14 



(33-) 7'Opilote (Buzzard). Coarse, long-grained wood, but of 

 compact layers when viewed in cross-section. It has a greenish 

 color. Turns and planes fairly well. The wood is used only in 

 neighborhoods where more useful woods are scarce. When unex- 

 posed or when well seasoned and protected with paint, it would 

 probably be a valuable wood, as it ran above the average in torsion, 

 was among the highest in compression, and stood well under 

 transverse stress. It attains a height of about 40 feet and I foot 

 in diameter. 



The Transverse Stress tests were made in the Transverse test- 

 ing machine of the Messrs. Fairbanks, designed for Prof. Thurston. 

 It consists of a Fairbanks' scale combination with a pointer and 

 beam at the end to secure perfect balance. The whole machine 

 rests upon a wooden frame ; upon the platform rests a cast iron 

 cross beam, and upon this slide the supports, which are set 

 and secured by bolts at the required distance apart. The 

 test piece is placed upon the mandrels, which rest on the supports 

 set at the required distance. The loads are put in the scale, and 

 equilibrium is established by the elastic resistance of the piece 

 offered against the pressure transmitted through the screw and 

 pressure block by means of the lever. The screw passes through 

 a nut, and terminates in a sliding piece. The cast iron columns 

 serve as guides to the pressure block. The whole is made stable 

 by wrought iron braces bolted to the wooden frame. The deflec- 

 tions of the piece are measured by the linear advance of the screw, 

 by means of a graduated wheel. The pitch of the screw was 

 found to be equal to 0-33297 inches, and the wheel was graduated 

 into 333 equal divisions, thus reading to j^-^ of an inch, with an 

 error of ooooi of an inch. The pointer clamped 1o the screw is 

 placed over the starting division in each case, and after the wheel 

 has been turned and equilibrium established, the reading is taken. 

 This is a very convenient way of reading the deflections, giving the 

 difference directly. The test pieces were planed by hand by an 

 expert, Mr. Kerr, and afterwards measured by means of a 

 micrometer screw reading to y^^ of an inch. 



With all the data required, we obtained the results in the same 

 way as in compression tests, by plotting the curve to each test, 

 with loads as ordinates and the deflections as abscissas. This 

 gives the elastic limit more exactly than in any other way. It also 



