3iS 



NATURE 



\yan. 31, 1889 



dent has sent to me an unmistakable specimen of annelid 

 piping from a rock at Killin. Now that we are all on the 

 scent, I have no doubt that my Loch Fyne worms will be found 

 at the head of a long procession. Argyll. ■ 



Inveraray, January 28. 



, Mr. Howorth en the Variation of Colour in Birds. 



Permit me to p )int out that the fact cited by Mr. Floworlh 

 {antea, p. 294), of the similarity between the birds of Western 

 Europe and those of Japan, which, accordinsi to him, ilkistrates 

 "from an unexpected quarter" the views expressed in his work 

 " The Mammoth and the Flood," is by no means so novel as 

 the general reader might, from his communication, suppoe. On 

 the contrary, it has been before the world for more than half a 

 century, and is, or ought to be, familiar to every well-informed 

 ornithologist, since it was in 1835 that Temminck gave (" Manuel 

 d'Ornithologie," Part 3, Introd. pp. l.-liii.) a list of 114 species 

 of birds, which he said were common to the Japanese Empire 

 and to Eurjpe. It has s-ubsequently been shown that this 

 number was exaggerated ; but, as observed in 1857 by Mr. 

 Sclater, in his classical paper on the geographical distribution of 

 birds, "there can be no question as to the general strong 

 resemblance of the Japanese avifauna to that of Europe " 

 (Journ. Linn. Sic, Zojlogy, ii. p. 134). 



That the birds of Siberia differ in appearance from thr.se of 

 Japan and of Western Europe has also been abundantly shown 

 by the long succession of illustrious explorers and naturali>ts 

 from whom we derive all but an infinitesimally small portion of 

 our knowledge of Siberian and Central Asiatic zoology ; but 

 Mr. Howorth does not seem to be aware that another hypothesis 

 has usually been 1 rought forward to account for that difference, 

 as well as for the similarity of the Japanese avifauna and our 

 own. In regard to the latter, this hypothesis has be^n pro- 

 pounded to my certain knowledge for more than five-and-twenty 

 years {Ibis, 1863, p. 1S9) ; and, whether true or false, has 

 jiot, so far as I am aware, been refuted. It is that the resem- 

 blance is an effect due to " the prevalence in both localities of an 

 insular, as opposed to a continental climate," for it has been 

 observed over and over again in various parts of tha world (an 1 

 not only in birds) that island forms possess a darker coloration 

 than their continental representatives {cf. Gould, Proceedings 

 of the Zoological Society, 1855, p. 78). while the fact that 

 neither British nor Japanese birds exhibit the brilliant hues of 

 their continental kinsfolk is notorious. 



Another remark upon Mr. Howorih's co.nmunication T would 

 offer. His theory, of which he has " little doubt," that "the 

 willow grouse of the Continent i:- an altered form, and that our 

 red grouse is the parent," is not new, but was definitely stated 

 some eight or nine years ago in the " Kncyclopse lia Britannica," 

 from the article "Grouse" in which I venture to quote a 

 passage showing that there is a side of the question which he 

 does not seem to have considered. 



" A very interesting subject for discussion would be whether 

 Lagjpus scoticus or /. al'')us has varied most from the com-non 

 stock of both. We can here but briefly indicate the more 

 salient points that might arise. Looking to the fact that the for- 

 mer is the only species of the genus which does not assume white 

 clothing in winter, an evolutionist might at first deem the varia- 

 tion greatest in its case ; but then it must be borne in mind that 

 the species of Z'r^(7///jr which turn white differ in that respect 

 from all other groups of the family Telraonidte. Furthermore, 

 it must be rememV)ered that every species of Lagopus (even L. 

 leucurits, the whitest of all) has its first set of rciiiges 

 coloured brown. These are dropped when the bird is about 

 half-grown, and in all the species but L. scoticus. white 

 remiges ?,x& \.\\&v\. produced. If, therefore, as is generally held, 

 the successive phases assumed by any animal in the course of its 

 progress to maturity indica'e the phases through which the 

 species has passed, there may have been a time when all the 

 species of Lagopus wore a brown livery even when adult, and 

 the white dress donned in winter has been imposed upon the 

 wearers by causes that can \iz easily suggested, for it has been 

 freely admitted by naturalists of all schools that the white 

 plumage of the birds of this group protects them from danger 

 during the snows of a protracted winter. On the other hand, it is 

 not at all inconceivable that the Red Grouse, instead of per- 

 petuating directly the more ancient properties of an original 

 Lagopus that underwent no great seasonal change of plumage, 

 may derive its a'lces'.ry from the widely-ranging Willow-Grouse, 



which, in an epoch comparatively recent (in the geologica 

 sense), may have stocked Britain, and left descendants that, 

 under conditions in which the assumption of a white garb would 

 be almost fatal to the preserva'.ion of the species, have reverted 

 (though d:)ubtless with some modifications) to a comparative 

 immutability essentially the same as that of the pri nal L.agopus." 



