PuaTE 31—CORALLOSPARTIUM CRASSICAULE. 
(THE CORAL BROOM.) 
Famity LEGUMINOS4.. | [Genus CORALLOSPARTIUM, Armstr. 
Corallospartium crassicaule, Armstr. in Trans. N.Z. Inst. xi (1881), 333; Cheesem. Man. 
N.Z. Fl. 108. 
Carmichelia crassicaulis, Hook. /. Handb. N.Z. Fl. 48. 
Corallospartium crassicaule is a most remarkable plant, easily recognized by 
its stout leafless branches, deeply channelled with numerous parallel tomentose 
grooves. It was discovered in the year 1861 by the late Sir Julius von Haast, 
while engaged on a geological exploration of the Mount Torlesse Range; and in 
the following year was also observed by him near Lake Ohau, in South Canterbury. 
Since then it has been collected in many localities between the Waimakariri 
basin and Central Otago, but can hardly be called an abundant plant; in fact, 
in some districts it is decidedly local. It appears to be confined to the eastern side 
of the Southern Alps, and, so far as I am aware, does not occur in any locality subject 
to the heavy westerly rainfall. In short, it is essentially one of those plants most 
at home in those dry and comparatively arid districts which he on the eastern flanks 
of the central chain of the Alps, the high peaks of which either intercept or drain 
the moisture from the prevalent westerly winds. Its altitudinal range can be 
roughly given as from 1,500 ft. to 4,000 ft. elevation. 
Our plant was first described by Sir J. D. Hooker in the ‘* Handbook,” and was 
placed by him in Carmichelia. But it may be doubted whether he would have 
made this disposition of it if he had been acquainted with the fruit, which differs 
from that of Carmichelia in wanting the persistent replum, and which is 1-seeded 
(or rarely 2-seeded) and indehiscent. On these grounds Mr. J. B. Armstrong 
proposed the genus Corallospartium for its reception. 
Mr. Armstrong states that the pod splits into two valves ; and Mr. Kirk, in the 
“ Students’ Flora,” says that the dehiscence is “sutural.” But I believe that both 
these are incorrect, and that the pod is indehiscent, the seed being liberated by the 
slow decay of the valves after the pod has fallen to the ground. Mr. A. W. Roberts, 
of Ranfurly, Central Otago, who has had good opportunities of becoming acquainted 
with the plant, states that this is invariably the case; and other competent 
observers hold the same opinion. 
An adult plant of Corallospartium, with its thick yellow-green branches devoid 
of all leaves, presents anything but a luxuriant appearance, and resembles a bundle 
of dead twigs much more than a living plant. Hence the name of “ Sticks ” which 
in 1897 I found was applied to it by the shepherds of some sheep-stations near the 
Mount Dobson Range, on the flanks of which it is not uncommon. I have often 
sought for seedlings, but have never succeeded in finding one in the wild state. 
They are well known, however, in plants raised from seed in cultivation, and have 
been described and figured in Dr. Cockayne’s memoir on “The Seedling Forms 
of New Zealand Phanerogams ” (Trans. N.Z. Inst. xxxi, 367, t. 31). 
Puate 31. Corallospartium crassicaule, drawn from specimens collected by Mr. A. W. Roberts 
near Ranfurly, Central Otago. Fig. 1, tip of branch from a cultivated plant in the garden of the late 
Mr. H. J. Matthews, Dunedin, showing leaves (x 2). Such leafy branchlets are often produced on 
cultivated specimens, but are rarely seen in wild plants. Fig. 2, flower (x 4); 3, calyx laid open, 
showing the stamens and style (x 4); 4, standard (x4); 5, one of the wing-petals (x 4); 6, keel 
(x 4); 7, ovary and style (x5); 8 and 9, longitudinal section of two ovaries (x 8); 10, ripe fruit 
(enlarged) ; 11, section of same, showing two seeds (the usual number is one) (enlarged) ; 12, embryo, 
showing the double flexure in the radicle (enlarged). 
