536 THE AMERICAN NATURALIST. [Vol. XXXVIII. 



trie with the growth rings and lie either in the spring or summer 

 wood, or in both. This is to be interpreted as a definite tend- 

 enc)' to aggregation which is nevertheless not fully expressed, 

 since in each case there are numbers of cells which are not 

 zonal in their distribution, but which conform to the law appli- 

 cable to Thujopsis and Podocarpus. In Juniperus the cells are 

 typically zonate, being also scattering in only one species. In 

 Abies only 63.6 of the species bear resin cells. These are 



neither scattering nor zonate in the sense of the previous types, 

 but It IS to be observed that in 50 f of such cases, or in 36.3 f 

 "t all species, they are aggregated in groups as a preliminary 

 -stci) t<) the formation of resin passages. On the other hand, 

 36.3 '/ of all species show the resin cells to be few. inconspicu- 

 ous, non-resinous and scattered along the outer face of the sum- 

 mer wood. This, for reasons which will appear more fully later, 

 IS to be regarded as a phase in distribution leading to the final 



