•wo NATURAL HISTORY OF PLANTS. 



terjjs, Rijssopterijs, and Hiptage, all belong to tropical Asia and Oce- 

 ania. There is not one species of the Maljnghieco and Gaudichaudiece 

 series that is not American. Nine genera of Banisteriœ and seven of 

 Hireeœ are so likewise. "Brazil," says A. de Jussieu, " seems the 

 trne country of the Muljnghiaceœ, so noticeable are they by their 

 number and variety in this part of the earth more than in any other." 

 There are, in fact, twenty-five of the American genera represented 

 in that country, and by a considerable number of species (nearly two 

 hundred and ninety). Some American genera, monotype or with 

 very few species, belong only to Guiana and Columbia, as Coleo- 

 stachys, Dlacidia, Loj^hopterys, Biplopterjjs^ and Jubclina. Mexico, 

 where the species of Malpighiaceœ (many of which remain to be 

 described) are not rare, owns the monotype genera, Lasiocarpiis and 

 Echimpterijs. To Chili or the neighbouring parts of Peru belong 

 the two exceptional (or perhaps congeneric) types Dinemandra and 

 Dincmagonum. Tricomaria is still more southern and also more 

 abnormal as to aspect and foliage, analogous to those of certain 

 Rhamnaceœ of arid countries. No Malphigiaccœ of North America 

 have been observed above 40° north latitude; there are only three 

 or four at the Cape, a couple in Australia, and at the most half-a- 

 dozen in the rest of Oceania. They are then, especially, plants of 

 tropical regions, and the majority American. 



Their aflBnities with the Erythroxijleœ and Nitraricœ have been 

 recognised by all Avriters. They rarely have the alternate leaves of 

 Erythroxylon, but in that case, they do not possess their so charac- 

 teristic intra-axillary stipules. Ancidoplms, on the contrary, has 

 petiolate leaves and intra-petiolate stipules ; but is distinguished by 

 • the geminate ovules in each cell, with the micropylc tm-ned dh-ectly 

 upwards and outwards. The Erythroxylece have an abundant fleshy 

 albumen in the seed. In Nitraria the leaves are alternate, the 

 petals valvate-induplicate and not twisted, and the ovule, suspended 

 next its fuuicle, turns its microijyle directly upwards and inwards. 

 But the Sapindaceœ especially are closely allied to the 3Ia/jiigh/accœ, 

 having nearly all their important characters particularly in the regular 

 types. The Ma^nghiaceœ, however, have no appendages to the 



