BUBIACE^j. 369 



north of America and Japan, the other series of Rubiacece are plants 

 of warm countries and three-fourths of them are tropical. The 

 Cinchonas, though belonging to the zone in America extending from 

 10° N. to 22° S. rise on the mountains so as to live frequently in a 

 temperate atmosphere. The zone of altitude, formerly fixed at " from 

 5000 to 8000 feet above the sea-level," has, by more recent observa- 

 tions, been extended to from 2600 to 11,000 feet, which affords an 

 explanation of the fact of their having been introduced from the 

 Andine regions to which they were long confined into so many other 

 countries, as Java, India, the Mascarene Isles, &c. Some genera, 

 as Spermacoce, Oldenlandia, are represented by weeds in all warm 

 countries of the globe. Many others are extremely limited in 

 geographical area. Half the Anthospermece are found only at the 

 Cape and in countries bordering upon it. Otiophora, Hymenocnemis, 

 Leiochilus, Nematostylis, Ghapeliera, Camphora, have been seen only 

 in Madagascar. The diandrous types, as Garlemannia and Silvimtthus, 

 are confined to Bengal. There are also Mexican genera that have 

 been seen only in a single very limited spot. On the other hand 

 Morindas have been found in tropical western Africa, Madagascar, 

 tropical Asia and Oceania and South America. Pcederia is now known 

 to extend from tropical Africa to the east of southern America, as 

 also various sections of Genipa, Sabicea, Lasianthus and the great 

 genera Uragoga, Ixora and Ourouparios 



Affinities. — Authors who admit the family Loganiacece think that 

 these difiier essentially from the Rubiacece only in the ovary being superior 

 instead of inferior. We have ranged the group Gcertnerece in the series 

 Uragogece notwithstanding the position of the ovary. The Valerianaceas 

 are very analogous to certain Caprifoliece ; they have the flower except 

 the number of stamens, which is less than that of the divisions of 

 the corolla only in Linncea, and the presence of a tuft at the top of 

 the fruit. Adoxa has been placed, I know not why, among the 

 Araliece. Wherever placed, it is true, it forms a very abnormal type 

 by some of its characters. There is a great resemblance between 

 some Rubiacece and the Compositce, although the inflorescence of the 

 former is never a true capitule, but only a collection of contracted 

 cymes; this arrangement may, indeed, be met with among the 

 Synantherece. The latter have only one fertile carpel ; but such is the 

 case here and there among the Rubiacece, even in a Genipa. The 

 Gornacece appear to us the group which approaches nearest to the 



VOL. VII. B b 



