November 3, 1904J 



NA TURE 



cooperation of the various railway companies, every ; 

 intelligent person in the district was made aware of ; 

 the impending- danger in ample time to make such j 

 preparations as they were able. _ | 



The floods of 1903 owed their inception to a series '< 

 of heavy rainfalls caused by a succession of storms 

 of the south-western type, the best rain-producing 

 quarter, coming on the top of the water derived j 

 from the melting of the snow on the mountains in the 

 upper reaches. | 



In the February flood in the lower Mississippi the 

 water rose in one long swell from Cairo to the Gulf of | 

 Mexico from 17.5 feet on the gauge on January 28, , 

 passing the danger point of 45 feet thirty-nine days 

 later, and 505 feet, or 55 feet above the top of the 

 banks, eight days afterwards. It remained above the \ 

 danger line for another twelve days, and then began 

 to fall. It will thus be seen that the water in the 

 river during the flood rose 33 feet. 



Although excessive rainfall was the original cause 

 of these floods, the effect was greatly increased by 

 works that had been carried out for the improvement 

 of the river and for providing means of inland trans- 

 port, necessitating the frequent crossing of the river by 

 railway bridges. Formerly a certain amount of relief 

 to the floods was afforded by the water flowing through 

 the numerous crevasses or breaches of the banks that 

 occurred, but during recent years the banks have been 

 systematically raised and strengthened. For example, 

 in the St. Francis system the levees have been extended 

 and raised 2 feet over a length of 173 miles, and the 

 area originally subject to being submerged reduced 

 4000 square miles. The same operations have been 

 carried on in other districts, so that the flooded area 

 which previous to 1897 extended over 30,000 square 

 miles in 1903 barely reached 7000 square miles. The 

 fight against this flood was also the most extensive 

 and persistent ever attempted in the history of levee 

 engineering. When a breach was likely to occur all 

 the help and material available was concentrated at 

 the point of greatest weakness. Kt one place a force 

 of more than 1000 men was employed both day and 

 night, in spite of which the bank gave way for more 

 than a mile. 



At another part of the river, about 36 miles below 

 New Orleans, a crevasse occurred at a place where 

 the river is 120 feet deep. The bank was all washed 

 away, and where it formerly stood a hole was scoured 

 out 60 feet deep. Owing to the precautions taken, 

 due to the warnings of the Weather Bureau, provision 

 had been made to meet such a catastrophe, and work- 

 men were at once concentrated on the spot, and train- 

 loads of material which had been provided in readi- 

 ness for such an emergency were brought to the place. 

 By this means the breach was successfully closed, and 

 the flooding of some of the finest sugar plantations in 

 ' Louisiana averted. 



Other causes that contributed to the greater rise of 

 the flood were the numerous railway bridges that had 

 been carried across the river without leaving sufficient 

 waterway for floods. In one place, where the natural 

 width of the river is 900 feet, the waterway had been 

 contracted to 400 feet by a railway bridge, the velocity 

 of the water through which rose to twelve miles an 

 hour. 



Encroachments by reclamation have also materially 

 interfered with the free flow of the river, the original 

 width of the channel in some places having been re- 

 duced one-half. 



The report of these floods contains numerous illus- 

 trations which give a very graphic idea of the ruin 

 caused in the flooded areas, and also of the works 

 carried on in repairing the levees. There is also a 

 map of the watershed of the Mississippi and of the 



NO. 1827, VOL. 71] 



flooded areas, and of the rainfall in the different 

 districts. 



Two other volumes issued by the Geological Depart- 

 ment relate to the floods of the river Passaic in 1902 

 and 1903, when the loss to the inhabitants of the 

 district was estimated for the two floods at about 3 

 million pounds. These two volumes also contain 

 numerous verv telling illustrations of the flooded areas 

 and of the damage done to houses and factories.' 



VJHA.T IS BRANDY? 



THIS question, which a few months ago greatly 

 exercised analytical chemists in this country in 

 consequence of the action of certain local authorities 

 under the Sale of Food and Drugs Acts, has recently 

 engaged the attention of the Technical Committee of 

 Oinology, instituted by the French Minister of Com- 

 merce by decree of March 22, 1904, and the committee 

 have adopted the conclusions of M. Rocques, the re- 

 porter of the subcommittee charged with the consider- 

 ation of the matter, whose report is published in extcnso 

 in the Moniteur Officiel du Commerce of June 30. In 

 view of the importance of the subject, it may be de- 

 sirable to give a short summary of the facts and argu- 

 ments which led the technical comrnittee to adopt the 

 conclusions of the special subcommittee. 



In the first place the committee, for reasons which 

 it is unnecessary to explain, object to the term 

 coefficient of impurities, hitherto employed by French 

 chemists, in conformity with a decree of the Minister 

 of Commerce of Mav 26, 1903, to designate the aggre- 

 gate proportion of 'the substances other than ethyhc 

 alcohol in brandy, and prefer to denote it by the term 

 coefficient non-alcohol, or more simply non-alcohol, by 

 which is to be understood the sum of the different 

 volatile substances, other than ethylic alcohol, ex- 

 pressed in grams per hectolitre of absolute alcohol. 

 These substances are the acids, aldehydes, ethers, the 

 alcohols higher in the homologous series than ethyl 

 alcohol, and the furfurol. 



The causes which influence this coefficient are many, 

 but in the main they may be said to depend upon 

 (i) the nature of the wine, (2) the method of distil- 

 lation, and (3) age. 



As regards the first cause, it is found that the pro- 

 portion, "as well as the character, of the volatile matters 

 vary according to the origin of the wine, the conditions 

 under which its fermentation has been effected, the 

 manner in which it has been kept, &c. The proportion 

 of acids and ethers is considerably augmented if the 

 wine becomes sour, and, speaking generally, the pro- 

 portion of aldehydes is higher in white than m red 

 wines. 



But it is mainly in the method of distillation that \ve 

 are to seek for the cause of the wide variations in this 

 coeflicient. This is readily understood if we examine 

 the manner in which the various substances, which 

 together constitute non-alcohol, behave during distil- 

 lation. It is known that these substances pass over 

 in very different proportion in the course of the distil- 

 lation. Thus the aldehvde and the more volatile ethers 

 are found mainlv in the first runnings {produits de 

 tete), whereas the taillings (produits de queue) contain 

 in largest quantitv the higher alcohols and the furfurol. 



The separation' of these various products— the pro- 

 duits de tete, the alcohol itself (de coeur), and the 

 produits de queue— is effected in a manner more or 

 less complete, depending upon the apparatus employed. 

 In the larger distilleries this apparatus is of a very 

 high order of perfection. But without further labour- 



1 The Passaic Flood of igo2. Water Supply and ImKation Paper 

 No. 88, and of 1903, Paper 92. (Washington : Government Printing Office. 



