RAMIFICATION OF BUDDLEIA AURICULATA. 203 
true they might be united in pairs ; and so four such bodies might 
be so blended as to form in appearance two only. 
An examination of the mode of development in the young bud, 
however, is sufficient to show that no such blending takes place. 
In the earliest stages there are two leaf-tubercles visible, one on 
each side of the central axial tubercle. As the leaves grow in 
length, their basal portions, instead of becoming detached one 
from the other and from the axis from which they emerge, remain 
in union for a certain time and to a certain extent, thus forming a 
leaf-tube surrounding the axis, homologous with the “ calyx- 
tube ” or * corolla-tube ” of certain flowers. There is thus neither 
* eonnation ” nor “adnation " in the strict sense of these words, 
but rather an arrest of development, in consequence of which the 
separation of parts which should take place does not occur; and 
they remain congenitally united, or, as I prefer to call it, insepa- 
rate. In the first instance, as already said, there are but two leaf- 
tubercles visible ; and in B. globosa no more are produced; but 
in B. auriculata the two lateral leaf-tubercles are very quickly 
followed by two others placed in an antero-posterior direction, 
and emerging from the margin of the leaf-tube between the bases 
of the lateral tubercles. The four tubercles thus form a whorl, 
although they are not simultaneously developed. 
It is clear, then, from this that the so-called auricles in this 
plant represent leaves of a whorl arrested in their development; 
and this view is supported by a comparison with other species of 
Buddleia and with other Loganiacez, wherein every intermediate 
condition between the mere connecting-line—the base whence the 
leaves spring, or, rather, appear to spring—and the verticil may 
beseen*. Itis true that the leaves of this verticil are, so far as I 
have seen, always unequal in size, probably from not being simul- 
taneously developed. The relationship to Rubiaces insisted on 
by many botanists is thus confirmed. 
The purport of the suppression or abortion of a portion of the 
leaves of a whorl may fairly be considered to be to serve as a means 
of obviating the overshadowing of the leaves below by those above. 
It is clear that leavesall arranged in the same plane are more fully 
exposed to the light than those which are placed crosswise in pairs. 
* Among the species which either constantly or occasionally show “ auricles” 
similar to those of B. auriculata are D. brachiata, Cham. et Schlecht., B. brasi- 
liensis, Jacq., B. diffusa, Ruiz & Pav., B. longifolia, B. verticillata, H. B. K., 
B. crispa, Benth., B. salviefolia, D. macrostachya. 
82 
