226 



NATURE 



[July 6, 1882 



have always been the aggressors, where the dark popula- 

 tions have always represented the passive or recipient 

 element. On the other hand, wherever it has escaped 

 from Malayo-Polynesian influences, or wherever it has 

 been able to preserve its original speech in spite of those 

 influences, this dark element will certainly be found 

 speaking languages organically distinct from the Malayo- 

 Polynesian. Mr. Man's recently published account of 

 the Andamanese dialects shows that they differ in their 

 morphology, in their glottology — in fact, in every respect, 

 from those of Malaysia. 



Mr. Lawes makes the same remark respecting the 

 Koiari people, who occupy the highlands back of 

 Moresby Bay in South-East New Guinea. And the 

 authors of the work under review now find that the Mafor 

 of Geelvinck Bay betrays, with many striking resemblances 

 to the Malayo-Polynesian, " an astounding peculiarity of 

 structure." 1 The " resemblances " are of a verbal cha- 

 racter, due to known contact with the Malays, who have 

 long frequented the waters along the north-west coast of 

 New Guinea. The " peculiarity of structure," involving 

 root modifications and something even approaching to 

 inflection (" Quasiflexion "), as understood in the Aryan 

 family, belongs to the organic Papuan linguistic type. 

 This type is thus demonstrated to be fundamentally dis- 

 tinct from the Malayo-Polynesian, which shows no trace 

 of these peculiarities. And thus also disappears the 

 fancied antagonism hitherto supposed to exist between 

 the linguistic and anthropological elements in the Oceanic 

 regions. A. H. Keane 



THE SOLAR-COMMERCIAL CYCLE 



IN an article printed in Nature (vol. xix., pp. 588-90) 

 I gave a table of the prices of wheat at Delhi, from 

 1763 to 1835, quoted, or rather calculated from data given 

 in a brief paper of the Rev. Robert Everest, contained in 

 the Journal of the (London) Statistical Society for 1843, 

 vol. vi. pp. 246-8. Between the years 1763 and 1803 there 

 was evidence of wonderful periodicity in the recurrent 

 famine and abundance at that part of India. When 

 recently engaged in examining more minutely the relation 

 between these prices and the variations of solar activity, 

 as indicated by Prof. Wolf's numbers, it has occurred to 

 me that an inference may be drawn which I overlooked 

 on the previous occasion. 



In the accompanying diagram I have exhibited the 

 prices in question together with Wolf's numbers as stated 

 in the Monthly Notices of the Royal Ast. Soc. vol. xxi. 

 pp. 77, 78. I have also indicated the dates of the Com- 

 mercial Crises of the time according to the article on the 

 subject in Mr. H. D. Macleod's " Dictionary of Political 

 Economy," vol. i. pp. 627-8. It need hardly be said that 

 the coincidence between the three classes of recurrent 

 phenomena is of a very remarkable character, and goes 

 far in supporting the relation of cause and effect which I 

 had inferred to exist, both on empirical grounds and from 

 the well-known fact that it is the cheapness of food in 

 India, which to a great extent governs the export trade 

 from England to India. But although the coincidence of 

 commercial Crises in Western Europe with high corn 

 prices at Delhi is almost perfect, it will be noticed that 

 after 1790, the correspondence of the solar curve with 

 that of prices is broken. Wolf does not recognise the 

 existence of any sun-spot maximum between 17S8 and 

 1804, and he believes that there wasa minimum at 1798. 

 According to Wolf's later researches {Memoirs Roy. Ast. 

 Soc., vol. xliii. p. 302), these dates are respectively, maxi- 

 mum 1788-1, minimum 1798-3, and maximum, 1804-2. 



But now arises the question to which I wish to draw 

 attention. If the eleven-year solar periodicity was really 

 interrupted in this long interval of i6'i years, how comes 



■ "("e-adedas Mafoor'sche aber wird in seinem Baue bei manchen auf- 

 falligen Aenlichkeiten eine erstaunliche Eigenthiimlichkeit im Bildungs- 

 prinzipe aufweisen," p. 4. 



it that the meteorological periodicity, as manifested in 

 the corn prices at Delhi, was not interrupted. It is true 

 that the price maximum of 1803 was a comparatively 

 small one ; but this was quite to be expected, considering 

 that if there were an intervening solar maximum, it must 

 have been a small one. May we not reverse the argu- 

 ment and infer that the evident relation between the 

 previous sun-spot maxima and the succeeding scarcities 

 at Delhi, would lead us to expect a minor solar maximum 

 about the year 1797 ? 



