1917] 



Bruce: A New Dendrometer 



59 



in part a matter of adjustment. Two opposed adjusting screws must 

 therefore control the rotation of one of the two mirrors. Adjustment 

 is simple. Some target of known diameter or breadth (a sheet of 

 paper against a dark background will serve) is observed with the 

 instrument set at the corresponding diameter. The mirror is then 

 rotated by its adjustment screws until the two edges appear in line. 

 This process is delicate, but neither complicated nor difficult. 



|llil| l i n |lill [ ili|li n i | li|i|n n| li n| ii n| i;il| n il| ni i| Mn[ ii n | M ll | l n l |nM| |l|i| i1 if| MI I||lll |M ll|i M l | i M i|ii n| |iil || ill|llll] 



A - Assembled Dzndrometer. Elevation 



%~; 



4 



| c e | d 



D- S/'de frcm £>e/orv 



3- P/an 



C- Cross Section 

 of Arm 



3 



LuW 1 I I I I 



Scale for A.B an4C 



f,! I I ll I I I f l ] 



Scale ford.E. andF 



_£"- Slide, tie. vat i en 

 a. — Arm. 



b. — Stationary mirror platform. 

 c. — Sliding mirror platform. 

 d.— Slide. 



F- Slide, from end 



e. — Fixed mirror and support. 

 /. — Adjustable mirror and support. 

 gg. — Adjustment screws. 

 h, i. — Springs. 

 Fig. 2 



The one error which dominates all others is that due to a failure 

 of the arm to be absolutely straight. This is unfortunately a matter 

 of instrumental construction and not of adjustment, and the difficulty 

 of making this arm straight is surprisingly great. It is obvious that 

 almost imperceptible deviations will result in slightly diverging or 

 converging sight lines and in increasingly serious errors in the 

 diameter readings, as the distance at which the measurement is taken 

 is lengthened. In the instrument described the maximum error from 

 this source is .6 inch when used at fifty feet ; it is doubtful if materially 

 better results are obtainable. This is not excessive. Even with a transit 

 read to the nearest minute, the diameters fifty feet away can be read 



