smaller collections, it has been increased by annual purchases of considerable size, selected with discrimination from the libraries of some of the world's greatest botanists, as these, through the death of their owners, have come into the market, until it now comprises over 16,000 volumes and 20,000 pamphlets, valued at over $60,000 and fairly symmetrical in all fields of botany and the sciences that must be taken into consideration in botanical work, as well as in gardening, landscape work, horticulture, forestry, greenhouse construction, and the like. Large as it is, however, it is so far from being satisfactorily complete for the uses it is put to that a sum greater than its present valuation could be spent on it within a few years, if the money were available and the works needed were in the market, without having even then a surfeit of the good things that such a library, destined for research purposes, should offer the busy man who, to use them, must be freed from too great care and delay in searching them out and placing his hand on their contents. Strongest in the library, as would be expected from the fact that the Garden is a garden and possesses a large herbarium and is at present more concerned with the systematic study of flowering plants than with other subjects, are the departments devoted to floras and monographs; those difficult things to have at hand, series of journals and proceedings in whole or in part devoted to botany; and the compendiums of gardening and garden plants; but there are in this country few fuller collections of treatises on plant morphology and physiology, and the works on the ecology of the flower are probably nowhere surpassed in completeness of representation. From the first establishment of the school of botany in the university, it has been the policy to spare no trouble or reasonable expense to make the library as complete as possible in any field in which special work is taken up, and the result is that each piece of research accomplished has been marked by a corresponding growth in the library.
A herbarium is an uninteresting collection to the average person who does not need to use it, whether he be a botanist or not. In envelopes, or glued or bandaged down on sheets of paper are thousands of more or less fragmentary plants, sometimes moldy or worm-eaten, for time works havoc with all organic matter, and usually much faded. And yet not even the library is more indispensable for the worker on the species that they represent; for they are the real plants, and not some one's interpretation of them. The choicest part consists always of the original specimens preserved by an author when describing and naming a species or genus, for however his description or figure may have erred, this type persists as a record showing the true generic and specific characters. An herbarium is not always a conclusive source to which to turn for final information, for the author may have and
