RESULTS OF A STUDY OF DOUGLAS-FIR SEFD 995 



classes is likewise shown by the weight per 200 seed, which was 2.875 

 grams, 2.130 grams, and 1.225 grams respectively. In each case 200 

 seed were sown in germination tests, with the following results : Large 

 seed, 29 per cent good; medium seed, 15 per cent good; small seed, 

 6 per cent good. It is clear, therefore, that the large seed of a tree are 

 of much better quality than the small ones. 



The practical application of these results depends on the interesting 

 fact brought out in the main Douglas-fir seed study, that large cones 

 produce large seed and small cones small seed. Since small seed are 

 inferior, it is well to collect only the larger of the cones on a tree. 



(4) What temperature is best for cone drying in a kiln? 



For this study a small kiln was built, in which the temperature could 

 be well controlled and the humidity and the air circulation regulated to 

 a considerable degree. The heating was done with a kerosene lamp 

 and the temperature recorded usually with a thermograph. The humid- 

 ity was increased by means of a steam jet from a kettle — or sometimes 

 by boiling water inside the kiln — and decreased by bettering air circula- 

 tion, through raising end curtains on the kiln. The humidity was 

 sometimes stated in general terms, as low, high, medium, and in other 

 cases determined by psychrometer readings and given in per cent of 

 relative humidity. The air circulation was classed as slight, good, etc.. 

 according as the kiln was tightly closed or otherwise. 



Altogether 23 tests of drying were made, each test comprising 10 or 

 15 cones. Three of the tests were of air drying and the remainder of 

 kiln treatment. The details of the experiments are contained in the 

 original notes, which are on file in the Forest Service office at Portland, 

 Oregon. In the following discussion it will be noted that the subject 

 broadened considerably when the tests started. It is possible to sum- 

 marize here only the main results, which for the sake of simplicity are 

 discussed under seven subheadings. 



(a) Amount of moisture which must he lost before a cone will 

 open. — Twenty-two tests, dealing with green cones, have recorded fig- 

 ures showing the loss of moisture during various periods of drying. 

 From these it is possible to compute the loss there must be before a 

 cone opens, though the conditions were too variable to permit other 

 than general statements in this regard. It was found that when less 

 than 16 per cent of the green weight was lost the scales did not even 

 start to spread ; when 19 per cent to 34 per cent was lost, the scales 

 spread more or less widely without really opening the cone ; and when 

 over 35 per cent was lost at least some of the cones opened well. 



