58 



NA TURE 



[Nov. 20, 1! 



observations 1S13 to 18S2. In selecting the years, those 

 only have been chosen in which the aggregate fall of rain 

 has been at leist 20 per cent, less than the average. The 

 table further gives the total fall and difference from the 

 average for the first ten months in each of these years, 

 and in the last column will be found the number of 

 months during which the rainfall has been deficient. 



From the first two columns it appears that the years 

 1S58 and 1864 claim the distinction of being the driest of 

 all, the total falls being only 173 inches and 174 inches 

 respectively, or 30 percent, less than the average. Next 

 comes 1S47, with a total fall of 177 inches and a deficit 

 amounting to 29 per cent. As regards the period of ten 

 months, the present year has been drier than any of the 

 past seventy-one, but in the year 1S47 the rainfall was 

 nearly as deficient. In the case of the other dry years the 

 aggregate fall for the ten months was at least an inch more 

 than in either 1847 or 18S4, and in the years 1832 and 

 1 837 it was three inches more. On comparing the returns 

 for the past seventy-one years, one more striking fact is 

 brought to light. Out of the whole series there has been 

 only one occasion on which the deficiency of rain has 

 continued through a greater number of months than it has 

 this year. This long period of drought commenced in 

 November 1846 and continued until November 1847, and 

 there were consequently no fewer than thirteen consecutive 

 months during which the rainfall in London was below 

 the average. Fredk. J. Brodie 



ANCIENT CHINESE GEOGRAPHY 

 ■jVJ OT long since the Chinese Ambassador to England, 

 -1- ' in the course of a remarkable speech at Folkestone, 

 twitted European scholars with the labours which they 

 freely bestowed on the study of extinct nations and races, 

 while the still existing civilisation of China, hardly infe- 

 rior in antiquity to that of any other race, received but 

 scant attention. Whether the charge is well founded or 

 not we cannot pretend to decide here ; but there is, we 

 believe, no doubt that there is still in Chinese literature a 

 vast mine, into which but few and trifling shafts have been 

 sunk. The wealth of the geographical literature of China, 

 for instance, is known to but a few scholars, and one of 

 these, M. de Rosny of Paris, in a work recently published 

 on the Oriental nations known to the ancient Chinese, 

 says that, among all the literatures of the East, that of 

 the Chinese probably contains the most valuable informa- 

 tion for the study of Asiatic ethnography, for a crowd of 

 nations which have disappeared, or which are unknown in 

 Europe, have been the subject of substantial notices by the 

 Chinese, outside which, probably, we know nothing of their 

 political history or of the annals of their civilisation. M. de 

 Rosny's work, which is published by the Ethnographical 

 Society of Paris, is devoted to the translation and piecing 

 together of extracts from old topographical works respect- 

 ing various countries known to the Chinese in ancient 

 times. Much of the labour in a work of this kind must 

 necessarily be devoted to identifying the places men- 

 tioned. In many cases this has not even now been satis- 

 factorily done. Thus, the origin of the name Ta-tsin, 

 applied to the Roman Empire, is wrapped in obscurity. 

 The latest theory is that it is the phonetical representa- 

 tion of Tarsus in Cilicia, whence Antoninus sent ambas- 

 sadors to Bactria, so that the name of Tarsus was the 

 first echo which China received of Rome. But although 

 there is much in M. de Rosny's volume which can only 

 interest the technical Sinologue, yet one can gather from 

 the text, as well as from the maps, a fairly accurate idea 

 of the knowledge of geography possessed by the Chinese 

 in early times. Of the maps, which are nine in number, 

 one contains the Indian Archipelago as known to the 

 Chinese, and six others Indo-China and Malaysia, ac- 

 cording to Chinese geographers, at various periods from 

 the twelfth century before our era down to 906 after Christ. 



The Chinese, then, according to M. de Rosny, have 

 from the most remote times occupied themselves with the 

 topography of the districts through which they migrated, 

 and have studied the geography of the neighbouring 

 countries. Yu the Great, who reigned in the basin of the 

 Yellow River twenty-two centuries before our era, was a 

 veritable geographer. The Shu-king, which contains 

 an account of the public works executed under his direc- 

 tion, contains the first rudiments of Chinese ethn graphy, 

 as Genesis does that of the Jews. A geographical work 

 which is probably not less ancient is the Shau-hai-king. 

 It is at least as old as the Choo dynasty — 1134 B.C. — and 

 some Chinese authors even carry its date back to the 

 twenty-seventh century before Christ. In a book of rites 

 of the Choo dynasty just referred to, it is stated that 

 twenty-four officials were specially charged with the ad- 

 ministration of a department for national geography. 

 It is, however, to the historians that we have to look 

 for accounts of the various peoples which early sub- 

 mitted to the preponderating influence of the Middle 

 Kingdom. The nomad hordes of the north and west, and 

 the States then in process of formation in the south, all 

 entered into relations with the Chinese. The ambassadors 

 whom they sent to the Court brought with them informa- 

 tion as to the people they represented, which was duly 

 consigned to the archives of the Empire by its historio- 

 graphers. The officials sent by the Chinese in return to 

 the peoples about them contributed their quota of geo- 

 graphical and ethnographical facts, until ultimately the 

 documents on the subject became so numerous that 

 native scholars judged it well to summarise them into one 

 great work. It was thus that the great encyclopaedia 

 associated with the name of Ma-touan-lin was formed. 

 Its first publication was in 1322. 



The limits of the world as known to the early Chinese 

 are stated by M. de Rosny to be : in the north, Southern 

 Siberia and Kamchatka ; in the east, the Kuri'.e Islands, 

 Japan, the Loochoo Archipelago, and that of the 

 Philippines ; in the south-east, Borneo and Celebes ; 

 to the south, Java, Sumatra, and Ceylon ; to the west, 

 Arabia, Persia, and the States bordering on the Caspian. 

 Some scholars have professed to discover the Roman 

 Empire under the name Ta-tsin, and America, which a 

 mission of Shamans are said to have discovered in the 

 fifth century, under that of Fousang. In the work before 

 us the writer gives, from Ma-touan-lin and other sources, 

 the statements of the early Chinese writers with regard 

 to the various races inhabiting these regions ; but he 

 warns us more than once that these ancient documents, 

 though of great value in teaching us about peoples little 

 known to us, must be used with the utmost reserve, and 

 only after undergoing a searching examination and criti- 

 cism. The present instalment of the work deals only 

 with the races to the south, south-east, and east, such as 

 the Japanese, Ainos, Siamese, &c. Its value as an 

 ethnographical and geographical work can only be known 

 to the one or two living Europeans who have made a 

 special study of the subject ; but it places beyond doubt 

 the fact that students of the ethnography and historical 

 geography of the Far East will have to reckon with the 

 works of their remote Chinese predecessors before their 

 knowledge can be regarded as complete. 



COLOUR 

 1\T M. A. ROSENSTIEHL has made an interest- 

 ■'■*-'■ • ing contribution to the science of colour in 

 the form of a brochure recently published under the 

 auspices of the Societe Industrielle of Rouen, and entitled 

 " Les premiers -Jeinents de la Science de la Couleur." 

 In this treatise, which is a model of brevity and of de- 

 monstrative clearness, the author shows that the empirical 

 methods which have hitherto prevailed amongst colourists 

 of all classes are radically imperfect. These methods are 



