GROWTH IN TREES. 9 



This ingenious method would enable the observer to detect changes 

 of 0.00005 inch in circumference, but as no recording devices are 

 provided the constant attendance of the observer was necessary. 

 Meanwhile my own results show that an actual amplification of 8 to 

 15 times the changes in diameter of the trunk is adequate for direct 

 observation and analysis. 



Friedrich had previously used a device for recording increments to 

 the thickness of trees as early as 1905. l The tree was encircled 

 with a steel band on roller bearings, and the ends were attached to 

 both registering and recording arms. No cognizance was taken of 

 errors due to changes in temperature, and no results obtained with 

 this apparatus have been seen. 



Some success having been attained in the measurement of the varia- 

 tions in stems such as those of the sunflower (Helianthus) and in the 

 variations in volume of succulent stems such as those of Opuntia, 

 by the use of mechanical levers, it was found desirable to extend the 

 observations to trees. It was thought that accurate measurements of 

 the changes in volume of these massive structures might not only 

 give information of value on the subject of hydration and growth, but 

 that some light might be thrown on the problem of ascent of sap. 



The attempt to measure the variations in large trunks, begun in 

 1918, raised many difficult mechanical problems. First it was obvious 

 that it would be practically impossible to hold the stem of any but 

 small trees rigidly in any position. This made it necessary to devise 

 means whereby the tree would support the instruments and carry 

 them with it as it swayed and leaned. It soon became evident that 

 such an appliance should consist of two parts; a base which would 

 remain firmly in whatever position near the horizontal it might be 

 placed about the trunk, and which would carry the more delicately 

 arranged members of the second part which would register variations in 

 diameter at some little distance from this base. 



Two forms of base or support were devised and tested. One form 

 consisted of a rectangular frame built up of bands of wrought-iron 

 2 by y inches and of oak wood 2 by 2>^ inches, which could be 

 clamped about the tree tightly by means of an iron rod, including a 

 heavy turnbuckle. A heavy block of seasoned oiled wood at one end 

 formed a base for the levers and recorders and a wooden screw 30 cm. 

 long, working in a smaller threaded block hinged to the outer end of 

 the larger base block, extended to a point on the trunk several centi- 

 meters below and formed a brace which kept the block rigid in its posi- 

 tion with respect to the tree. Flexible supports carried a floating 

 frame which served to register variations in thickness. This form of 

 base was heavy, unwieldy, and by reason of the metal was subject to 

 temperature variations which might influence the record. 



1 Friedrich, J. Zuwaehaautograph. Centralb. fur das gesammte Forstwesen. 31, pp. 45&- 

 461. Nov. 1905. 



