RESEARCH METHODS IK STUDY OF FOEEST ENVIRONMENT. 99 



The conditions under which the hygroscopic coefficient should be 

 determined, as most recently worked out by Alway, Kline, and 

 McDole (103), are briefly as follows: 



1. The absorption box is of wood, 12 by 9 by 8 inches, the interior 

 surfaces being paraffined to prevent absorption of water and warp- 

 ing. Larger boxes were found to be more difficult to keep saturated. 



2. In the bottom of this box is placed a snug-fitting galvanized-iron 

 tray. 3 to 4 inches high, to hold the water. The walls of the box are 

 then lined with blotting paper, the edges of which project into the 

 ves-el of water. This insures rapid dissemination of vapor through 

 the interior. 



3. A wooden table is held by metal supports, 1 to 2 inches above the 

 surface of the water in the tray. On this table are placed the two 

 trays which hold the soil samples. 



-f. Metal trays are accepted as most satisfactory, because they ab- 

 sorb no moisture and hence do not retard absorption by the soil. 

 These trays are of aluminum or copper, 7 inches long, 5 inches wide, 

 and 0.75 inch deep. 



5. The soil is carefully sifted over the bottom of the tray to a depth 

 of 1 millimeter. This naturally precludes the use of coarser material. 

 It was found that there was little or no change in soils from careful 

 grinding which would barely permit the coarser particles to pass a 

 1 -millimeter sieve. It was also found that oven drying at 105° to 

 Hi) ('. did not appreciably affect the absorbing capacity of any of 

 tin' soils tested. In any case, however, previous drying should be 

 ;i voided when possible. 



6. The exposure to vapor in the boxes is for 24 hours. At the end 

 of tin- period the soil tray is removed from the box as quickly as 

 possible and emptied into a stoppered weighing bottle, since exposure 

 to the air beyond a few seconds would cause appreciable loss of mois- 

 ture. 



7. Undoubtedly the most important consideration in securing re- 

 liable results is a suitable room. This must be, in most cases, a cellar 

 room not subject to daily fluctuations of temperature or heating from 

 one side, or even to Localized heating from bright light. These pre 

 cautions are absolutely vital, if condensation is to be prevented. As 

 ;i matter of theory, it is altogether- probable that the need is to pre- 

 vent even momentary complete saturation of the vapor in proximity 

 to the oil-, since thismighl give rise to t he creation of liquid moisture 

 in them, and cut trely alter' their condition. 



-. A temperature of about CO F. may be considered a standard. 



At a lower temperature there will be fewer water molecules reaching 

 oil, and, necessarily, a dower rate of absorption. 

 Under these circumstances fairly constant results may be expected 

 in hygroscopic coefficient determinations. In the absence of any other 



