364 



NATURE 



[February 16, 1893 



improved by a firmer grasp of scientific principles. Tlie 

 commercial statistics are, as might be expected, much 

 fuller, better arranged, and more serviceable than those 

 relating to physical geography ; but we imagine that few 

 members of the Imperial Institute, likely to make 

 use of the book, are without the original records relating 

 to their own department. The difficulty of propor- 

 tion and perspective is rather seriously apparent 

 in the treatment of India, which has to be passed over 

 more lightly than the colonies, because equal detail 

 would involve the sacrifice of much space. Thus the great 

 internal trade of India is scarcely touched upon, and the 

 wants and tastes of consumers in the ultimate Indian 

 market, by whom imports are finally absorbed, are not 

 laid before the British merchant. 



Beneath Helvellyn^s Shade. By Samuel Barber. (London: 



Elliot Stock, 1892.) 

 This book consists of notes and sketches in the Valley 

 of Wythburn, and is brightly and attractively written. 

 Perhaps the best chapters are those on clouds, the 

 various forms of which have been carefully studied by 

 the author. He has also many interesting remarks on 

 various aspects of Cumberland scenery, on the customs 

 of the people, and on antiquities. Occasionally, perhaps, 

 Mr. Barber adopts too much the tone of a preacher, but 

 his impressions and ideas are for the most part 

 fresh and vivid. The book will especially please 

 those who have themselves felt the charm of Words- 

 worth's country. 



LETTERS TO THE EDITOR. 



[ The Editor does not hold himself responsible for opinions ex- 

 pressed by his correspondents. Neither can he undertake 

 to return, or to correspond with the writers of, rejected 

 manuscripts intended for this or any other part of iihlVR.^. 

 No notice is taken of anonymous communications. "] 



Dr. Joule's Thermometers. 



Respecting the question asked by Mr. Young (Nature, 

 vol, xlvii. p. 317), I am glad to have an opportunity of stating 

 that shortly after Joule's death I obtained' the sanction of his 

 son to examine tjfie scientific apparatus that were left in his 

 house. 



I found a number of thermometers, and amongst them the 

 two chiefly used by Joule in his researches. These thermo- 

 meters have been placed in my charge for the present. I have 

 made careful comparisons of them with a standard of the 

 " Bureau international des Poids et Mesures," and therefore 

 indirectly with the air or hydrogen thermometer. A standard 

 issued by the Technische Reichsanstalt has also been used as a 

 check. I spent a good part of last winter on the work and am 

 now only waiting for an opportunity to repeat some of the 

 measurements. The results will be published in due course, 

 and I think will prove of interest. As Joule compared his 

 thermometers with one used by Rowland, we shall in this way 

 have an indirect comparison of Rowland's air thermometer with 

 those by which the Berlin and Paris standards have been in- 

 dependently fixed. 



One question arises on which I should be glad to have some 

 information, and I should be grateful to any of your readers 

 who could help me. The glass of which Joule's thermometer 

 is made does not behave like the English glass now in use ; 

 and it would be important to know the probable composition 

 of glass used in England about the year 1840 for thermometric 

 purposes. As my experiments are not concluded I do not wish 

 to speak with too great a certainty ; but I believe it will be found 

 that if we could return to the glass of Joule's thermometer, we 

 should have a substance as well and possibly even better adapted 

 to the manufacture of thermometers than the modern Jena or 

 French thermometer glass. 



I am sorry I cannot give a very definite answer to Mr. Young's 

 question. Joule does not, as far as I know, anywhere give the 

 actual readings of the freezing point, but only its changes. 

 Rowland, in quoting the comparison between Joule's thermo- 



n6. I 2 16, VOL. 47] 



meter and his own, gives 22 "62 as the actual reading of Joule's 

 zero. I have not at the present moment access to Rowland's 

 paper, and have no note of the date at which this comparison 

 was made (ehher 1879 or 1880). 



