30 



tubular and concentric. In both cases the new leaves are formed 

 at the center of this false-stem, completely hidden from view, 

 and only push out from the top when they are practically ma- 

 ture. In respect to the position of their inflorescences, the two 

 types of plants differ fundamentally. The single inflorescence 

 which culminates the growth of each banana "tree" is formed 

 in the center of the false-stem and, after performing a long up- 

 ward journey, emerges at the top, while in Arch onto phoenix, 

 once the tree has reached reproductive maturity, an inflores- 

 cence is generally formed in the axil of each leaf, and they are 

 exposed successively as the leaf-sheaths split open and fall 

 away. 



While the youngest leaf stands almost erect, the older ones 

 gradually droop, and finally the oldest, hanging vertically 

 downward, begins to die and turn brown. The increasing pres- 

 sure of the young leaves and inflorescences, continually being 

 formed within, finally causes the thick, leathery sheath, over 4 

 feet long, to split from top to bottom down the side opposite to 

 the attachment of the lamina. Meanwhile the base of the sheath 

 is becoming detached from the stem so as to leave a circular 

 scar. As the dying leaf falls away the elongated spathes which 

 enclose the inflorescence are revealed standing upright in the 

 axil of the sheath. The first spathe may split down either the 

 inner or outer side, and falls away, according to my observa- 

 tions in Panama, from 8 to 16 days after the frond has dropped. 

 The spadix is still enclosed in a second, similar spathe, which 

 generally bursts and falls a day or two after the first. These 

 spathes are green, thick and leathery, and completely closed, 

 and the inflorescence is so crowded within that there is not the 

 slightest waste space. The flower buds are sharply angular as a 

 result of the pressure to which they have been subjected during 

 their development (Fig. 1). The spathes are burst simply by 

 the swelling of the spadix within them, and there is no pre- 

 formed line of dehiscence. 



Only one of the five fruiting palms was low enough for me 

 to reach the inflorescences when standing on top of an eighteen 

 foot stepladder, and this tree was kept under observation from 

 January untiFJune, 1929. Once they had escaped from their 

 spathes, the numerous, slender, whip-like branches of the in- 

 florescence hung beneath the short, thick, horizontal main axis. 



