were between more distinctly marked forms of structure, 

 connecting- them with each other, but not very obviously 

 corresponding with any. 



It was originally supposed to be a Rutaceous plant, and 

 the twisted style, the definite numbers among the several 

 floral envelopes, and the few seeded ovary favoured this sup- 

 position ; but when it was more exactly examined, the inde- 

 hiscent fruit, unsymmetrical flowers, curved embryo, great 

 hypogynous disk at the back of the stamens, and undotted 

 leaves, were opposed to the opinion of its forming part of 

 that order. 



Upon the publication of his Enumeration of Baron Hugel's 

 plants, Dr. Endlicher stationed the genus among Sapindacese, 

 led to that conclusion it is to be presumed, by its unsym- 

 metrical polygamous flowers, large disk, frequently 3-celled 

 ovary, as also by the structure of the embryo, and the trace 

 of an arillus found upon the seed. Its habits, however, 

 correspond so ill with Sapindaceae, that nothing but a very 

 exact examination of characters would have led to its being 

 stationed in that order. 



In the place above quoted, I suggested that it would 

 prove to be an anomalous form of Capparidacese ; my reasons 

 for which were, that it has altogether the habit of a Cleome, 

 that its stipitate ovary, glandular hairs, declinate stamens, 

 and especially its large dimidiate disk, are all in accordance 

 with that order, with which the seeds do not materially 

 disagree in structure. The objections however, to a reference 

 of Diplopeltis to Capparidaceae, namely, the discrepancy 

 between the number of the sepals and petals, and the pluri- 

 locular ovary are great ; the latter in particular, in our 

 present ignorance of the true value of characters, forms an 

 obstacle to which the mere habit of a plant and circumstances 

 of external structure are not equivalent. 



Upon the whole then I would agree to referring Diplo- 

 peltis to Sapindacese, because it is more like that order than 

 any thing else, but being stationed there it must be regarded 

 as an outlying genus, tending to connect the order very 

 closely with Capparidacea^. 



Fig. 1 is a male flower, deprived of its petals ; 2 is a 

 female in the same state : 3. is a vertical section of the ovary. 



