558 MR. J. PARKIN ON THE 
by redneing the subtending foliar organs to b ‘acts, the intercalary inflor- 
escence of the usual D. Winteri type would be formed; and, further, by the 
suppression of the terminal vegetative part an inflorescence, falsely terminal 
in origin, like that of D. Howeana and its associates, would arise. 
The axillary inflorescences of Drimys would seem to be of dichasial or 
pleiochasial origin, and have become umbellate in form through internodal 
compression. 
Ruracgzg.—An example of an intercalary inflorescence, after. the style of 
the one found in Drimys Winteri, is afforded by Choisya ternata. It is 
a half-hardy evergreen shrub. 
Boronia, another genus of the same family, also provides examples of this 
kind of inflorescence. 
Myrracea.—This family includes some of the most pronounced instances 
of intercalary inflorescences, viz., those to be found in the Australian genera, 
Callistemon and Melaleuca. The line separating the flower-bearing from 
the foliar part of the main axis is sharply marked. 
Ertcace#.—Several examples of intercalary inflorescences are to be found 
in this family. 
The genus Frica is of especial interest, as the inflorescence is often of the 
interealary type, and in some cases this feature is almost as marked as in 
the Australian Myrtaecous genera just considered. The British species, 
E. cinerea, has a raceme-like clustering of its flowers, but the main axis of 
the inflorescence continues to grow vegetatively beyond the flower-bearing 
part. In the other common British species, ZZ, Tetraliz, the axis does not 
continue so, and the clustering of its flowers resembles an umbel. It would, 
on comparative grounds, appear highly probable that this species has evolved 
a pseudoterminal inflorescence from an intercalary one. 
From a cursory examination of the genus rica, as,a whole, it would 
appear that the interealary inflorescences have arisen, in the first place, by 
the grouping together of short lateral leafy shoots, ending in single flowers 
or in dichasial groups of flowers; and, in the second place, by these shoots 
becoming shorter and losing their foliar character: thus a cluster of flowers 
is formed without any intervening foliage—an inflorescence, in fact, of the 
interealary order. It is a short step from this to the pseudoterminal 
inflorescence. 
From the study of intercalary inflorescences, it would appear hardly 
warranted to consider any apparently racemose inflorescence, occupying a 
torminal position, as truly such, unless by the comparative method it can be 
clearly shown to have evolved from a panicle with a terminal flower, that is 
to say, from a eymose grouping. 
Racemose-like inflorescences should consequently be investigated from 
both points of view. In some cases the comparative method may fail to 
show from which class (interealary or eymose) the inflorescence has arisen, 
