CLASSIFICATION, 5 
$2. Sketch of the attempts to classify the species of Dalbergia. 
The year 1834* is marked by the first real step in advance as regards the internal 
classification of the species of Dalbergia. То have thrown out of Dalbergia all 
foreign elements save the Brachyptera was to have done much; and if it may be 
said that we owe this to Graham's assortment of the Wallichian Dalbergias in 1839, 
we have to recognise that Wight and Arnott went still further when, while retaining 
Derris scandens in Dalbergia, on account of its pod, they treated it, because of its 
opposite leaflets and its versatile anthers, as the type of a distinct subgenus. More 
important, however, was their proposal to subdivide the true Dalbergias into two distinct 
groups according to the monadelphous or isodiadelphous arrangement of the stamens, 
because this proposal became the basis of the first atterapt made by Bentham іп 
18517 to divide the genus into natural sections. Bentham at the same time 
reduced to Dalbergia the Malayan Lndespermum proposed by Blume, the African 
Podiopetalum proposed by Hochstetter, and the American Triptolemea proposed by Martius, 
subdividing the whole into three sections:—(1) Sissoa, with tke stawens monadelphous 
and the pod long and straight; (2) Selenolobium, with the stamens of Sissea, but the 
pod short and lunate; and (3) Dallergaıia, with the pod of Sissoa, but the stamens 
isodiadelphous. Тһе subdivision was convenient, and at the time it was proposed 
appeared to be natural, There were, however, only 29 species then known, and of 
these only two—JD. spinosa and D, torta—were known to have the pods of Seleno- 
lobium, though, as a matter of fact, two other species, whose fruits were then unknown 
to Bentham —D. parviflora and D. reniformis —have Selenolobioil pods. Miquel, when in 1855 
he described the Malayan species of Dalbergia, adopted Bentham’s three sections and 
observed, what Bentham had overlooked, that D. parvifera must be a Selenolobivm, He 
did not, however, suggest any alteration in the principle of classification. In 1860 
Bentham returned to this subject in the course of his classical essay on the 
Dalbergieae,§ a model of lucidity and method. The pod, as he there explains, which 
had formerly been chiefly relied on in separating Dalbergia from other Leguminosae, must 
now be abandoned as a primary guide, А limitation based entirely on the рой 
involves, as we have seen iu the historical review of the genus prior to 1834, the 
inclusion of many species with different foliage and inflorescence and, above all, 
with very different stamens— species that are much better referred to Derris and to 
Lonchocarpus. If insisted upon, it would logically invoive the suppression of various 
distinct and apparently quite natural genera like Platymiscium and Leptolobium. Оп 
the other hand, if rigidly applied, the character afforded by the pod would involve 
the exclusion from Dalbergia of the species for whcse accommodation Bentham Las 
proposed the section Selenolobium, and would further emphasise the exclusion of the | 
quite artificial genus Ecastap/yllum. Restricting in this fashion the generic importance 
attached to modifications of the pod, Mr. Bentham explains that, had he for the first 
time been grouping the species of this genus, he would have divided it into three 
artificial secticns—(a) Dalbergia, with a long straight pod; (b) Selenolobium, with the pod 
thicker and lunate or reniform; and (е) Zeastophyllum, with the pod orbicular or 
nearly so. The actual delimitation adopted was a modification of this design, 
t Flor. Ind. But. 1. 1, 126. 
i : dr. Penins. Ind. i. 264. 
* Wight & Walker-Arnott, Prodr. Penins I exe Lie Sok ir. Ran 
+ Plante Junghuhniane i. 264. 
