488 



NATURE 



[September 22, 1892 



Cynosurus cristatiis is included. This latter species is 

 of limited value, and in permanent pasture only. 



Genera are given pp. 4-1 1, and the following fifty pages 

 are occupied by an account of those species which have 

 up to the present been found in Great Britain or Ireland. 

 The rest of the book is devoted to the sedges. The deri- 

 vations of the names of genera are mostly given, as well 

 as the French and German synonyms of the different 

 species discussed. The illustrations are satisfactory, and 

 are in general given for those species which are most 

 common. That of Triticum repens, on p. 32, is perhaps 

 exceptional. The beginner very often confuses the spike 

 of this grass with that of some varieties of rye-grass. 

 The spikelets of the latter are set edgewise to the rachis. 

 whilst those of the former have their flat sides to the 

 rachis ; if the beginner is still in doubt the rootstock can 

 be examined ; this is stoloniferous in the case of couch- 

 grass. 



Elementary Plane Trigonometry. Clarendon Press 

 Series. By R. C. J. Nixon, M.A. (Clarendon Press, 

 Oxford, 1892.) 



The author in his preface informs us that in writing this 

 book he has tried to free his mind as far as possible from 

 all current text-books, and to base this one solely on his 

 experience of twenty-five years. That he has done this 

 is soon seen when glancing through the pages, for the 

 order of arrangement and general basis differ very con- 

 siderably from those usually adopted. The line of 

 demarcation he draws between elementary and higher 

 works lies in the use and non-use of the symbol V — i 

 thus avoiding here altogether the use of imaginaries. 

 An omission which may seem rather questionable is that 

 of the theory of logarithms, which is here excluded as it 

 does not appertain to trigonometry proper ; the beginner 

 is not left entirely without logarithms themselves, for 

 there are two chapters in which he can make a slight 

 acquaintance with them, together with one on the adap- 

 tation of formulae to logarithmic calculation. Through- 

 out the work the author has made a strong point°of 

 giving in their fulness and generality all definitions and 

 proofs, while he has added also numerous examples, 

 some of which are worked out to serve as specimens, 

 while others are accompanied with hints as to their 

 solution. 



If any fault be found in the book it is perhaps that it 

 has been expanded to too great dimensions : excluding 

 the answers at the end there are no less than 364 pages, 

 which, for an elementary work of this kind, is un- 

 doubtedly a large number. At any rate the error is made 

 on the right side. In all other respects the book can be 

 decidedly recommended, for the propositions are all 

 neatly proved, and the get-up, as regards the figures and 

 letters, could scarcely be surpassed. 



Paraguay : The Land and the People, Natural Wealth 

 and Commercial Capabilities. By Dr. E. de Bour- 

 gade La Dardye. English Edition. Edited by E. G. 

 Ravenstein. (London : George Philip and Son, 1892.) 



Every one who has any reason to be interested in Para- 

 guay ought to read this book, which is in most respects a 

 model of what such a work ought to be. The author 

 spent two years in the country, so that he had ample op- 

 portunities for making himsslf acquainted with its leading 

 characteristics. His impressions, upon the whole, were 

 very favourable ; but there is not the faintest atter^pt to 

 convey an exaggerated idea either of Paraguay's re- 

 sources or of the use she is making of them. M. de Bour- 

 gade writes in a spirit of scientific impartiality, bringing 

 out the facts exactly in accordance with what he believes 

 to be the most trustworthy evidence. He begins with an 

 account of the geographical exploration of the country, 

 NO. 1 195, VOL. 46] 



then presents a geological survey, and describes the 

 basins of the Parana and the Paraguay, and Paraguay's 

 vegetable and animal life and minerals. Next there are 

 chapters on various aspects of social life— government, 

 and laws, financial position, real property, population, 

 and immigration. A section on "Labour" includes 

 chapters on means of communication, the soil, stock- 

 breeding, agricultural products, tobacco, timber, textile 

 plants, various raw materials, yerba-mate, and the orange. 

 On all these subjects the author writes clearly and with 

 full information. The work is enriched with a map and 

 illustrations, and of the translation we need only say that 

 it has been done carefully and adequately. 



LETTERS TO THE EDITOR. 

 \Tht 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 oj Nature. 

 No notice is taken of anonymous communications.^ 



Thunderstorms and Sunspots, 



About six years ago Prof, von Bezold laid before the 

 Bavarian Academy a memoir relating to lightning-flashes that 

 had done damage to houses in Bavaria. In that kingdDm the 

 fire-insurance of buildings is entirely in the hands of the State, 

 and a long series of statistical data on the subject was available. 



Two things appeared from this inquiry— first, that those 

 damaging lightning- flashes had enormously increased in the last 

 fifty years (to 1882), much more than the increase of houses ; and 

 second, that there was apparently some relation between the 

 phenomena and the sunspot cycle. To each maximum of sun- 

 spots corresponded a minimum of damaging lightning-flashes or 

 thunderstorms (only in two cases one year displaced) ; but 

 between each pair of minima was another secondary minimum 

 not far from the minimum of sunspots. The curve of lightning 

 damage, in fact, shows a double oscillation for each sunspot 

 period, maxima of sunspots corresponding with the better - 

 defined of the two minima of lightning damage, A somewhat 

 similar result had been arrived at by Prof. Fritz from a study of 

 thunderstorms in the Indian Archipelago, but he considered it 

 adverse to the idea of a causal relation between sunspots and 

 thunderstorms. 



In an earlier paper to the Bavarian Academy (1874), Prof, 

 von Bezold, from a study of several thunderstorm records, came 

 to the conclusion that " high temperatures and a spotless solar 

 surface give years abounding in thunderstorms." This supposed 

 relation between sunspots and thunderstorms does not seem to 

 have attracted much attention of late years. The object of this 

 note is chiefly to show some curves and figures from thunder- 

 storm records, which, it appears to me, yield further evidence of 

 the relation. 



In the diagram herewith are two curves, one for Berlin from 

 1850, the other for Geneva from 1852. The numbers of days 

 of observed thunder are taken and grouped in averages, each 

 yearly point of the curve representing an average of five years. 

 The vertical scale-figures are to the left. Below is an inverted sun- 

 spot curve, with scale-figures to the right. The upper points of 

 the latter are minima, and it will be observed how maxima of 

 the thunderstorm curves occur over them or nearly so ; and 

 similarly with sunspot maxima and minima of the other curves. 

 There is not always exact coincidence, but a very considerable 

 correspondence will be noticed. (I do not here reproduce the 

 figures yielding those curves. ) 



It is to be regretted that the official Greenwich records do not, 

 so far as I know, contain any tabulated series of figures relating 

 to thunderstorms in a long course of yeats. From an examina- 

 tion of the Greenwich Observations and the Weekly Return, I 

 am enabled to present a table of the number of days on which 

 thunder was observed during the sis months April to September 

 in each year from 1850 to 1891. The actual figures are given 

 in one column, and another column gives smoothed values (five 

 year averages). In the curve made from these smooth values, 

 we find maxima corresponding closely with the sunspot minima 

 of 1856 and 1878, and it is now, apparently, near another pro- 



