These experiments are originally described in Circular No. 15 and other publications of the U. S. Forestry Division, also in "Materials of Construction," by Professor J. B. Johnson.

(2) Experiments conducted for the Tenth U. S. Census by Mr. J. P. Sharpless at the Watertown (Mass.) Arsenal upon specimens botanically selected by Professor Sargent. These experiments are less complete in detail, and averages are based upon infinitely [p005] fewer tests for each species. So far as known most specimens were from butts. Nothing is known of moisture conditions save that specimens were "carefully seasoned." Tests were upon about twelve hundred specimens divided over four hundred and twelve species, allowing but a small number for each. The series is most valuable in that the species attempted were so numerous as to present an almost complete American series; in that the botanical identity of the specimens was beyond question, and because it gives a general idea of relative values. The results are frequently quoted and appear on the accompanying pages in spaces immediately following those occupied by, or set apart for, "Forestry" figures or their alternates. The tests are characterized as follows:

They are originally described in Vol. IX, Tenth U. S. Census; Executive Document No. 5, Forty-eighth Congress, First Session; in Catalogue of the "Jesup Collection," by Professor C. S. Sargent, and elsewhere.

(3) Experiments conducted upon full-sized pieces. The most reliable investigations under this head were either conducted by Professor Lanza, of the Massachusetts Institute of Technology, or else are noted by him in his work, "Applied Mechanics" (ed. 1895, pp. 673-711). They are valuable in that specimens were selected on a commercial rather than on a scientific basis. Professor Lanza claims that such actual pieces are less perfect and show approximately one half the unit strength developed by the more carefully selected smaller specimens. These experiments are characterized as follows:

(4) All other experiments. Many experiments have been made from time to time which, while valuable, are not distinguished by any particular method or principle, such as separate the investigations noted in the preceding articles. Data as to selection of specimens, moisture, and other conditions are either incomplete or else absolutely lacking. Such tests are referred to as are noted in works of Hatfield, Trautwein, Lazlett, Rankine, Thurston, and others. Some of these series are exceedingly valuable comparatively. Professor Rankine and Mr. Lazlett experimented principally upon foreign woods. [p006]

The figures established by the United States Division of Forestry and alluded to in item 1, [page 4], appear, so far as they exist, upon the following pages. Where they do not exist, the leading spaces set apart for them are left vacant for other insertions as preferred. All coefficients are in pounds per square inch. Fractions of pounds in weight and lower figures in coefficients have been omitted as superfluous.

It is not always easy to determine the species of living trees, because forms in the forest differ from those in the open, because bark varies with age, and because fruit and leaves of many trees are lacking in the winter. It is easier to tell genus than species—that a tree is an oak, than whether it is a red or a pin oak. Experience is required in this connection.