,NiSr» JjOt'RNAU W IN^RKSTRY 



ri\e plots were divitled into types n\\i\ the types named on ilu- ba>iis 

 of the species prevaihnjf in total cubic vohtme of the stand in each 

 plot, "rhe chestnut plots were divided into two types, those which were 

 ""M> per cent or more chestnut" in volume and those which were "less 

 than 50 i>er cent chestnut" in voUmu\ The following is a tabulation 



A toiiai v^! V 'r v.v^:-^ v>r «■> ,; per ocni oj all ihc pii>is u-Vi wimimi u\c 



s<w>e site class in cubic volun^e as in heij^ht. In const ructinji Uxird- 



f«>t vt>h\me curws otily the selected \>lots \wre used> as will also be 



the cn^e in workii>§j out the other constituents^ of the yield table, such 



\ r of trees, awragx^ diameter, etc, 



V w^re Wi wnselecttNl plots which did not corresj>ond in cubic 

 vv>U\me and heijjht. and of these "i^ j>er cent fell l>elow in volume 

 rc\5\nv\<nent?i for their site as det«n«med on the basis of height, and 

 only ^1 j>er cent fell in a higher class by volume than by height. This 

 indicates that, in some cases at least, the plot^ were either understtx^ked. 

 or that i>erhaixs in .some cases the deteni^ination of the awrage dom- 

 inant height may haw been t^>o high. In regard to t\i>es, the scarlet 

 ivik atu! white- t>ak t\»t>es had the gTeate.st proi[X>rtion of plots which 



. \ibic \^>l«me set by the generalised 



:e curvxs weix^ drawn so as to include the greatest 

 p ', \>i:s within a gi\ien quality of site as determine*! 



bN , son>e\vhat uneven intervals in x'olume between 



tlu . - > was tv> l>e e>q[>ectetl. esj>eciaUy as ev^n inter\^ls 



are use<cl in the i\ge-heigbt sitf class curves. It w^s found that by draw- 

 ing regttlar inti?rc?ils in \\>l\]ime ov^r 50 per cent of the plots \\^>nld 



