EDITOEIAL. 503 



even the best ordered library must speedily degenerate into chaos. 

 The station library in particular, irrespective of such administrative 

 details as its organization as a part of the college library, its mainte- 

 nance as a separate entity, or its partition into departmental collec- 

 tions, needs careful supervision because of its peculiar nature. 

 Merely from the business standpoint, now that the stations are ex- 

 pending, as in recent years, an average of about five hundred dollars 

 annually for books, periodicals and binding, it is easy to see that in 

 even a few years the library comes to represent a considerable invest- 

 ment, justifying the careful preservation of its peculiarly perishable 

 material and its housing with greater regard to the hazard from fire 

 and other losses than it commonly receives. 



With a large and ever-increasing library of this sort in constant 

 use, there must be regular and systematic attention. The conditions 

 render desirable the employment of a librarian, trained if possible. 

 The usefulness of the library will often depend very largely upon 

 this one factor. Even in the acquisition of the material the services 

 of some one who can make it his special business is well nigh indis- 

 pensable. It is doubtful if there is yet a full realization of the im- 

 portance of securing and preserving complete files of the publications 

 which constitute the bulk of the station library. In the case of public 

 documents the impression is still widespread that such documents if 

 lost can be easily replaced, and in consequence too often their acquisi- 

 tion is neglected, or once obtained they are not properly cared for. 

 This is evident from the conditions frequentl}^ found and from the 

 many requests for back numbers which come to this Department. 



As a matter of fact it is well known to all who have had occasion to 

 attempt such work that the replacing of these documents is not only 

 difficult but in many cases well nigh impossible. Editions are small 

 and soon exhausted ; and such documents are not as easily acquired 

 through book dealers as most other kinds of publications. 



Thus in spite of very great efforts by this Office and others to secure 

 an absolutely complete set of the publications of the experiment 

 stations in this country, it is improbable that any such set exists or 

 can ever be brought together. A similar condition exists with refer- 

 ence to the publications of this Department. In both cases there are 

 sets substantially complete, but those in the possession of many col- 

 leges and stations are quite far from it. Of necessity in dealing with 

 publications issued at irregular intervals and oftentimes in unnum- 

 bered series and even undated, there must be constant watchfulness 

 and persistence to obtain material and preserve it. The current lists 

 of publications of the stations and this Department now make it 

 feasible to follow this matter systematically. 



When the library has installed at its head an assiduous collector 

 and an efficient caretaker there has been a long step forward, but the 