Such a formula as that given by Mr. Young can, however, 

 only have a limited application. The zero of a thermometer 

 depends on the temperature at which the thermometer has been 

 kept previous to its immersion into ice, and with properly- 

 annealed thermometers the secular changes are much smaller 

 than the temporary ones. Last winter Joule's thermometer showed 

 changes in zero from 23*51 to 23 00 on the arbitrary scale, the 

 original temperatures varying from ^° to 30°. 



All observations lead to the conclusion that the secular 

 changes of a thermometer gradually vanish, so that the zero 

 corresponding to any temperature approaches a limit. Mr. 

 Young's formula would make the zero rise indefinitely. 



Arthur Schuster. 



Dust Photographs and Breath Figures. 



Your two correspondents on February 9 add interesting in- 

 stances of these phenomena. I am sorry that one of my statements 

 was not clear. In saying " Two cases have been reported to me 

 where blinds with embossed letters have left a latent image on 

 the window near which they lay," I meant to describe them as 

 not in contact. 



I have questioned my neighbour Dr. Earle again as to his 

 case. The plate-glass window of an hotel in London has on the 

 inside a screen of ground glass lying near but not touching : 

 upon the latter are the words "Coffee Room" in clear un- 

 frosted letters. One day as he was at breakfast the screen was 

 taken away, but the words were left plainly visible on the 

 window, and no washing would remove them. The other case is 

 curiously similar, but each narrator was ignorant of the other's 

 tale. A friend, Mr. Potter, asked me if I knew whether a house 

 in which he was lodging had been an hotel, for on misty days 

 they saw " Coffee Room " on one of the windows. I remembered 

 the house had been an hotel two or three years previously, and 

 there had been brown gauze blinds with gilt letters. 



Mr. Thiselton-Dyer's observation appears not so much akin to 

 these two as to the dust picture of a water-colour drawing of 

 which I spoke in my former paper. 



I look forward to seeing the effects at Canterbury. 



Winchester College, February 13. W. B. Croft. 



Fossil Plants as Tests of Climate. 

 Mr. De Range's note relating to the above subject in 

 Nature, p. 294, mentions that "Heer has determined a mag- 

 nificent flora of more than 350 species from these northern 

 tertiaries, and that he at once pointed out the absence of tropical 

 and subtropical forms." My contention, founded on an attentive 

 study of his determinations and of the original specimens in 

 London and Dublin, and to some extent in Copenhagen, is that 

 not fifty, or perhaps not the half of fifty, of these determinations 

 are entitled to the smallest weight ; and again that though at first 

 he saw nothing subtropical in the flora, he subsequently declared 

 the presence of palms, &c., upon utterly insufficient data. While, 

 however, wishing to rid the *' magnificent " flora of 300 or more 

 useless and misleading encumbrances, I am far from wishing to 

 depreciate the extraordinary significance and value of that which 

 remains, and which clearly shows that in early Eocene times the 

 coast of Greenland supported in certain places forests which 

 included the redwood, the plane, and even the magnolia, asso- 

 ciated with many more northern forms. This is consistent with 

 the tropical vegetation existing during a part of the possibly con- 

 temporary lower tertiary period in the south of England. Both 

 facts are sufficiently inexplicable, but there is no occasion to 

 magnify the difficulties they present. As to the Greenland floras 

 they have not been proved to contain any forest trees that might 

 not, and which in fact do not, flourish in their modern repre- 

 sentatives, when planted in certain favourable spots on the west 

 coast of Ireland, and even of Scotland. We are not even obliged 

 to assume that Greenland as a country was characterised by such 

 vegetation, for this might be as erroneous as to regard Ireland or 

 Scotland as countries generally characterised by forests of 

 arbutus. The flora of a country is in fact most likely to be pre- 

 served in its most sheltered spots, in lake bottoms like parts of 

 Killarney, or where small rivers quietly steal into the tidal waters 

 of deeply recessed bays like those of Bantry and Kenmare, in 

 forest pools like some in the Mount Stewarts' woods of Bute,and 



