NATURAL ORDERS. 135 



genus I have formerly alluded to as consisting of TtuoUia 

 idifpnom and U. balsam ra -^ and a new species of B/n- [va 

 pliai'}^. All these genera exist in India, and none of them 

 have yet been found in America. 



CONVOLVULACE^. The herl)ariuiu of Professor 

 Smith contains twenty-two species of this order, among 

 which, however, there is no plant that presents anything 

 remarkable in its structure ; the far greater part belonging 

 to Ipomoea, the rest to Convolvulus. 



In the herl^arium there is a single species of llydrolea, 

 nearly related to Sagonea palustris of Aublet, which 

 would also be referred to this order by M. de Jussieu. 

 But Hydrolea~ appears to me to constitute, together 

 with Nama, a distinct family {Ilj/drolcce) more nearly 

 approaching to Polemoniaceae than to Convolvulaceae. 



SCROPHULARTN^. The collection contains only 

 ten plants of this family, of which two form new genera, 

 whose characters depend chiefly on the structure of 

 antherae and form of corolla. 



The LxVBIATiE of the herbarium' consist of seven 

 species, three of which belong to Ocymum, a genus 

 common to equinoctial Asia and Africa, but not extend- 

 ing to America ; an equal number to Ib/pfis, which is 

 chiefly American, and has not been observed in India ; the 

 seventh is a species of lloslinulia, a genus hitherto found 

 only on the west coast of Africa, and which, in its inflo- 

 rescence and in the verticillate leaves of one of its species, 

 approaches to the following order. 



VERBENACE/E, together with Labiata? form one natu- 

 ral class,^ for the two orders of which it has already become 

 difficult to find distinguishing characters. 



In the Congo herbarium there are seven Verbenaceac;, 

 consisting of three beautiful species of Clerodendron ; two 



» P/-oc7/-. Flor. Nov. JToll 1, p. 17S. - Vid. op. citat, p. 482. 



^ Flinders' Voj/. 2. p. 505. {A)iie,p. 38) 



