Species of Xeronema — Moore 
357 
moorei they are all similar (Fig. 4A, D). Dif- 
ferences in capsule shape were emphasized by 
Oliver and are shown in Figure 4B, E; New 
Zealand capsules are more strongly stipitate 
than those seen from New Caledonia. A 
striking difference not recorded by Oliver, 
since it was not suggested by Brongniart and 
Gris, either in description or figure, is seen 
in the old capsules. In X. moorei the style 
remains slender and becomes twisted like the 
long staminal filaments which, together with 
the shrivelled persistent tepals, form a thready 
tangle about the opened capsule. 2 In X. 
callistemon the style thickens and dries firm 
and straight and stiffly erect so that the 
raceme permanently retains its brush-like ap- 
pearance (Fig. 4B, E, Fig. 5). In both species 
the spreading of the capsule valves at de- 
hiscence tends to split the style base into its 
three constituent parts, with frequent break- 
age in X. moorei. 
Brongniart and Gris (1868: 3) describe fer- 
tile seeds of X. moorei as 1.5 mm. long, "uno 
latere convexa, aculeis brevibus apice inflatis 
truncatisque exasperata, altero nudo lateraliter 
raphe carinato.” This description and their 
accompanying figures agree rather well with 
seeds of X. callistemon. Seeds from the cap- 
sules recently received from New Caledonia 
(Fig. 4F) are regularly longer than those of 
X. callistemon and almost oblong in outline; 
processes cover the surface except on rather 
vaguely defined furrows at the sides of the 
prominent keel which is itself echinate along 
its ridge. In X. callistemon (Fig. 4C) the seeds 
taper somewhat towards one end, the keel is 
smooth and the inner faces are plane and 
sharply distinct from the outer curved face 
which alone is echinate with semitransparent 
blunt cylindrical processes; processes in X. 
moorei are equally blunt but darker, more 
opaque and more inclined to be bent. 
2 "Les filets staminaux longs, secs et persistants du 
Xeronema , rneme a l’epoque de la maturite du fruit, lui 
donnent un aspect tout particular d’ou nous avons 
tire le nom generique” (Brongniart and Gris, 1868: 5). 
Compare the erroneous explanation in Cranwell and 
Moore, 1938: 25 and Moore, 1953: 26. 
Fig. 2. X. moorei. Montagne des Sources, New 
Caledonia; on top of rocky ridge. 1955. Photograph 
by L. J. Dumbleton. 
The pollen of X. callistemon has been de- 
scribed and figured (Cranwell, 1942: 293; 
1953: 49, pi. 5, text fig. 38). That of X. 
moorei is slightly smaller but with similar ex- 
ceedingly characteristic reticulation of the 
exine (Cranwell, 1953, from herbarium speci- 
men, confirmed by N. T. Moar (personal 
communication) from fresh pollen). 3 
ECOLOGY AND DISTRIBUTION 
The habitats of the two species are probably 
as nearly alike as their different latitudes can 
provide. X. moorei in New Caledonia (Lat. 
22° S) "seems to do best on rather rocky nar- 
row ridges more or less in the cloud belt" 
(Dumbleton, in litt.). Selling (in Guillaumin, 
3 Chromosome numbers of Xeronema have been counted by 
Dr. J. A. Rattenbury of Auckland University College. X. 
moorei from Montagne des Sources has a somatic number of 
2n = 72 (unpublished) and X. callistemon from Hen Island 
a somatic number of 2n = 36 (Roy. Soc. New Zeal. Trans. 
(1957) 84 (4)). Specimens from which root tips prepa- 
rations were made are growing at Auckland University College. 
