344 BRYOLOGY OF NEW ZEALAND. 
character, viz., the stems naked in Hypnodendron, tomentose Ny 
Mniodendron, is contradicted by M. Korthalsii Bry. jav., and M 
parvum CM, while the chan niee of the branching (comose in 
rayon frondose in Hypnodendron) is equally unreliable. H. 
ron "for example, having usually quite definitely comose or 
patente. vedi ing. My inclination would be to unite the two, but 
if maintained distinct, I think the naked or tomentose stipes should 
be the crucial character, which would at least be an eee one, 
and would leave Mniodendro on, at any rate, a small and e 
genus, and Hypnodendron no more heterogeneous tes it is a 
present. 
In any ease the New Zealand species of Mniodendron have all 
tomentose stems, while those of Hypnodendron have them naked. 
The New Zealand species of Mniodendron a been much mis- 
understood; principally, I think, owing to Wilson and also Mitten 
having formed a wrong conception wee M. Siebert “(G. M. ); and a great 
number of the specimens in Wilso and Hooker s herbaria are 
eertainly wrongly named. After a rch tudy of the eeeTIpuERe 
and plants I have come to certain fairly definite conclus: 
find no plants with the back of the nerve as 1 as 
is claimed for M. comatum and M. Sieberi. Frequently, or perhaps 
normally, the ine w part of the nerve is toothed or spinose, but 
that is equally true of uM. comosit 
The length i. the setae and a number on a stem is of no 
value whatever, striking as is the difference at times. Thus certain 
plants of M. comatum have the setae always single, and from 4 to 
em. ie while the usual case is to have sore ero a up to 6 
and more together, and only about 2 em. long; but I an find abso- 
lutely no characters sation Sr with these different eordttions. and 
all intermediate stages oceur; the same gathering, for i instance, will 
show a single seta on one stem and 8 or 9 on an adjoining one; while 
with the long form of seta, three stems have a single seta each, while 
the fourth has three. 
Nearly all the plants I have seen fall into two distinct groups; 
M. comosum, a very robust plant with peas straight, more or less 
erect branches, a very stout nerve, and margin of leaf more or less 
bi-stratose ; and comatum, often taller but always less robust, with 
the branches i in a more distinet ecomal tuft, much more slender, often 
nerve much weaker, the margin of the leaf unistratose. 
species the Seiad of Gentenor of the leaf margin and the excurrent 
arista 
es very grea 
The question remains what is M. Sieberi? According to C. 
Mueller this > very close indeed to M. comatum, but differs in the 
n bein subinerassate, ”’ and Brotherus places it, wit 
Sactsiiie. in Pais See n having margin of leaves bi-stratose, and 
nerve at back ataady spinose. I have seen no plant with nerve 
spinose at. of leaf, and the only plant I have seen that departs 
at all widely from the two types described above is a plant in herb. 
Hooker ‘‘ N. Hold., ex herb. Drummond,’’ which has a nerve inter- 
