and Louisiana State University some years earlier. They were again emphasized by Dr. A. C. True in 1912. Where they prevail, the library is the actual center of the intellectual life of the institution.

The data collected in the survey were used to ascertain in so far as possible (1) the extent to which the libraries of the land-grant colleges are meeting the requirements mentioned, and (2) recommended methods by which the requirements may be met. In the following section of this report the actual use made of libraries at the present time is given first consideration as a necessary introduction to successive studies of books, buildings, administrative control, personnel, and financial support. The term instructor is used in this report to signify any member of the instructing staff.

Chapter II.-Usability of Libraries

Use of Books the Sole Purpose of Libraries

The efficiency of a library is measured by the service which it renders to faculty and students. Librarians understand—what was not understood 50 years ago that large book collections, wellarranged buildings, scholarly and willing personnel are essential solely in order that books may be used. Measurements of the actual use of the libraries included in this survey are statistically possible. Such measurements as were reported or otherwise ascertained provide the basis for the conclusions reached and the recommendations made for improvement of service. The measurements are considered in succeeding paragraphs in terms of (1) proportion of students using the library daily, (2) loans for home reading, (3) books for assigned reading, (4) number of interlibrary loans, and (5) seating capacity and utilization of seats.

Measurements of Use

Proportion of students using the library.-The number of individuals who use the library is an indication of the quantity and perhaps also of the quality of service. On account of the work involved actual count was not requested in the survey. A few libraries, however, reported statistics. Excluding students who come for classes or for "dating" parties, one librarian reported that the average student used the library four times a week. Another institution reported an average of six times per week. These figures can be used by any institution as a basis for comparison with the attendance in its own library. The two reporting libraries represent a high standard of use, judged by the figures reported of books loaned for home reading and use of seating capacity, both of which are far above average. The figures, therefore, may be taken as a tentative standard. It should be noted in addition that these two libraries are open 13 and 14 hours a day, respectively. A personal examination of some other libraries justifies the statement that the attendance of students at these libraries is generally far less than that of the two reporting libraries. This conclusion is further strengthened by the fact that these other libraries are open 7 hours per day, instead of 14, and are closed during most or all of the evening hours when students are most free to use books.

Number of books borrowed for home reading, not overnight use.— Data were requested as to the number of books loaned for reading outside of the library building, excluding books borrowed solely for overnight use. In order to obtain figures which would be somewhat comparable for the various institutions, the number of books loaned at each reporting institution was divided by the number of full-time students at the institution. The results are given in Table 1. The figures obtained are, it is true, a general rather than an exact criterion of the use of the libraries and of the reading by students. Not all books loaned by a library are loaned to students. Nevertheless, as the basis is the same for all institutions, the results have certain significance, especially in the extreme cases of loans less than an average of 7 a year by each student or greater than 15, figured on the basis of the total number of books loaned divided by the total number of full-time students. The average number of books loaned each student by the various institutions varied from 3 to 33 a year. The average was 12.

TABLE 1.-Total number of books loaned per year, excluding overnight loans, divided by number of students enrolled1 as of October 31, 1927

1 The number of students was taken from U, S, Bureau of Education Bulletin, 1929, No. 13, p. 30.

A comparison of financial expenditures for libraries with the number of books loaned for home reading was made in order to determine if any relationship existed. The expenditures were reduced to an average for each student in order to form a basis for comparison. Nine of the 14 institutions that expended annually less than $12 per student for library purposes showed an average loan of 81% volumes per student per year. The remaining 5 could not report the number of such loans. The average number of loans to the average student by 6 institutions-Massachusetts Agricultural College, University of Hawaii, Iowa State College, Universities of Tennessee, Wyoming, and Illinois-which reported annual library expenditures of more than $20 per student was 20 volumes per annum. In certain of these institutions the average student withdrew from five to ten times as many books a year as is the case of the group with the lowest expenditures. Two of the three institutions with the lowest average number of loans for each student were also the lowest in financial support, although other factors than expenditures influence the use of the library by students, it is apparent that the amount of expenditure bears some relationship to use.

Too definite conclusions can not be drawn solely from the number of books loaned. Several factors may affect such loans. A large number of departmental libraries may lessen the withdrawal of books for home use. Also, it may be assumed that inadequate library buildings with limited seating capacity would increase the number of books borrowed, while on the other hand adequate seating facilities would lessen the necessity of withdrawal of books for use outside the building. This supposition does not seem to hold true. Five libraries with known inadequate facilities as to library buildings and also with limited financial support showed an average loan of fewer than 7 volumes per student as compared with an average of more than 12 for the whole group of land-grant colleges. It is apparent that inadequate buildings, as well as lack of financial support, are associated with relatively slight use of books. On the other hand, four of the six institutions mentioned above, with loans per student much above the average, have recently erected new buildings.

Amount of assigned reading.-The number of books withheld from circulation for assigned reading in the library may provide a rough indication of the use of books in a library. It might be supposed that a large number would cause a decrease in the loan of books for home use. The opposite is true. In the six institutions previously mentioned with annual expenditures of more than $20 per student and with average loans per student of 20 volumes per annum, an average of 90 books was on reserve for each 100 students. In the eight institutions with annual expenditures less than $11 per student and with average loans of 7% volumes per student, 60 books were on

reserve for each 100 students. Indeed, if Kansas State Agricultural College, which seems to have developed extensive use of its library with limited financial resources is excepted, the books in the assigned reading room for the second group is fewer than 50 per 100 students— not much more than one-half as many as in the case of the libraries in the first group. Libraries, in general, with loans of books above average have also a greater number of books available for assigned reading per student than do the libraries with loans below average. Use shown by books borrowed from other libraries.-The number of books borrowed from other institutions also gives some indication of the use of library material in a given institution. A library with inadequate collections would ordinarily, it would be supposed, need to borrow much more material from other libraries than would institutions with strong collections. This supposition also proves to be contrary to the facts. Institutions borrowing more than 200 volumes each in 1927-28 from other libraries were University of Illinois, Iowa and Kansas State Colleges, Massachusetts Institute of Technology, Universities of Minnesota, Missouri, New Hampshire, and Ohio State. These are chiefly institutions, in so far as the libraries reported figures, with the stronger collections and with use of their own collections generally much above average. On the other hand, the seven colleges which reported loans much below average and also with book resources known to be most limited borrowed an average of eight books per year from other institutions. Again it is apparent that faculties in this latter group are not using published material as are the faculties in the first group, even after allowances are made for the size of the institutions.

The inference seems logical that a number of libraries are not supplying material which is demanded of libraries in other land-grant institutions. Possibly the faculties in the institutions of the second group have become discouraged in their attempts to obtain material, or possibly they are content with poor library service and lack of material. If the latter is the case, the quality of the instruction may be open to question.

Ratio between seating capacity of library and number of students— Exhaustion of seating capacity.-Some indication of the use of libraries may be obtained from the number of available seats in proportion to the number of students, compared with reports as to whether seating capacity is sufficient. The University of Florida, with one seat for every five students, reports crowding at certain periods, especially in the evening. Iowa State College, with a ratio of one seat for every eight students enrolled (560 seats for 4,200 students), reports all seats filled at times. Oregon Agricultural College, with the same percentage (480 seats for 3,780 students), also reports seating capacity not sufficient at peak times; University of