578 JOURNAL OF FORESTRY 



individual and local requirements must be unintelligently sacrificed 

 for the sake of system, but I do believe that the Branch of Research 

 can take even the existing data and find out the principles of tree form 

 of the American species, make better tables than we now have, and 

 eliminate further duplication. This must be done naturally in collabora- 

 tion with the Administrative Branch of the Service, which is to use the 

 tables. It is for the latter Branch to define what it wants, and then 

 for Research to go ahead and make it. 



There are a number of questions as to the system to be adopted that 

 will arise at the very start, and I should like to outline some of them. 

 Those which I will speak of in turn are the following : 



1. Local vs. general volume tables. 



2. Normal or forest-run trees as the basis. 



3. Allowance for defect. 



4. The diameter basis. 



5. Total vs. merchantable height. 



6. A fixed or a graduated top diameter. 



7. The length of the sections. 



8. The assumed stump height. 



Local vs. General Volume Tables 



I spoke a minute ago of the practice of some of the cruisers in the 

 Forest Service of making local volume tables for each separate projecL 

 This practice is now less common than a few years ago. 



Such tables were based on a few trees, and in them was inculcated 

 the cruiser's individual method of cruising — the intensity of utilization 

 that he assumed for that job, the local defectiveness that was culled 

 etc. These local tables were undoubtedly excellent for the maker's use 

 in the one locality, but should not have been used by anyone else or in 

 any other locality. With the large amount of cruising that the Forest 

 Service is doing, it strikes me as a great extravagance to make local 

 volume tables, though something can be said for their superiority for 

 local use and their fitness for a particular job. The Service is now 

 pretty well committed to standard volume tables, namely, volume tables 

 based on a large number of trees from various localities. I believe that 

 this is good policy, and that if the tables follow a carefully worked out 

 plan, as discussed below, standard tables for broad regions will work 

 very well. 



There is undoubtedly a limit to the applicability of a standard table, 



