82 CIRCULAR 3 3 3, U.S. DEPARTMENT OF AGRICULTURE 



(c) Where two or more tables have been independently prepared from the 

 same data they should be checked against each other, as, for example, by 

 computing the board foot-cubic foot ratio or the percentage difference between 

 the volumes by the Scribner and International rules. 



38. The following information should be published with every volume table : 



(a) Species (common and scientific names). 



(b) Region or locality. 



(c) Author and date. 



(d) Unit of measure (log rule). 



(e) Portion of tree measured. 



(f) Basic data: Number in each d.b.h. class, number in each height class, 

 location of data by heavy lines. 



(g) Method of computation. 



(h) Method of measurement (instrument and length of sections). 



(i) Average deviation. 



(fc) Aggregate difference. 



(Z) Age. 



(m) Site, if yield tables exist; if not, physiographic location of material. 



39. It is premature to standardize methods of preparing taper tables. The 

 Jonson and Behre formulas are the most promising, but demand a further 

 study of the factors of bark thickness and root swelling and of the method of 

 field determination of form classes. The conventional methods are of doubtful 

 accuracy. 



40. The number of trees required for a satisfactory table depends on two 

 factors : 



(a) The average deviation. 



(&) The number of entries in the final table. The number of trees measured 

 should be directly proportional with each of these factors, but it is probably 

 premature to set fixed limits. Perhaps 300 or 400 could tentatively be accepted 

 as satisfactory where neither of these factors is abnormally high. 



41. In checking the applicability of volume tables a relatively small number 

 of trees may be used. If their aggregate difference and average deviation from 

 the table are calculated, the test for satisfactory accuracy is an average devia- 

 tion of the same order of magnitude as that for the basic data on which the 

 table was prepared and an aggregate difference which does not exceed two 

 times this average deviation divided by the square root of the number of trees 

 used in the test. In cases of exceptional importance a check of this sort should 

 be made separately on the smaller and the larger sizes. 



NORMAL YIELD TABLES FOR PURE EVEN-AGED STANDS 

 FIELD DATA 



42. The basic data for the preparation of an adequate yield table should 

 consist of from 100 to 300 plots for each major type involved, well distributed 

 throughout the range of site classes and ages involved. 



43. The plots should be well distributed over the region to be studied so as to 

 include all characteristics of the type's range. 



44. All plots should be essentially even-aged, the definition of " even-aged " 

 depending on the reproduction period of the species studied. By " reproduction 

 period " is meant the average time required to obtain sufficient young growth 

 on the ground to insure a fully stocked stand at maturity. 



45. The size of the plots should be such that they will include from 100 to 

 300 trees each. 



46. The shape of the plots need not be rectangular, but there should be at 

 least four sides, and all angles should be greater than 60°. 



47. Each plot should be surrounded by an isolation strip of similar timber 

 wherever possible, and where not, as large a proportion as possible of the 

 perimeter of the plot should be thus protected and the boundary on exposed 

 sides should be laid out with exceptional care to include the whole area occupied 

 by the border trees. 



48. The survey should be made by staff compass and tape or other means of 

 equal accuracy. 



49. Plot areas should be computed on the basis of horizontal distances. 



50. Wherever practicable one semipermanent corner on each plot should be 

 set and tied to some easily identified landmark. A sketch map of the region 

 showing location of plots should also be prepared, 



