48 
unique in the order, and it seems to have outweighed all the 
oan enemy which must have tended towards the separation of 
G 
the gen The authors of the Genera Plantarum may have 
hesitated t to separate the African ‘“ aie ” from the Malayan 
for bslenap of sufficient material ; at a e, their diagnosis of 
] 
any r 
Kickxvia agrees very well with Blume’s ae of his genus, 
but ee fits the set pey plant referred to it. The latter, 
indeed, appears rather as a kind of appendix to the former, no 
better place having been ipl for it for the time 
~ development of a flying apparatus in shape of a tuft of 
hairs or lumose awn attached to the seeds is a universal 
aintelyanes in Hchitidee. The tufts eit either from the 
chalazal end of the seed, and then they are termed basal, or from 
the micropylar end, when they are styled apical, or they caatiiate 
from both ends. Sometimes they are transformed into plumose 
awns by the lengthening of the axis of the tuft, and often also by 
the intercalation of a naked stalk between the seed proper and the 
e commonest form is an apical tuft. Basal tufts 
without aig ones are characteristic of Wrightia; basal and apical 
tufts occur together in lsonema, Adenium, and Haplophyton ; 
_ apical awns and basal tufts together are found in all the numerous 
species a Strophanthus ; basal wns aaa in Kickxia and 
Funtum Where two tufts or a tnt and a plumose awn occur 
iivaltandotely. the Teed! tuft is often early deciduous, and does 
not leave the follicle with the seed ; nevertheless its occurrence 
eee that there isa more general disposition towards dev veloping 
e flying contrivance from the chalazal end of the seed than is 
pecans assumed. At the same time we see that the presence of 
this peculiar disseminative organ is not confined to genera which 
are admittedly close allies, as a glance at the different attempts 
to group the genera of create will show. To summarise 
briefly, the basal awn of the seeds of Kickria and Mad daaits 
is unique in the order in so te as in no other case known, n, the 
peculiar modification, but it has its homologue in several not 
closely allied genera, and therefore cannot be considered as a 
character in itself indicative of close relationship 
What is true of the basal awn may be said of the apical plumose 
awn of Strophanthus, Laubertia, Stipecoma, Urechites, ete. Tt 
is the homologue of the usual apical tuft of the majority of 
nai, and occurs also in genera sehowwiee not closely linked 
oge 
It is quite ai that the basal awns in Kickria and 
Funtumia have evolved from the basal tufts of two types 
which had little ee’ in common than those characters which bind 
foliaceous or planoconvex cotyledons being the rule. So far as 
I know, the genera Wrightia and Holarrhena are the only ones 
