214 DR. M. T. MASTERS ON THE MORPHOLOGY 
be a new species, R. Harveii, mihi, this creeping rhizome is more 
clearly marked than in most other species. 
Culm.—All writers who have described species of this order 
speak of the branches thrown up from the rootstock as “ culms,” 
and although Bischoff and other terminologists restrict the use 
of the word “culm” to the order Gramines, and apply the term 
“ calamus " to such stem-structures as we have in Restio, I have 
deemed it advisable to retain the word “ culm ” for this family, as 
sanctioned by the invariable practice of Linneus, Thunberg, 
Robert Brown, and I believe of all systematic writers. In the 
face of such overwhelming authority it would be pedantie, to say 
the least, to employ another term, and it would moreover lead to 
confusion. 
The culms then ascend more or less vertically upwards from 
the rhizome,—if of large diameter, then in comparatively small 
numbers ; if small and filiform, then generally they are numerous 
and tufted. In either case they are simple or more or less 
branched, the branches bearing the inflorescence at their ex- 
tremity or being merely sterile. The sterile branches are often 
distinct in habit and appearance, as in function, from the fertile 
ones. In form, the culm and its branches, whether fertile or 
barren, are generally cylindrical, but may be more or less com- 
pressed or angular. Where the branches are numerous and 
crowded, the compressed form is evidently due to that accommo- 
dation process of which the order presents so many marked illus- 
trations. R. tetragonus, Thunb., is the only species, so far as I 
am aware, whose culm is markedly tetragonal; in this case the 
branches are slightly compressed, and their sides fit into the 
grooves of the parent culm. The surface of the culm and its 
subdivisions is either smooth, tuberculated, or studded with mi- 
nute depressions. The colour varies even in the same specimen, 
and greatly depends on the age of the plant; that of the dried 
specimen differs considerably, so far as I have had the opportunity 
of observing, from that of the living plant. 
‘The disposition of the branches affords occasionally a useful 
means of discriminating the species. The culm ordinarily divides 
di- or trichotomously, but occasionally the subdivisions are more 
numerous, and then we have the branches coming off in fascicles 
or tufts, while at other times, as in R. subverticillatus, the branch- 
lets are verticillate. The main culm divides near to its base, or 
towards the centre, or quite at the upper portion, thus giving a 
distinctive aspéct to the plant, although the characters so afforded 
