1918. j The N.Z. Journal of Science and Technology. 303 
treatment, but seasoned timber showed variable results immediately on 
withdrawal from the cylinder. Checking was badly marked in some pieces 
in the pressure steam tests, but not in the low-pressure and creosote trials. 
It might be here noted that in similar trials using zinc chloride, and sodium 
fluoride in aqueous solution, splitting seemed to be affected by the rapidity 
with which the vacuum was drawn. In seasoning and impregnation tests 
with kamahi (Weinmannia racemosa) the most marked change was con¬ 
traction, the timber actually shrinking into smaller bulk. This shrinkage 
varies in degree according to the portion of the tree included in the piece, 
and thus change of shape arises. It would seem that bulk seasoning in 
this way—f.e., by steam treatment before cutting—might minimize this 
well-known fault of kamahi. Again, straight treatment while green, with 
zinc chloride, seems to prevent warping, as well as being germicidal, 
although one cannot be decided on this point. 
Summary .—For application to cylinder treatment of New Zealand 
beech the following general statement may be made, making allowance 
for the larger size of the timber:— 
(1.) Average time to raise interior of sleepers to boiling-point of water, 
three to four hours, using steam at 201b. 
(2.) Green sleepers tend to check badly under high-pressure steam and 
vacuum treatment; seasoned timber -does not. 
(3.) Beech stood pressure treatment with aqueous zinc chloride and 
sodium fluoride with little or no checking when seasoned. 
~ wv a 
(4.) There is nothing to be gained by holding the vacuum for more 
than forty minutes. 
(5.) The rapidity with which the vacuum is drawn together with the 
maximum temperature attained governs the degree of checking. 
High temperature and rapid evacuation favour checking, low 
temperature and slower application of vacuum preventing it. 
(6.) The interior temperature never quite attained that of the medium. 
TABULATED RESULTS OF TEMPERATURE MEASUREMENTS. 
Fig. 5. —Diagram showing changes in interior temperature under steam 
at 20 lb. pressure. Curve A is an average one, while I and II are 
average curves for seasoned pieces. 
