Maech 30, 1917] 



SCIENCE 



315 



and Proper Motions: 0. D. Perrine, Obser- 

 vatorio JSTacional Argentine, Cordoba. 



Hydrology of the Isthmus of Panama: 

 Brig. Gen. Henry L. Abbot, United States 

 Army, retired. Extensive tables for rainfall, 

 outflow, evaporation, etc., are given and dis- 

 cussed. 



The Meteor System of Pons-Winneclce's 

 Comet: Charles P. Olivier, Leander McCor- 

 mick Observatory, University of Virginia. 

 The elements of the meteor's orbit are deter- 

 mined from more than 1,000 observations. 



Improvements in Calorimetric C omhustion, 

 and the Heat of C omhustion of Toluene: 

 Theodore W. Richards and Harold S. Davis, 

 Wolcott Gibbs Memorial Laboratory, Har- 

 vard University. The improvements are: 

 Means of effectively closing the bomb with 

 less risk to the lining and cover; means of 

 burning volatile liquids without loss ; a method 

 of automatically controlling the temperature 

 of the environment; means of evaluating the 

 incompleteness of combustion. The heat of 

 combustion of toulene is determined as 10,155 

 calories (18°) per gram. 



The Mass of the Electric Carrier in Copper, 

 Silver and Aluminium: Richard C. Tolman 

 and T. Dale Stewart. A continuation of ex- 

 periments on currents produced by aecelera- 

 . tion in metals. 



The Silver Voltameter as an International 

 Standard for the Measurement of Electric 

 Current: E. B. Rosa and G. W. Vinal, U. S. 

 Bureau of Standards, Washington, D. C. A 

 s urmn ary of eight years' experimental work 

 which has shown how the voltameter can be 

 used as a reliable current standard and as a 

 means of checking the constancy of the value 

 of the Weston normal cell. 



Edwin Bidwell Wilson 

 Massachusetts Institute of Technology, 

 Cambridge, Mass. 



SUGGESTIONS FOR THE DEVELOP- 

 MENT OF SCIENTIFIC LIBRARIES; 

 WITH SPECIAL REFERENCE TO 

 AUTHORS' SEPARATES 

 The communication from Tracy I. Storer 

 in Science for November 24, on the care of 



pamphlet collections, brings up a matter which 

 has interested me for several years. It has 

 been my task to rearrange a few scientific li- 

 braries and my privilege to consult several 

 others, and I have found in nearly all of them 

 considerable room for improvement, especially 

 in the method of handling authors' separates 

 and other pamphlets. 



It seems to be a common, if not the prevail- 

 ing, custom in private and semi-private scien- 

 tific libraries to arrange pamphlets alphabet- 

 ically by authors, as Mr. Storer recommends. 

 This has the advantage of obviating the men- 

 tal exertion of classifying them by subjects 

 (which ought to be an important considera- 

 tion with that apparently increasing class of 

 persons who prefer to follow a mechanical 

 routine rather than exercise judgment) and of 

 keeping together the writings of one's friends, 

 so that if a friend comes for a visit one can 

 see at a glance just how many of his papers 

 one has. But in most other respects the alpha- 

 betical arrangement is an undesirable expedi- 

 ent. For there is no important difference be- 

 tween a pamphlet and a book, and no good rea- 

 son for putting a pamphlet on a different shelf 

 from a book on the same subject (unless of 

 course there happens to be a considerable dif- 

 ference in size of page or the pamphlet be- 

 longs to a series of bulletins). 



Large libraries use some sort of subject 

 classification, and this is equally desirable for 

 small ones. When one is carrying on a par- 

 ticular line of research one likes to have all 

 the literature bearing on it together as far as 

 possible. It is out of the question to keep in 

 mind every one who has written on a given 

 subject, and unless a library is pretty fully 

 catalogued (which is not usually the case vsdth 

 private working libraries), some pamphlets are 

 pretty sure to be overlooked if they are not 

 classified by subjects. Another objection to the 

 alphabetical arrangement is that every scien- 

 tist receives many pamphlets on subjects that 

 he is not particularly interested in,^ and with 

 any other system than a subject classification 

 these will always be in the way, unless they 

 are relegated to a special " limbo." 



1 See Torreya, 16, 101-102, April, 1916. 



