reference to the rest of the procedure, and its relation to 

 so-called "formal " logic, I diffisr from him. For I think that 

 an inductive generalization maybe set out syllogistically ; e.g., 



What has once produced X will always produce X ; 



A has once produced X ; 



I •. A will always produce X (= all A is X). 



If space allowed, I should like to consider the justification for the 

 major premiss, and also to say something about the grounds on 

 which the minor (which indicates the hypothesis or discovery) 

 asserts causation [or concomitance] in a given instance. 



(3) Mr. Dixon says : " We do not, in mathematics, conclude 

 a universal proposition from a single concrete instance." But 

 it appears to me that, as far as my own experience goes, in every 

 concrete mathematical proposition which I understand this is 

 exactly what happens ; and I do not see how, on Mr. Dixon's 



NO. II 64, VOL. 45] 



view, mathematical formuloe could ever have been constructed. 

 "A mathematical formula," Mr. Dixon remarks, " does not 

 imply the existence of any instance whatever of its application, 

 any more than a definition implies the reality of the thing 

 defined." But if a definition is always of a thing, what more is 

 wanted? The definition is admitted to be oi something ; and 

 what is something must, I suppose, exist somehoiv. 



(4) I still think that in the passage in Mr. Dixon's letter 

 which I referred to under (4) he is not consistent. For if, as he 

 asserts, the definition oifoura.s = i-f- 1 + i, makes it false to say 

 that Tzyice two are four, this is surely because \he facts referred 

 to hy four are no longer what they were when the statement in 

 question was true. If definitions were purely arbitrary, as Mr. 

 Dixon holds, what would prevent my saying that Four (i-f i-fi) 

 means twice two (i + i) + (i + i)? it is surely only the refer- 

 ence to things which makes it absurd— (and, however/o«r (4) 

 may be defined, how is one (i) to be understood, except by refer- 

 ence to things?). 



That words and symbols used intelligibly do, and must, refer 

 to something beyond themselves, seems to me indisputable. If 

 they did not, no assertion of the form S is P could ever be 

 made, for the symbol S is certainly not the symbol P. And for 

 any statement, of the form .S' is P, to Ifc possible and significant, 

 it is further necessary that S and P should have identical appli- 

 cation, but diverse signification. If application and significa- 

 tion were the same, we should get 6' is S and P is P ; if applica- 

 tion were tiot the same, we must say, S is not P. Hence, no term 

 can ever be taken in mere denotation (or application), nor in 

 mere connotation (signification) ; but both momenta of each 

 term have to be taken into account in every assertion. If (to 

 take a case given by Mr. Dixon in his " Essay on Reasoning," 

 p. 8) we "define" metal as "the list of denotation, iron, 

 copper, tin, zinc, lead, gold, and silver," then iron, &c., can 

 only be pointed out by taking some specimen of iron, and 

 saying, This and all other things which are -like it in certain 

 respects. An absolutely arbitrary denotation can be given only 

 if the whole of the objects denoted are severally pointed out ; 

 and even then, unless they are labelled, they can only be re- 

 membered and identified by means of their characteristics ; if 

 labelled, by that characteristic. 



Mr. Dixon objects to my attributing to him the view that 

 "mathematical truths in as far as ' real ' are obtained by induc- 

 tion, and are therefore not necessary." But in his letter of 

 December 10 he says :—" For example, the assertion 'Two 

 straight lines cannot inclose a space' is certainly not a 'neces- 

 sary truth.' Either its terms are defined by connotation, so that 

 its truth depends solely on those definitions, or else its terms 

 are defined by denotation, as representing real things in space ; 

 and the trzith of the assertion can only be proved by induction 

 from actual experience with those things. In the first case, the 

 truth is arbitrary, not necessary ; and in the second case it 

 might conceivably be false, as was shown by Helmholtz." It 

 was this passage which led me to the opinion which I expressed. 

 Cambridge, January 31. E. E. C. Jones. 



Vacuum Tubes and Electric Oscillations. 

 I have not had the advantage of hearing the lecture of M. 

 Nikola Tesla nor of seeing his experiments, but it does not seem 

 out of place to recall the attention of your readers to an article 

 by Dr. Dragoumis in your issue for April 4, 1889, in vol. xxxix. 

 p. 548. Oliver J. Lodge. 



THE NEW STAR IN AURIGA. 



CINCE our last article w^as vv^ritten the weather has 

 '--^ continued very bad for astronomical observations. 

 The only new results obtained which have reached us 

 consist of a paper read by Mr. Norman Lockyer at the 

 Royal Society on Thursday last, and an important 

 telegram from Prof. Pickering, which appeared in 

 Wednesday's Standard. 



We will take these in order. Mr. Lockyer's com- 

 munication to the Royal Society was dated February 8 ; 

 it stated that two more photographs, containing many 

 more lines than the former ones, were taken on Sunday 

 night, February 7, and it went on to make the important 

 announcement that " The bright lines K, H, h, and G are 



