Rhodesian species of Fuirena , Hesperantha , and Justicia . 201 
describes so well for the grasses is exactly paralleled in the protective 
tunics of the bulbs, corms, or rhizomes of the representatives of those 
orders in those regions. The ‘ Faser- ’ and ‘ Fasernetz-Tuniken 5 are perhaps 
the most common, but in some genera, viz. Geissorhiza and Hesperantha , 
abnormally thickened c Stroh-Tuniken 5 only are found. In the latter case 
the membranous unthickened parenchyma surrounding the strongly lignified 
internal portion of each superposed tunic must effectively increase the 
sponge-like efficiency of the whole water-holding structure. This would 
be a very important point for a plant like H. matopensis , which has to 
compete for water, at the most inauspicious time of the year, with perennial 
plants, having abnormally well-developed root systems already in possession 
of, and in intimate relation with, the soil particles. 
JUSTICIA ELEGANTULA S. Moore. 
This most interesting little plant is very widely spread in Southern 
Rhodesia and is recorded from Nyassaland as well (Nicholson, Herb. 
Kew). It was first collected by Dr. Rand in 1897 at Bulawayo and at 
Salisbury, and described by Mr. Moore 1 from those specimens, consisting 
of flowering shoots only. In 1902 it was collected again by Mr. Eyles in 
the Matoppo Hills, who sent a complete specimen to the British Museum ; 
it did not, however, arrive until after the present investigation had been 
completed. In the present case 2 it was of very general occurrence in the 
Matoppo Hills, and a large quantity was noted on one of the sidings on the 
way up to the Victoria Falls, a fact which testifies to its wide distribution, 
though herbarium material is at present restricted to the specimens 
enumerated above. 
The author first found it in August, growing in large patches on sandy 
veld, forming colonies suggestive of surface rooting origin. The little 
tufts of flowering shoots, scarcely a decimetre high, were dotted over 
these areas, where they were very conspicuous in the general deadness of 
surface vegetation, owing to their bright rosy pink flowers (Fig. 8). 
Horizontally inclined, dead branches radiated from each tuft of 
flowering shoots (Text-fig. 8, d.b.). The latter, when taken up, were 
found to arise from winter resting buds of fleshy white radical leaves, 
forming sessile rosettes on a thickened rhizome beneath the surface (Text- 
fig. 8, r.l.). These rhizomes were not continuous, being limited to each 
group of shoots, and it was rather difficult to explain their origin. Finally, 
after some hunting, a vegetative shoot was found, with leaves still on it, 
which apparently explained matters, as it bore some roots at one of the 
nodes (Text-fig. 8, n.r). On the flowering shoots the cauline leaves 
succeeding the hypogeal radical ones are at first small and linear, bearing 
1 Moore, Journal of Botany, XXXVIIT, 1900) p. 204* 
P 
2 Gibbs, 1. c. p. 461. 