In conckision, let me recommend those interested in the local 

 viriation of colouring in the plumage of birds to study Gloger's 

 ^'Das Abiindern der Vbgel durch Einflass des Klima's" 

 (Breslau, 1833), a treatise which, though naturally out of date 

 in many respects, contains much that ornithologists would be 

 the better for no! overlooking. Alfred Nevvtox. 



Magdalene College, Cambridge, January 26, 



Constitution of the Chlorides of Aluminium and the 

 Allied Metals. 



In Nature of Dacember 27 (pp. 198-200), Dr. Young has 

 directed the attention of your readets to the formulae of the 

 so-called hexachlorides, RgClfi, and bodies of similar constitution. 

 As this question possesses a great theoretical interest, I may be 

 allowed to say a few words about it. 



The constitution of the chlorides RoCl,, was regarded by Friedel 

 as analogous to that of carbon hexachioride, Clj^C — C^Cls,' 

 and the metals themselves were said to be tetravalent. Mende- 

 lejeff, in his classical paper on the periodic law— a gem of 

 chemical literature— was the first, to point out that it is not 

 necessary to regard aluminium in AL^CIg as tetravalent, but 

 that a compound of this formula represents only a molecular' 

 combination — a polymeric state— of AICI3. This view was 

 very little regarded by chemists. It was, ten years later (Proc, 

 Vienna Acad., 1882), extended by the author of the present 

 lines, to other bodies of similar constitution, and it was shown 

 that several chlorides give a vapour consisting partly or entirely- 

 of polymeric modifications of the simple chloride. So we have 

 the molecules Sn,,Cl4— SnCl^, Fe._,Cl4— FeCly (N'oO^-^NOa), and . 

 it was supposed that other substances of similar constitution 

 would split up into simpler molecules, if they could bear 'highei' 

 temperatures, especially AI.jCl,;, aluminium belonging to a natural' 

 group of trixd elements, whose other members, vi2. B and In, 

 give chlorides of the simple and molecular formulae, BCI3 and 

 InClj. As further instances of the above case may be quoted 

 to-day, HoF,— HF, Be.^Cl4— BeCI,, GaaClg— GaCl^. 



It may, however, be usefid to those readers of Nature who 

 cannot follow completely the current chemical literature, to 

 know that the values of vapour-densities obtained by Dumas's 

 method a-e not strictly compa-able with those obtained by V. 

 Meyer's method ; for, a; has been shown lately by Victor Meyer 

 and his pupils, especially Biltz, the vapour density of dissociable 

 j substances is found smaller by V. Meyer's apparatus than by 

 that of Dumas at one and the same temperature. So Dumas's 

 method shows, for the above chlorides, the existence of double 

 molecules at a temperature at which that of V. Meyer indicates 

 single molecules. It was already pointed out by Dr. Young 

 that the single molecules, GaCU, InCl.j, InCI), and CrClj, were 

 found by V. Meyer's method only, and there is no doubt that 

 some of them may be found double by Dumas's method. 



The result of the vapour-density determinations of aluminium 

 chloride and the chlorides of the allied metals, carefully collected 

 by Dr. Young in Nature, is of great theoretical importance, 

 for, as the chlorides RCI3 exist as single molecules in the 

 gaseous state, the respective metals are undoubtedly trivalcnt in 

 these CO npounds. 



But the question arises, what is the valency of these metals 

 when their chlorides have the double molecule RoCl,j ? 



It is impossible to regard these compounds as analogous to 

 carbon-hexachloride and the said metals as tetravalent, for that 

 compound, when heated to a higher temperature, does not split 

 up into two molecules, CCL,, as the chlorides R.,C1b do. More- 

 over, it must be admitted that the theory of valency which was 

 developed by the study of organic compounds {caiboti deriva- 

 tives) does not strictly hold good in the case of the remaining 

 elements. It is impossible to state the number of " bonds " by 

 which the chlorides RCI3 are kept together in double molecules, 

 R.^Cly. This number is certainly not one (CI3R— RCI3) ; just as 

 two molecules of SiiCl., in Sn._,Cl4 cannot be regarded as being 

 united by one bond, CI.,Sn"' — Sn"'Cl2, for no one will assume 

 that tin is trivalent in this compound. If we admit any bonds 

 at all between the two Sn atoms, we must admit two, CLSn^z 



