OX THi: CONSTRUCTION OF HoofB) CANN«-N. 4<J 



of about 1000° is required to produce this annealing < fiV t. For iron, it slimil.l 

 be carried to a full rod Ik it, whatever temperatnie that may be. 



Now, for the purpose of ascertaining, with some d< ree of pr< ision, <Jjo differ- 

 ence in hardness, tenacity, and elasticity between a piece of iron subjected to 

 various degrees of heat from 400° up to that which produo< a thoroughly 

 annealed stale, that is, a full red heat, and the same iron after hiving been 



subjected to some one of the hardening process*'* before mentioned, I have 



made a great many experiments upon iron wire of various iiea> and in various 

 states as produced by previous working and heating. I wiU now relate the 

 mode of making a few of tin e experiments and the results obtained from them, 

 which results were in accordance with numerous others obtained by the 

 method of operating. 



-ame 



These experiments were made upon pieces of iron wire about fifteen 1 et 

 long and of different sizes. The instrument lor performing the experimenta con- 

 sisted of a long horizontal frame, to one end of which was affixed a strong 

 steelyard, which was bent into the form of a bell-crank; and the shorter arm of 

 which was vertical, while the longer arm, upon which hung the poise, was hor- 

 izontal. One end of the wire to be experimented upon was connected with 

 the shorter arm by being turned a few times about a ring which was con- 

 nected with the arm by a free joint. The other end of the wire was fastened, 

 by similar means, to a strong bolt fixed to the frame at a distance of lift* n feet 

 from the steelyard. Connected by cramping it with the wire, near the end last 

 described, was a stiff wooden rod, which lay upon the frame, and pa d, by the 

 side of the wire and parallel with it, to near that end which was connected with 

 the steelyard. To the neighbouring end of the wooden rod was fixed a smooth 



o o 



tm plate, about one foot long and five inches wide. Tin plate lay Opon the frame, 

 immediately under the wire and nearly in contact with it; and upon the sur- 

 face was registered, by a fine needle-point, the changes in the length of the wire 

 under different tensions. To do this, a short straight-edge, or ruler, was firmly 

 cramped to the wire, directly over the register-] late, so that, when a line w;i< 



drawn upon the register-plate by the 



was laterally pressed 



the edge of the ruler, its direction would be across the register-plate, or at a right 

 angle with the axi- of the wire. The distance from the point upon the wire 

 where the rod connected with tin register-plate was cramped to it, to the point 

 where the straight-edge or ruler was cramped, was exactly 1 10 inch ^; and it will 

 be seen that, by the arrangement here do 5 ri bed, whatever yielding or springin 



Mo. Bot. Garden* 



1897. 



