l..<; 



NATURAL HISTORY OF PLANTS. 



The plants of this family have long been distinguished as 

 constituting a natural group either on account of their aspect or their 

 properties, or because of some prominent character, as the form of the 

 malvaceom corolla, or the organization of the cohimniferom fruit. 

 From Zaluzian 1 until Linnaeus, 2 authors have made particular 

 mention of this group. But it is necessary to refer to the Genera 

 of A. L. de Jussieu, 3 in order to see united in one and the same 

 order all the representatives then known of the different series we 

 have enumerated. These are thirty-two in the work we have 

 just cited. But the successors of A. L. de Jussieu soon divided his 

 order Malvacea into several secondary families. Ventenat 4 separated 

 from it Sterculiaca, and R. Brown 5 from Buettneriacece. In 1824 

 De Candolle, 6 in making these two into one, admitted Bombacea? as 

 a separate family. The multiplication of these groups is carried 

 as far as possible in the works of Endltcher, 8 and above all of 

 Lindley. 9 But as the characters by which the three principal 

 types of Malvacece, Sterculiacece, Buettneriacece, are distinguished 

 from each other, are far from being constant and absolute,' we see 



1 Meth. Herb. (1592), cl. 16. The Mallows. 

 This class is distinguished, according to him, 

 among others by J. Bauhin in 1650, by 

 Johnston (in 1661), by Magnol, Morison, 

 &c. 



2 Fragm. Meth. Hat., in Cl. Plant. (1738), 

 Ord. XXXIV. (Columniferi). 



3 271, Ord. xiv. (1789). 



4 Malmais., ii. (1790), 91. 



5 In Flind. Yoy. (1814), ii. 540; Misc. 

 Works (ed. Benn.), i. 11. 



6 Prodr., i. 429, 475, 481. 



" K., Piss. Malvac. (1822), 5. Kunth in- 

 cludes in one and the same general group, Mal- 

 vaceae, Sterculiacea, and Tiliaca. After which 

 he secondarily divides Sterculiaceai into series 

 corresponding to most of those which we have 

 enumerated. 



8 Gen. Plant., 978-1012. The author divides 

 his class L, that of Colwnmiferce into four orders, V 

 Malvaceae (209), Sterculiaceai (210), Buett- 

 neriacece (211), and Jiliaceae (212). Ster- 

 ctdiacete comprise according to him Bombaceae 

 and llelictcrece ; and he adds to Buettneriacea', 

 Lasiopelaleae, Dombeyeae, Hermannieae, Prio- 

 tamea', and Philippodendreae. 



9 Veg. Kingd., 359. The author admits also 

 Sterculiacea>, Buellneriacere, Malvaceae, as dis- 

 tinct families, defining them as Endlicher has 

 done. 



10 To abridge the examples, we see that Lindley 

 characterizes Sterculiaceae as " Malval Pxogtns, 

 with columnar stamens all perfect, and 2-celled 

 anthers turned outwards ;" and that this family, 

 moreover, includes Matisia and Quararibea with 

 one-celled anthers properly placed in this group, 

 because they are inseparable from the Myrodias 

 with two-celled anthers ; Helicteris, of which 

 the anthers are sometimes those of Myrodia, and 

 sometimes those of Matisia; Plagianlhus and 

 Hoheria, which have anthers really 1-celled; also 

 allthe Bo»;Z((7cfrt > ,arealso constituted the same. The 

 Buettneriacece are defined as : " Malval Fxogens, 

 with 1-adelphous stamens, in most 'cases partly 

 sterile, and 2-celled anthers turned inwards." This 

 group, moreover, includes several Basiopetaleae, 

 with extrorse anthers, almost all the Pombeyeae 

 having anthers also extrorse,as have mostPerman- 

 nieee, and Buettnerieae, and Philippodendron, which 

 is a Plagianthus. It may even be said, that the 

 extrorse anthers constitute the exception in this 

 family, such as Lindley defines it. I do not 

 speak of the numerous plants destitute of sta- 

 minodes which are necessarily comprised here. 

 Bentham and Hooker have, without doubt, 

 recognised the insufficiency or the inexactitude 

 of these various characters, for they have pre- 

 served {Gen., 195, 214) but two orders, those of 

 Malvaceae and Stereuliaceae, according as the 

 anthers have one or two cells. But if such a 



