70 
Gardner — Conirihidions to Flora of W.A, 
The new sjun-ies ])el(m^’s to the section Uaconosm.: of lientlianij 
■witli nfliiiity to G. floribuudinn, (t. microairpum and Ct. oxylohioides. 
It is closest to G. jiorihundiim^ from which it differs in the mucli 
smaller acute thick leaves, much shorter })etioles, somewhat longer 
(coni])aratively much longer) stipuh's. which are perhaps also 
Iwoa.der, and e(jual and larger calyx-lobes, and in the racemes. The 
style is also ha.iry in the lower half, hrom G. microcarpum and 
G. oxylohioides it differs in habit, mnch thicker, smaller and more 
crowded leaves, wliich aia* also not glaucous, denser and shorter 
racemes, and other })articulars. 
The delineation of the s])ecies of (kist rolohium is so artificial, 
being based in some s]>ecies on measurements and indiinientum, that 
it Ijecomes a difficult matter to define exactly the limits of these 
spe(‘ies. Such diff('rences, for example, between G. microcarpum 
and G. oxyl(d)ioidcs^ are very much less than those wliich separate 
the varieties of G. spinosum. The ])roposed new species is very 
different in habit from any of the Gastrolobiunis known to me, and 
the rigid crowded leaves are suggestive more of a Vidtenaca than a 
(rasiroJobium. 
Gastrolobium dci)sif<di inn is a ])oison plant. An investigation 
into its toxicity was undertaken by 11. W. Bennetts, Veterinary 
Pathologist of the l)e])artnient of Agriculture, Perth. The material 
used was in a dried condition, and was. fed to guinea-pigs. The 
(‘xperiment on the first animal was discontinued on the fourth day. 
The second guimm-pig, which at(‘ 7 grams of leaves in three days, 
was dead on the fourth day. The symptoms shown by the first 
animal were, rough coat, ina])])etence. shivering and erratic inove- 
numts. The animal coni]>letely recovered on the sixth day. On 
^lost moitem examination the second animal showed congestion of 
the organs, notably liver, kidneys and lungs. 
Explanation of Plate VII, 
Fncohjpius DicJsii Gardner — The type. 
A. — i natural size. 
B, (\ D. and E.— 1 i times natural size. 
P.-~ Pront and linck view of anther (enlarged). 
Printed for the Royal Society of Western Australia by R. S. 
Sampson, 971-973 Hay Street, Perth. 
