PREFACE. xi 
and cultivated as many as he could, and also sent a large living 
collection to Kew. Lady Barkly and Miss E. B. Barkly made 
water-colour drawings of them as they flowered. Of these 
copies were sent to Kew, together with specimens in alcohol, 
accompanied by copious descriptive notes. An account of 
this material was published in Hooker’s Jcones Plantarum, 
tt. 1901-25; and coloured figures of four of the species that 
flowered at Kew appeared in the Botanical Magazine. The 
whole of Sir Henry Barkly’s material was a contribution to 
the study and elaboration of the group only second perhaps in 
value to that of Mr. Pillans. 
I may be permitted more personally to express my indebted- 
ness to Lieut.-Colonel Pratn for kind and unfailing assistance 
in many ways, without which the task of editing a work of 
this kind at a distance from the resources of Kew would be 
one of peculiar difficulty. 
The critics of niceties of nomenclature, which often seem to 
obscure the interest of larger problems, will probably notice 
that the Kew tradition has been adhered to, and the supposed 
right of priority of the original specific epithet has been 
disregarded where an existing name is available which has 
correctly placed a species in the genus to which its affinity is 
most obvious. The principle was laid down by Sir Joseph 
Hooker in 1872 in Zhe Flora of British India. Its 
justification is based on technical grounds equally with those 
of common sense. It may be convenient to briefly state 
them :— - 
i. The so-called binominal nomenclature which we employ 
was devised by Linneus, and, as with everything he did, on 
a logical and definite basis. Nothing but confusion can arise 
by departure from this. To the specific epithet, apart from its 
proper function, Linnzus attached no importance at all. He 
saw that the scientific problem was to get the species into its 
right genus. ‘Nomen specificum sine generico est quasi 
pistillum sine campana.” The specific name taken alone is 
the clapper without the bell. A Linnean name, then, though 
it consists of two parts, must be treated asa whole. “ Nomen 
omne plantarum constabit nomine generico et specifico.” And — “ 2 
