360 PROCEEDINGS OF THE ACADEMY OF [June, 



have seen 'some of the lower branches curved, which must have 

 occurred late in life, but this seems exceptional. 



Flatanus. 



In P. occidentalis I have found no sign of adult curving, but in 

 P. 07'ienkdis, the European species, the laterals change from 

 angular divergence to horizontal positions, and even recurving in 

 many cases. 



Juglans. 



The Walnut family, I believe, retain their early character 

 through life, though the English Walnut has a tortuous appear- 

 ance. 

 Quercus- 



It is in QuevcHs, or the Oak family, Ihat we find the most inter- 

 est, as the numerous species afford material for comparison. The 

 angular divergence and ultimate positions seem so characteristic 

 that we may have a reasonable assurance of the species when a 

 tree is seen from a rapidly passing railroad train. Qiiercus pa/us- 

 iris, the Pin Oak, is a good illustration of what occurs in many 

 species (Plate XVII). 



The upper branches grow at an acute angle, but after about 

 five years proceed to decline, the point of departure being at the 

 base, near the main trunk. By fifteen years they have l^ecome 

 horizontal, and continue declining until the divergence again 

 reaches an acute angle. Much the same procedure marks Quercus 

 Phellos, the Willow Oak. Quercus tindoria, the Black Oak, has the 

 lateral brauches horizontal at an early age, but I have never found 

 any go lower. This is true of the Quercus coccinea, the Scarlet 

 Oak, though to a less degree (Plate XVII), Quercus bicolor, the 

 Swamp White Oak, departs from its angular position toward the 

 horizontal when about ten years old, but never gets below the hori- 

 zontal line. Quercus macrocarpa (the Burr Oak), Q. hjrata (the 

 Over-cup), Q. Prinus (the Kock Chestnut), Q. rubra (the Red), I 

 have not found to vary from the original framework. This also 

 seems true of the European Quercus Robur, Q. Cerris (the Turkey 

 Oak), and of the Japan Q. dentata. Q. alba, with iis frame of 

 branchlets almost as huge as its trank, has been a puzzle to me. I 

 am inclined to believe that the angular divergence of the lateral 

 branches has not changed from the original plan. It imiy be noted 



