PLATE 9.—NASTURTIUM LATESILIQUA. 
Famity CRUCIFER %.] [Genus NASTURTIUM, R. Br. 
Nasturtium latesiliqua, Cheesem. in Trans. N.Z. Inst. xl (1911), 179. 
Cardamine latesiliqua, Cheesem. in Trans. N.Z. Inst. xv (1885), 298; Man. N.Z. Fl. 35. 
The first specimens of this fine plant were gathered by myself in the summer 
of 1881, during a botanical expedition to the Mount Arthur Plateau, Nelson. It 
was then abundant on limestone rocks on the northern side of Mount Arthur, 
ranging from a height of 3,800 ft. to 5,500 ft., or within a short distance of the 
summit of the mountain. Two years later I found it equally plentiful on the 
northern slopes of Mount Owen, in the Upper Buller Valley. Since then it has 
been gathered by Mr. F. G. Gibbs and others on several of the high peaks in north- 
west Nelson, almost always on limestone rocks. 
When publishing the species in 1883 I referred it to Cardamine, being mainly 
influenced by its evident relationship to Hooker’s C. jastigiata, from which it differs 
in the villous leaves, larger flowers, and much broader more turgid pods. But, as 
pointed out in the Manual, both C. fastigiata and C. latesiliqua, together with the 
allied C. Enysiz, differ from the typical species of Cardamine in the 2-seriate seeds. 
Consequently Dr. Schulz, in his elaborate monograph of the genus published in 
Engler’s ‘‘ Botanische Jahrbuchern ” for 1903 (vol. xxxii), excludes all three, but 
makes no other disposition of them. But although there is no difficulty in justi- 
fying the exclusion of the species from Cardamine, it is far from easy to decide 
in what other genus they should be placed. Personally, I look upon it as quite 
possible that the three species in question, together with the Australian C. radicata 
Hook. f., will ultimately form a separate genus. But before taking a step of that 
kind a careful comparison should be made with the whole of the genera constituting 
the subfamily Arabide, a work which can hardly be properly carried out without 
access to the large public herbaria and libraries of Europe. Partly from that reason, 
and partly because the characters of Nasturtiwm, although not exactly conformable, 
still approach very closely to those of the four species mentioned above, I have 
decided to transfer them to that genus for the present. (See some remarks in a 
paper of mine published in the ‘ Transactions of the New Zealand Institute,” 
vol. xlint (1911), 178.) 
Prats 9. Nasturtium latesiliqua, drawn from specimens collected by Mr. F. G. Gibbs and myself 
on Mount Arthur, Nelson, at an altitude of 4,000 ft. Figs. 1 and 2, flowers (x 3); 3, petal (enlarged) ; 
4, stamens and pistil (x5); 5, pistil (x8); 6, section of portion of pod, showing 2-seriate seeds 
(enlarged) ; 7, section of seed, showing position of embryo (enlarged) ; 8, 9, 10, 11, 12, various embryos, 
variously magnified. 
