184 BOTANICAL NOTES. 
species, though T. venosa approaches that condition. The first 
three can at best be considered names of convenience to group 
the very numerous forms of one variable species. TJ. carnea 
is a well marked species in the type, but is intimately con- 
nected to T. ixioides by intervening links, and T. cyanea forms 
an intermediate species between TY. ixioides and T. venosa. 
The flowering period of the first five species is the same, 
namely, October and November, and they grow usually 
together, except that T. cyanea shows a partiality for damp 
situations. The flowering period of T. venosa is December 
and January. In most descriptions the habit in reference to 
degree of robustness is relied on as an aid to identification, 
but an intimate acquaintance has completely destroyed that 
assistance. In all, except the typical form of 7. carnea, 
specimens may be found from a few inches high to nearly 
two feet, and from an excessively slender state, with an 
almost filiform leaf grooved on the upper surface to a thick 
robust condition, with a broad, thick, fleshy leaf, concave 
above, and from bearing one flower to a raceme of 8 to 12. 
T. carnea has a marked habit, distinguishing it from the 
other memters of this group, namely, that the stem is bent 
at each bract, and in a totally unreliable degree. TT. venosa 
is usually a more robust species than T. cyanea and T, 
aristata, with large forms of J. izioides, are more robust 
than the common forms of TY. longifolia. In size and 
colour of perianth there is little really reliable. T. ixioides 
may be blue, pink, or rarely white, it is often spotted 
with dark blue, or dark pink, but not always so. ZT. 
aristata and longifolia has a pale blue to white perianth with- 
out spots or veins. T. carnea has a fleshy-pink perianth, but 
in varieties approaching J. izioides, it is spotted with dark 
pink or blue. T. cyanea and YL. venosa has a blue, pink, or 
white perianth, veined with darker colour; the column also 
is striped with dark coloured bands. In all the perianth 
varies from 4 to 10 lines long, the greatest variation being 
found in T. izioides and T'. aristata. 
The only other means of distinction at our disposal, namely 
the shape of the column, yields but little evidence to favour 
the right to consider all these forms as reliable species. 
The column in this group is short. The stigma is peltate 
and near the base, and the anther varies inversely as the 
development of the column-wing from small, and placed elose 
above the stigma at the back of the rostellum to large and 
protruding and freely removed above the stigma. The 
column is surrouuded at the sides and base by a com 
paratively broad wing that is continued at the back, and 
more or less above the anther into variously formed lobes or 
processes. In 7’. venosa the wing is not continued behind the 
anther, but ends in two narrow lobes bent forwards and 
