March 5, 1886.] 



SCIENCE. 



211 



Committees were appointed as follows : Messrs. 

 Dun woody, Meil, and Upton, to prepare forms for 

 records to be used by state services and volunteer 

 observers ; Messrs. Davis, Thomas, Mell, Dun woody, 

 and Woodruff, to report on a system of weather- 

 signals for local display throughout the country ; 

 Messrs. Mendenhall, Fuertes, Dunwoody, Upton, 

 and Payne, to consider plans for a permanent 

 organization of the conference. 



The attendance at the conference represented so 

 many parts of the country, that its recommenda- 

 tions will doubtless have due weight in securing 

 the desirable end of uniform methods of work in 

 the state services now in operation, and in those 

 yet to be formed. Among the members of the signal 

 service, there were present Professor Mendenhall, 

 Lieutenants Dunwoody, Woodruff, Finley,Walshe, 

 and Day, Professors Ferrel, Abbe, Hazen, Russell, 

 and Marvin, and Mr. McAdie. The state services 

 were represented by Professor Thomas of Ohio, 

 Professor Payne of Minnesota, Professor Young 

 of Nevada, Professor Mell of Alabama, Messrs. 

 Henderson and Redding of the bureau of agri- 

 culture, Georgia, Professors Upton and Davis and 

 Messrs. Rotch and Ellsworth of New England, and 

 Professor Huston of Indiana. Professor Fuertes 

 of Cornell university, and Mr. Gillingham of Vir- 

 ginia, volunteer observers of the signal service, 

 were also present. 



The conference adjourned, to meet again at the 

 call of the committee on permanent organization. 



At the meeting of the committee on permanent 

 organization, held after the adjournment of the 

 conference, it was decided to organize under the 

 name of the 'Association of local weather services," 

 and to hold meetings annually in February. The 

 object of the association is to encourage and pro- 

 mote the mutual co-operation of the local weather 

 services and the general weather service of the 

 United States. Its membership is limited to the 

 officers of local services or duly appointed delegates, 

 together with representatives from the chief offices. 



METHOD OF STATING RESULTS OF 

 WATER - ANALYSES. 



The Chemical society of Washington, at the 

 meeting of Nov. 12, 1885, appointed a committee 

 to consider the present state of water-analyses, 

 and to present a method of stating analyses 

 adapted for general use, in order that those here- 

 after published may be readily compared with 

 each other and with future work. This commit- 

 tee reported Feb. 11, 1886, and was authorized to 

 prepare an abstract for publication, in order to 

 call the attention of chemists to the subject. 



The society earnestly recommends the adoption 



of the scheme which is herewith briefly presented. 

 The full text of the report will be published in the 

 next bulletin of the society. 



Water-analyses are usually made to answer one 

 of three questions : viz., 1°, Is the water useful 

 medicinally? 2°, Is it injurious to health? and, 

 3°, Is it suitable for manufacturing purposes? 

 Many books relating to water were published 

 during the eighteenth century, but accurate 

 chemical analysis was not attempted until about 

 18^0. As the earlier analyses w r ere isolated, rare, 

 and made for special purposes, the form of the 

 statement was of little importance, if it was only 

 intelligible. At the present time, however, w^ater- 

 analyses are very numerous. An examination of 

 about a thousand shows some forty-two methods 

 of stating quantitative results, there being some- 

 times three different ratios in the report of one 

 analysis. Such discrepancies render comparisons 

 difficult and laborious. 



The various methods of statement may be 

 classified under the following general forms : — 



1°. Grains per imperial gallon of 10 pounds, or 

 70,000 grains. 



2°. Grains per U. S. or wine gallon of 58.372 + 

 grains. 



3°. Decimally, as parts per 100, 1,000, 100,000, 

 or 1,000,000. 



4°. As so many grams or milligrams per litre. 



The last two would be identical if all waters 

 had the same density ; but as the densities of sea- 

 water, mineral waters, etc., are much above that 

 of pure water, it is plain that the third and fourth 

 modes are not comparable. 



The committee therefore unanimously recom- 

 mends — 



1°. That water-analyses be uniformly reported, 

 according to the decimal system, in parts per 

 million, or milligrams per kilogram, with the 

 temperature stated, and that Clark's scale of 

 degrees of hardness, and all other systems, be 

 abandoned. 



2°. That all analyses be stated in terms of the 

 radicals found. 



3°. That the constituent radicals be arranged in 

 the order of the usual electro-chemical series, the 

 positive radicals first. 



4°. That the combination deemed most prob- 

 able by the chemist should be stated in symbols 

 as well as by name. 



The abandonment of Clark's scale has been 

 recommended by Wanklyn and Chapman ; and 

 the recommendation made by the committee does 

 not involve the disuse of his method, but merely 

 the bringing of it into accord with the decimal 

 system, — the changing from grains per gallon to 

 milligrams per kilogram. 



