128 



.VA TURE 



[Jink 6, 1895 



of the second decimal. Ver)' concordant determinations of 

 density gave as a mean number 1 9 '90. 



Argon, therefore, shows no sign of association on cooling, nor 

 'if dissociation on heating, as Prof. Bevan thinks it might. 



Kayi.eigh. 



Terrestrial Helium (.'i. 



Prof. Paschex and I have lately made a careful determina- 

 tion of the wave-length of the strong yellow line emitted by 

 clevcite when heated in a Pliicker tube. \Vc owe the mineral to 

 the kindness of Prof. Rinnc. My large R^iwland concave grating 

 of 6'5 metre radius, clearly shows the yellow line to be double. 

 Its less refrangible component is much weaker, but comes out 

 quite bright, when the stronger one is brilliant. We photo- 

 graphetl the two lines together with the second order of the 

 spark spectrum of iron. There are a number of iron lines on each 

 side that are included in Rowland's list of standard wave-lengths 

 (Phil. .I/af. . July 1S931. From these we interpolated the wave- 

 lengths of the yellow lines by micrometric measurement. Three 

 <lifferent plates taken on different days gave us : 



.Strong component. Weak component. 



5875-894 5876-216 



5875-874 5876-206 



5875-880 ... 5876-196 



Mean 5875-883 Mean 5876-206 



We think an error of more than 0-025 ^'^'V improbable. 

 Now Rowland's determination of D, (Phil. Mag., July 1893) 

 is : — 



5875-982 



the result of three series of measurements which he believes to be 

 accurate to o-02. 



The difference between this value and the wave-length of the 

 strong component is much too large to be accounted for by an 

 error of oliscrvation. 



We do not therefore agree with the conclusion, drawn by Mr. 

 Croukes, that the unknown element helium causing the line Dj 

 to ap|)ear in the solar spectrum is identical with the gas in cleveite, 

 tiiilisi /)j /'.; shman to he douhU. Perhaps Prof. Rowland will 

 tell us if this might have escaped his notice, l-'rom his note on 

 Dj in Phil. Mag. , July 1 893, it appears that Dj cannot have been 

 so wide as to include both lines, because he would then not have 

 considered his determination accurate to 0-02. As for dispersion, 

 'ine may see in his table of solar spectrum wave-lengths that he 

 has frequently measured three and even four lines in an interval 

 as large as the one beUsecn the components. 



Ilinnover Techn. Hochschule, May 16. C. RnNr.r.. 



The Origin of the Cultivated Cineraria. 



I HAD hojied that it would not be necessary for me to say 

 anything more upon this subject. But Mr. Bateson's last letter 

 seems to require a few remarks on my part. 



I confess that I find it very difficult to follow his train of 

 arguments. .\ll I can do is to restate once more my 

 ipriginil position, and endeavour to see how far Mr. Bateson has 

 b'cn successful in impugning it. I am sorry that Mr. Bateson 

 thinks I have " treated " him " to some hard words," though I 

 confess he seems to me, in that matter, quite able to take care of 

 himself. 



I asserted then (a) that the cultivated Cineraria only differs 

 from the wild form, pulling colour changes asirle, in dimensional 

 .liflVrtii, ..,. I Ijthcve that in saying this I am expressing 

 rate opinion of the Kew staff, the mcmocrs of 

 \\ is human nature, would have no hesitation in dis- 

 .i4t.-vin;{ with their chief, if they thought otherwise. To this 

 )i ■in; I <lo not understand that .Mr. Bateson advances any serious 



■ (b) I asterted that these dimensional differences had 



,iii .,-,,,„. ,i,i..,| T„ (his I understand Mr. Bales<m 



-ee that he has brought forward n 



■ the contrary. 



N.ja (.,f Mt. liiii-i'iii s own p.isilion. He.asserls, in common 



with other auth .riiii-«, ihii ih<- modern Cineraria is of hybriil 



origin. I ' : conclusion. And here I 



m»y qu'ji' . K. R.S., the well-known 



editor of ,,. i,., . „ , .,,,,. who in that paper for 



Januiry 24, 1891, p. loS, states ;—" Carnations and Picolces, 



NO. 1336, vol.. 52] 



again, which originate from ones(X'cies, vary from seed but not 

 from buds; and the same may be said of the Cineraria, the 

 offspring of one species." 