Standing alone, the presumption thus created would, 

 doubtless, be of a somewhat slight character. But it is 

 in the first place well known, that the data upon which 

 Wolf based his numbers about this time, are less conclu- 

 sive than in other parts of his series. His results, too, 

 from 1801 to 1807 are expressly marked as doubtful, so 

 that extrinsic information which might have little weight 

 where there was abundance of reliable solar or magnetic 

 observations may come in very usefully where doubts 

 already exist. Now it happens that the late Mr. J. A. 

 Broun inquired very carefully into the facts known about 

 the solar variation at this time, his results being given in 

 the Transactions of the Royal Society of Edinburgh, vol. 

 xxvii. pp. 563-594, and in his article printed in Nature 

 (vol. xvi. pp. 62-64). Broun inferred from the observa- 

 tions of Gilpin, and from other data, that there was a 

 small maximum about 1 797, and that there were grounds 

 for believing that the subsequent maximum " may really 

 have occurred after 1806, when Gilpin's series terminated." 

 Now, what Broun deduced from totally different data, is 

 exactly what we should infer from the Delhi prices. If 

 we are to believe that Indian meteorology depends upon 

 solar variations, then it almost follows that there was a 

 solar maximum about 1797. The consequence of this 

 inference, however, is very important, because it goes to 

 support the views of Lamont, Broun and others, that the 

 solar period is about io£ (10-45) years and not in as 

 calculated by Wolf. It should also be pointed out that 

 the temperature observations of Prof. Piazzi Smyth lead 

 to a like result. The epochs of the heat waves are, accord- 

 ing to him (Nature, vol. xxi., p. 248), 1826-5, 1834-5, 

 1846-4, 1857-9, and 1868-8, giving an average interval of 

 1057 years. 



I may take this opportunity of asserting that the 

 progress of events confirms belief in the eastern origin 

 of the great commercial Crises. 1 In his important work, 

 the " Prdcis du Cours d'Economie Politique" (vol. i. pp. 

 604-5), M. Cauwes while partially accepting the doctrine 

 of periodicity criticises the particular views here advo- 

 cated. He says : — 



" Depuis longtemps les economistes ont signale" la 

 pe'riodicite' de ces evolutions : MM. Juglar et Jevons pre'- 

 tendent meme pouvoir la calculer d'une maniere precise. 

 Selon M. Jevons, l'ensemble des phenomcnes serait 

 renferme' dans tin cycle de dix anne"es et demie. De fait, 

 les grandes crises e"conomiques du siecle (1806, 1S17, 

 1825-7, 1836-37, 1847, 1857,) s'echelonnent a dix annees 

 d'intervalle ou a peu pres, mais les dernieres, 1866 et 

 1873, seraient venues un peu avant l'heure, et celle de 

 1873 s'est prolonged au dela de toute attente." M. 

 Cauwes in short accepts the six earliest crises of this 

 century as sufficiently agreeing with the theory. The 

 crisis of 1866 no doubt came about a year before it would 

 be expected, which is a divergence of reasonable amount. 

 The year 1S73, however, is one which it would be impos- 

 sible to introduce into the series. Now there doubtless 

 were both in America and England in that year, a state 

 of commercial stringency, a relapse of prices and other 

 disturbances which might be mistaken for the signs of a 



1 As it is impossible to reproduce the explanations and qualifications con- 

 tained in the article quoted above, or that at pp. 33-37 of the same volume of 

 [ii ( v "1. xix.), it is assumed that this article is read subject to those 

 qualifications and explanations. In p. 588 col. /' of the same volume, a seer 

 of wheat was by a typographical oversight stated to be equal to 21 lbs. instead 

 of the true weight 2 lbs. 



