June 17, 1909J 



NA TURE 



455 



plague, diseases rarelv seen in this country nowadays, 

 we' think that brief accounts of Mediterranean fever 

 and cholera niitrht have been included with advantage. 



The book is profuselv illustrated, and some of the 

 photographs, though only in black and white, give a 

 remarkably good idea of the characters and distribu- 

 tion of rashes. R- T. H. 

 £e.';chrijving en Ondcrzoek van den gyroscopischen 



Horizon Flettriais {Model Pontliiis et Therrodc). 



By L. Roosenburg. Pp. 94; 3 plates. (Utrecht: 



Kemink & Zoon, igog.) 

 In- this pamphlet the author describes some improve- 

 ments introduced into the form of gyrostatic horizon 

 proposed by Fleuriais. In the original construction, a 

 top rotated in a chamber from which the air had been 

 removed, and the whole could be fixed to a sextant in 

 front of the horizon glass. Upon the top was placed 

 a glass scale, with arrangements for reflecting the 

 divisions of the scale into the sextant telescope in a 

 direction parallel to the equator of the top. The 

 iingle subtended by the divisions of the scale was ten 

 minutes, and the position of the object was estimated 

 •on this scale. 



In the new form here described, a temporary 

 vacuum onlv is made, and the chamber can be opened 

 Tor the inspection of the parts, and renewal of the top 

 point and the cup in which it rotates. The top is set 

 in motion bv an air-pump, which also creates the 

 vacuum. Observations are possible for fifteen or 

 twentv minutes. After the top has been rotating 

 some six or seven minutes and the precessional effects 

 rendered negligible, the sextant is clamped with the 

 sun or star in the field of view, and a considerable 

 number of readings taken of the position of the object 

 on the scale. Lastly, the reading of the sextant is 

 taken. 



The author insists on the necessity of a large 

 •number of readings in order to get good results, ap- 

 parentlv to eliminate the effect of irregular motion, 

 which in unfavourable circumstances can amount to 

 iV in three seconds. It is contended that though 

 practice with the instrument is necessary, it is not 

 difficult to use, is, in fact, easily mastered, and is 

 equally available for stars as the sun. The results of 

 more than 200 observations are given, and, with a 

 Tew exceptions, the errors of altitude are always less 

 than 3'. The author concludes that it is a trust- 

 worthy and very serviceable instrument for the deter- 

 mination of position at sea, preferable to other forms 

 of the same class. 

 Revue de Giographic anuncUe. Publi^e sous la 



Direction de M. Ch. \'elain. Tome ii., Ann(^e 



igoS. Pp. 730. (Paris : Ch. Delagrave.) Price 



15 francs. 

 This volume of the " Revue " ranges no less widely 

 than the preceding one. As regional geography we find 

 classified " Etude analytique du Relief de la Corse." 

 by J. Deprac, and " Le Perou," by C. Guibeaud. In 

 the mathematical department G. Perrier deals with 

 the figure of the earth and important geodetic opera- 

 tions, and .\. Berget writes on " Les M(?thndes et 

 les Instruments du Geographe Voyageur. " INI. Zim- 

 merman provides a review of half a century of 

 European colonisation, and P. Girardin studies the 

 subject of glaciation in the most recent geological 

 ■epoch. 



The first of these papers, that on Corsica, is an 

 Important contribution to the geography and geology 

 of an island which has not been as closelv studied as 

 might be supposed from its accessibility. M. Perrier 

 •tleals principally with the new measurement of the 

 arc of the meridian of Quito which is in the hands 

 of the Service geographique de I'Arm^e. It has been 



NO. 2068, VOL. 80] 



in progress for nearly ten years, of which the field 

 work alone occupied five, and its results are far from 

 complete as yet. 



The article on Peru by M. Guibeaud is a general 

 geographical study, most useful in its way. First it 

 provides a short survey of the country according _ to 

 natural regions, and then passes on to a consideration 

 of its chief economic, ethnographical, and political 

 aspects. This article is particularly well illustrated. 

 M. Zimmerman's study of colonisation is a careful 

 collection of facts and theories, with copious refer- 

 ences to authorities, vi'hich should form an excellent 

 foundation for the investigation of this subject of 

 world-importance. The volume, judged on French 

 standards, is particularly well printed and produced. 

 It is heavy and bulks large, and not a few readers 

 would no' doubt like to be able to obtain one or 

 other of its component articles separately. 

 Notes on Dynamics. By Sir G. Greenhill. Second 



Edition. Pp. 221. (London : His Majesty's 



Stationery Office, 1909.) Price 3^. 

 This cheap issue from His Majesty's Stationery Office 

 of a second edition of Sir George Greenhill's notes, 

 prepared for the advanced class of the Ordnance 

 College, Woolwich, will, we hope, become known to 

 teachers and students. The title is modest, the book 

 has never been advertised, and few people are aware 

 of its great value and originality. It contains many 

 excellent numerical examples, rather different from 

 those which teachers usually set in elementary dynamics 

 classes, but the reader will be even more interested in 

 letting the author carry him occasionally into problems 

 which are quite outside any ordinary curriculum. 

 When he deals with problems which are dealt w-ith 

 in the text-books, he takes a way of his own in each 

 case, and gives us new ideas. The end sections deal- 

 ing with the stability of rigid bodies moving in fluids 

 are of great interest. J- P- 



LETTERS TO THE EDITOR. 

 [The Editof does not hold himself responsible for opinions 



expressed 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 0/ Nature. 



No notice is taken of anonymous communications.] 

 The Temperature of the Upper Atmosphere. 



Like Dr. Chree (Nature, June 3), I do not think the 

 term " stratosphere " a suitable one, and isotherm.il layer 

 is obviously open to criticism. We should all be indebted 

 to Dr. Ctiree if he would suggest a better and more 

 accumte term. Isothermal column appears to me 

 sufficientiv accurate to describe the phenomena over a 

 single station, but cannot be applied to the whole upper 

 part of the atmosphere. Some single word implying the 

 absence of vertical circulation is required. 



There is no reasonable doubt that the daily temperature 

 variation becomes insignificant at a height of i km., and 

 hence one .is apt to infer that it is negligible at 10 km. ; 

 but the observations are not sufficiently well distributed, 

 and in my opinion the effect of solar radiation on the 

 balloon, if not on the instruments, is too uncertain for 

 us to speak positively about a daily variation at such 

 heights. The two years' observations in England have 

 shown no annual temperature variation above 10 km., and 

 I do not believe that there is any definite change from 

 summer to winter. It is probable that the mean monthly 

 temperatures at 10 km. do not differ greatly either with 

 latitude or with the season, although all the observations 

 yet available at 15 km. show lower temperatures over the 

 tropics than over temperate latitudes at that greater 

 height. 



I am glad to see that Dr. Chree does not assert that 

 errors of ±10° F. are the usual thing. Probably few of 

 those who use the instruments would assert that such an 

 error might not occasionally occur. If the figures for 