.Mr. Bateson complains that I do not give "any specific 

 answer" to the historical evidence. I thought I had made it 

 sufficiently clear in my last letter that: (iill doubted its value for 

 scientific purposes : (h) I set it aside as irrelevant on account of 

 the impossibility of proving the descent of the moilern Cineiaria 

 from its supposed ancestors. Both Prof. Weldoii and I have 

 shown that the historical evidence can be handled both ways. 

 But I prefer to set it aside altogether in the Atce of objective 

 facts. 



Mr. Bateson's next step is one to which I most seriously 

 demur. He transforms a proposition of mine into terms to 

 which I could not assent, and then proceetls to attack it. He 

 makes me say that '- to improve a plant the only safe way is by 

 selecting," &c. 1 absolutely never said anything of the kind. 

 "Iinprove" in horticulture is a word of large connotation. I 

 confined myself to the production of dimensional changes, and 

 I believe that what I said was in accordance with horticultural 

 experience. 



To demolish my position, Mr. Bateson has to get over the 

 fact, which seems to me incontestable, that there is no essential 

 morphological difference between the cultivated Cineraria and 

 the wild C. iiiitiila. To do this he trots out the Himalayan 

 rabbit. I cannot but admire his courage. What possible analogy 

 can there be in the two cases? Two "breeds' of rabbits are 

 crossed and produce a third different from cithi-r. If the modern 

 Cineraria is of hybrid origin, then it has eliminated traces of all 

 but one of its parents. The principle of economy of hypothesis 

 makes me slow to believe this, .\nyhow the Cineraria has 

 clearly not produced anything analogous to a Himalayan rabbit 

 which differs from both its iiarenls. 



.\s to Mr. Darwin's account of the origin of the Cineraria, I 

 must frankly take the responsibility. I have no doubt he worked 

 with ordinary garden kinds. He wrote to me for information as 

 to their origin. .\t the time I was entirely ignorant of the sub- 

 ject. I wrote to Mr. Thomas Moore, who was considered the 

 I best authority on such matters, and he sent me the traditional 

 \ account. I p.issed it on to Mr. Darwin, with the opinion, no 

 doubt, that I thought the information trustworthy. So I am 

 afraid Mr. Bateson is only appealing in this ca.se from Philip 

 sober to Philip ilrunk ; i.e. from my own considered opinion 

 to my unconsidere<l one. 



I will now wind up all I have to say on the subject with a few 

 I miscellaneous remarks. 



There can be no two opinions as to the importance of the 

 study, from the point of view of organic evolutii n, of the changes 

 which can be brought about in plants under cullivation. But it 

 must be conducted with scientific precision. Thisdiscu.ssion will 

 not have been fruitless if it directs attention lo the subject. .\ 

 beginning h.as already been made. M. Bornel has worked on 

 the genus Cisliis at .^ntibes, and has reconslrucled some of the 

 forms, as to the origin of which there was only "historical evi- 

 dence," described ami figured by Sweet. My friend Count 

 Solms-Laubach is eng.iged on the cultivated forms of Fmhsia, 

 and 1 am quite siire that any results he arrives at n>ay be 

 .acceiiled with implicit confidence. .\s he has asked me for 

 species of Cineraria, I hope he may look into this matter also. 



1 must repeat my caution as to theilanger of .accepting horti- 

 cultural evidence as to hybridity. I will give a few recent 

 instances. I could easily give a long list with chapter and verse 

 for each. 



((») Thuya fitiformis was long considered lo be a hybriil 

 hnWiUKnJtini/ennTirgiiiiaiiaanA a Thuya. Il is now known 

 to lie a " growth-sl.age" of Thuya orieiitalis. The history i^ 

 discusiicd by Sir Joseph Hooker in the Ganlentrs' Clirouiile for 

 June 22, 1861, pp. 575, 576. It affords a ilelightful commentary 

 on the hybridisitioii fall.acy and the value of "historical 

 evidence. " 



(h) Some years ago we received at Kew bulbs of what professed 

 tr) be a hybrirl between .Amaryllis Helladoinia and lirunr.'igia 

 foH(['hiu,e. When it flowered, it was evident that il was no 

 hybrid at all, but imly a very fine form of the former specie-. 

 This is rarely propag.aled from seed. In this particular case 

 seminal variation had come into play with correspomling iliinen- 

 sional change. The hybrid origin is recorded in the Gardeners^ 

 CA;-OH/,/<' for September 4, 1875, p. 302. It will, no doubt, be 

 dug out hereafter as "historical eviilcnce. " 



(<•) The la.st number rjfthe Gardeners' Chroiiitle(]\m^ 1. iSOj, 



