3 1 6 Arber . — The Leaf Structure of the Iridaceae , , 
in M. polystachya , opposite the larger bundles there are smaller bundles 
inversely orientated. I am unable to confirm this description ; I find the 
bundles to be all normally orientated and arranged in a single series. • 
The most striking ; peculiarity of the dorsiventral Moraeas is the 
unusual character of the midrib region. This region, in the species which 
I have examined, is thinner than the remainder of the leaf, and its bundles 
are smaller than the rest and somewhat irregularly placed. This feature is 
not confined to the dorsiventral species. Figs. 12 A and B, p. 304, show 
the absence of the median bundle in the ensiform M. Macleai ; the shape 
of the leaf of this species makes the lack of midrib even more conspicuous 
than in the bifacial forms. 
The genus Moraea differs from Iris chiefly in the absence of a perianth- 
tube and the fact that the stamens generally cohere. It is commonly said 
that the Moraeas are found to the south and the Irises to the north of the 
Equator. But this can only be maintained if the Barbary Nut, Iris 
Sisyrinchmm , L., be ranked as a true Iris. This plant, which ranges from 
Portugal and Morocco to North-west India, resembles the Irises in 
possessing a perianth-tube, but in many other points recalls the Moraeas. 
Though Sir Michael Foster 1 placed it unhesitatingly in the genus Iris, 
and both Pax 2 in Engler’s ‘ Pflanzenfamilien ’ and Baker 3 in the ‘ Handbook 
of the Irideae ’ take the same line, other authorities have long held that it 
ought to be regarded as a Moraea ; it was figured as ‘ Moraea Sisyrinchmm^ 
the European Moraea ’, in Curtis’s ‘ Botanical Magazine ’in 181 1 (vol. xxxiii, 
No. 1407). Mr. W. R. Dykes kindly tells me that his long study of the 
genus Iris has convinced him that this plant should be excluded from the 
genus and transferred to Moraea , which it resembles in its corm and in 
the sheathing leaves of the inflorescence axis, while Ross 4 and Chodat and 
Balicka-Iwanowska 5 have reached the same conclusion from the leaf 
anatomy, which is the point that specially concerns us here. The leaf of 
the Barbary Nut (Fig. 38 a) recalls some of the Moraeas in consisting 
of a flat bifacial region (leaf-base) terminating in a cylindrical apex (petiole). 
The apical appearance of the leaf exactly corresponds, for instance, to that 
of Moraea edulis , Ker-Gawl., as figured in Curtis’s ‘Botanical Magazine’, 
vol. xvii, No. 613. The internal structure is shown in Figs. 38 B-D, and 
it will easily be seen that in the anatomy — especially in the peculiarities 
of the midrib region — there is an exact reproduction of the features of such 
a species as Moraea polystachya (Figs. 36 A and B). ■ The leaf structure 
thus tends to confirm the attribution to Moraea . 
1 Foster, M. (1892). 2 Pax, F. (1888). 
3 Baker, J. G. (1892). 4 Ross, H. (1892-3). 
5 Chodat, R., and Balicka-Iwanowska, G. (1892). 
