THE LEAF. 



223 



pation of the region of the axis close to the apex which 

 it then completely absorbed ; that it is a single leaf, 

 being cylindrical only in its external contour, is shown 

 by the presence within it of a single collateral vascular 

 bundle. The trees of this species at Kew have some- 

 times short-shoots in which the original two needles 

 had reappeared ; one of these was slightly shorter than 

 the other, and both were flattened on the inner side 

 (PL XVI, fig. 3). These two plants seem to afford the 

 only two instances known of normally terminal foliage 

 leaves. 



d. From Dorsiventral to Radial Symmetry. — In the 

 yew (Tax ns baccata) the leaves on the primary or main 

 stem have an even and orderly spiral arrangement ; on 

 the lateral horizontal or ascending branches, however, 

 the leaves, while still preserving the spiral arrange- 

 ment (although this may be slightly less regular), 

 become, owing to the twisting of their leaf-stalks, 

 apparently arranged in two rows along the shoot. 

 The fastigiate variety or Irish yew represents a rever- 

 sion to the primitive type of leaf arrangement. This 

 primitive type is also very frequently induced at the 

 tips of the lateral shoots by the prick of the Cecido- 

 myious insect Friopkyes jpsilajpsis (Phytoptns taxi), 

 forming a characteristic gall. This would seem to be 

 an interesting case of a traumatic reversion. 



2. Terminal. — These are of two kinds, viz. : (1) 

 those formed of a single leaf and (2) those composed 

 of two congenitally-united leaves, but transitions 

 between these occur. 



( 1 ) Single Leaf. — The most remarkable instance of this 

 is that cited by Morren for Gesnera Geroltiana* where 

 an abnormally large foliage-leaf terminated the axis of 

 the flowering-plant ; the structure here is undoubtedly, 

 from its shape, a single leaf, but its unusual size, as 

 well as its position, leads us to suppose that it really 

 represents two leaves. The terminal leaf of Buddleia 

 Hemsleyana represented in PI. XXIII has a basal 



* The genuineness of this case is, however, open to doubt. 



