ON AN EARLY MENTION OF THE DOUBLE WALLFLOWER. 31 
between them except that of colour. But if the reference here is to 
Stocks it may be that the statements to be found in other parts of the 
" Book of Agriculture " in regard to the kheiri were intended to apply 
to the Stock as well as the Wallflower. 
We find a good deal of further information in chapter xxvii., the 
subject of which is "The Culture of Aromatic and Sweet-smelling 
Plants, such as the Kheiri, the Lily, the Water-lily, the Buphthalmum, 
the Narcissus, the Chrysanthemum or Matricaria, the Althaea* the 
Sweet Basil, and other plants of the kind." Article 1 deals with the 
culture of the kheiri (Fr. tr. T. II. p. 256, Sp. tr. P. II. p. 266). 
Clement-Mullet appends a note here to the effect that the Arabic 
name kheiri is synonymous with Cheiranthus in general,! i.e. giroflee 
in all its colours. He then proceeds to ask the question — " Are we 
to take it that giroflees are intended by the Leucoion of Dioscorides 
(lib. 3, cap. 138), the Viola of Pliny (xxi. 14), and the Ion of the 
Geoponics (xi. 22) ? " — and answers it with the remark that M. Fee 
doubts it, and that he himself shares that view. In my earlier account 
I ventured to express the contrary opinion that both the Wallflower 
and the Stock are intended by Dioscorides in his description of 
Leucoion.% This question, however, is not vital to our present purpose. 
The above-mentioned article on the kheiri proceeds to state 
that according to a certain Arab writer, Abou'l Khair, there are 
eight kinds in cultivation, viz. : 
(1) The kheiri of gardens, which is well known, the flower of which 
is purple (Fr. tr. purpurine, Sp. tr. purpureo). 
(2) The kheiri of gardens with flowers white. 
(3) The kheiri with yellow flowers. 
(4) The kheiri with flowers variegated red and white. 
(5) The kheiri with flowers of a violet (? = purple) tint (Fr. tr. 
violacSe, Sp. tr. turquesada). 
(6) The kheiri with flowers of a very deep red. 
(7) The kheiri with flowers dingy (tawny) yellow. 
(8) The kheiri with sky-blue flowers. 
In addition to these iight forms there is, he adds, a wild kheiri 
the flower of which is purplish and small, and also the one known 
more particularly under the name of the water kheiri with purple 
the flowering season, is usually attributed to unfavourable conditions such as 
damp and cold, conditions less prevalent in the Mediterranean region, where 
possibly the greater rarity of the phenomenon may have been the reason 
why efforts were made to induce it artificially. 
* I have rendered the rose de Chine of Clement-Mullet (rosa chinesca of 
Banqueri) simply by the genus. As to the species here intended see discussion 
of the subject in another work by Clement-Mullet, " Iitudes sur les noms Arabes 
de diverses families de vegetaux." — Journal Asiatique, 1870, p. 45. 
| It may be mentioned in passing that Stocks and Wallflowers were in- 
cluded together under the Linnean name Cheiranthus until as late as 181 2, when 
Robert Brown divided the genus and gave the name of Matthiola to the Stock. 
In Linnaeus' Species Plantarum the garden Stock appears as Cheiranthus incanus. 
X The work of Dioscorides here referred to is his Materia Medica, supposed 
to have been written in the first century of the Christian era, a manuscript copy 
of which is preserved at Vienna. A facsimile reproduction of this manuscript, 
which is illustrated by drawings, published in 1906, now renders this work 
more accessible. 
