1901.] NATURAL SCIENCES OF THILADELPHIA. . 359 



familiar archipg was attributed to the weight of fruit. This I have 

 found erroneous. The curving of the large branches does not 

 begin until the tree has reached nearly the height it is (o eventually 

 become. It is not seen on younger trees, no matter how much 

 fruit they may have to bear. This is also true of the Apple. The 

 Cherry does not curve. The Cerasus serotina, Wild Cherry, shows 

 no disposition to curve, but in old specimens of C. Padm the 

 maturer lower branches often curve considerably. In Cratcegus, 

 Amelanchier, and allies no tendency is observable. 



Cornus. 



Old bushes of Cornus Mas have the branches curved considerably 

 with age. Xyssa, the Sour or Black Gum, starts horizontally from 

 the first. 



Diospyros. 



Diospyros Virginiana, the Persimmon, has its branches more or 

 less tortuous at an early stage, but in about ten years the laterals 

 become more horizontal, and the ends of the branches curve up 

 from the fifth year's growth. 



Halesia. 



In H. tetrajitera and H. diptera there is no apparent curving, 

 but H. Meehani, a remarkable seedling sport in my garden from 

 H. tetraptera, has erect branches that recurve remarkably when a 

 few years old, giving the ti-ee a curious wind-swept appearance. 



Catalpa. 



Catalpa hignonioides and C. speeiosa, old trees, give the impression 

 of curvature with age, but I have not been able to decide this to 

 my satisfaction. The tortuosity may be due to early growth 

 alone. 



Laurus. 



The branchlets of Laurus Sassafras start out in early life at an 

 angle of about 45°, but after about five years assume a horizontal 

 direction. The succeeding branch system is tortuous, but seldom 

 becomes decumbent. 



Morus. 



The Mulberry family shows little disposition to depart from the 

 original plan of divergence. In some specimens of Morus alba I 



