730 



(1) That iu order to establish reliable data as to mecLanical properties of our tim- 

 bers, it is necessary to make a verj- large mimlter of tests, l)y which the range as 

 well as average capabilities of the sjtecies is deteruiiued. 



(2) That in order to enable ns to make the most etiicieut jiractical ai»idi<;ition of 

 the data thus obtained, it is necessary to know the physical and structural condi- 

 tions of the test material and bring these into relation with the best results. 



(3) That in order further to deduce laws of relation between mechanical jtroper- 

 ties ami the ])liysical antl structural conditions, as well as the conditions under whith 

 the material was jiroduced, it is necessary to work on material tlie history of which 

 is thoroughly known. 



Brielly, then, to solve the i)roblems before us it is necessary to make our test on a 

 large number of specimens of known origin and known physical condition. While 

 the tests in themselves ajijical at once and tirst to tin- engineer, inasnnich as by their 

 great number they w ill furnish more reliable data regarding the i apabilities of the 

 variotis timbers, the chief value and most imiiortant feature of the work lie in tl>e 

 attempt to relate the mechanical i)roperties to the structure of the material and to 

 the conditions under which it was protluced. 



Tlie organization and niotliods of tlu' invt'sligation liavc been biit'Hy 

 described iu Circular No. 7 of the Division, as follows: 



There are four departments necessary to carry on the work as at present organized, 

 one for each of the following lines of work: (1) (.'ollectiug, (2) mechanical tests. (!t) 

 physical and microscopic examinati<»n of the test material, and (4) compilati*)U and 

 final discussion of results. 



The collection of the test material is done by experts (Dr. Charles Mohr of Mobile, 

 Alabama, for Southern timbers). The trees of each species are taken from a number of 

 localities of dirtennt soils and climatic conditions. From each siteliv*' trees of ea<li 

 species are cut up into logs and disks, each piece being carefully marked, so a.«* to 

 indicate exactly its position in the tree; four trees are cho.sen as representiitive of 

 the average growth, the fifth or " check tree " lining the best-develtiped specimen of 

 the site. 



Disks of a few young trees, as well as of limb wood, are also collected for biological 

 study. The disk ])ieces are 8 inches iu height and contain the heart and sapwooil of 

 the tree from the north to the south side of the iieriidieiy. From fifty to seventy 

 disk i)ieies and from ten to fifteen logs are thus collected for each species and site. 



A full account (dtlie louditiiuis of soil, climate, asjtecf nu-asuremenfs, and deter- 

 minable history of tree and forest growth iu general accompanies the collection from 

 each site. 



The disks are sent, wra]>ped iu heavy paper, to the botanical laboratory of the 

 University of Michigan at Ann Arbor (Mr. F. Koth, iu charge), to be studied as to 

 their jihysical jiropertiea, macroscojtic and microscojtic structure, rate of growth, 

 etc. Here are determined (1) the specific weight by a hygrometric method; ('J) the 

 amount of water and the rate of its loss by drying in relation to shrinkage; (3) the 

 structural diflerences of the diflerent pieces, especially as to the distribution of spring 

 and summer wood, strong aiul weak cells, open vessels, nu'dullary rays, etc.; (4) the 

 rate of growth and other bicdogical facts which may lead to the tiuding of relation 

 between physical appearance, condition of growth, and mechanical properties. 



Tlu' material thus studied is jireserved for further examinations and tests as may 

 api>ear desirable, the history of each piece being fully known and recorded. 



The logs are shipped to the St. Lonis Test Laboratory, in charge of I'rof. .1. H. 

 Johnson. They are stenciled off for sawing and each stick markeil with dies, cor- 

 responding to sketch in the record, so as to be perfc<tly identified as to number of 

 tre»', and thereby its origin, and as to ])ositiou iu tree. After sawing to sizi', the 

 test i)ieces are stacked to await the testing. One half of every log will be tested 

 green, the other half after thorough seasoning. At the time of testing a deteruiiuatiou 



