246 MR. G. BENTHAM ON EUPHORBIACE.B. 



IV. ORIGIN AND GEOGRAPHICAL DISTRIBUTION. 



Among Dicotyledons, Euphorbiaceae stand fourth in point of 

 number : — Composite being first, with (in round numbers) 10,000 

 species in 800 genera ; next, Leguminosae, under 7000 species in 

 400 genera ; thirdly, Kubiaceae, above 4000 species and rather 

 under 350 genera ; Euphorbiaceae having above 3000 species in 200 

 genera. Further discoveries are likely to raise the numbers of 

 the last two of these Orders more in proportion than those of the 

 first two, but yet not sufficiently to alter their relative position. 

 Labiatae, which follows as the fifth, have under 3000 species in 

 140 genera ; and their geographical relations are not such as to 

 promise so many additions as in the case of Rubiaeeae and Euphor- 

 biaceae. The relative positions of the five orders may therefore 

 -be regarded as definitively fixed. In definitiveness of circum- 

 scription, Euphorbiaceae rank with Compositae and Leguminosae. 

 Each of these three Orders has become perfectly isolated, without 

 any intermediate forms remaining to bridge over the interval which 

 separates them from the surrounding ones — having no trace of 

 those gradual modifications which so closely connect Rubiaeeae on 

 the one hand with Caprifoliaceae, and through them with Corna- 

 cese, and on the other, with Loganiaceae, and through them with 

 the great mass of dicarpellary gamopetalous orders, or of those 

 which allow of no very definite line being drawn between Labiatae 

 and Verbenaceae. EuphorbiaceaB, therefore, like each of the two 

 above-named primary Orders, can be treated as a definite whole ; 

 and it is thus that the study of their origin and geographical dis- 

 tribution acquires a peculiar interest. 



In investigating the origin of Euphorbiaceae, the geological 

 record is unfortunately of no assistance. The generally herbaceous 

 genera of the northern temperate regions are not such as to leave 

 any permanent traces of their existence ; and the leaves of arbo- 

 rescent Euphorbiaceae have, in general, no peculiarity by which 

 they can be distinguished from those of other Orders ; besides 

 that, the palaeontology of those tropical regions where the majority 

 of them are to be found is but little known, and I can find no 

 authentic record of a single fossil Euphorbiacea. Ettingshausen 

 has indeed referred about a dozen impressions of leaves to various 

 genera of that Order ; but Schimper (Paleontologie V6getale, iii. 

 290) doubts the identification of most of them ; and not one is 

 positive enough to found any such conclusion as that there was 



