TERN STBCEM1 ACE 'M. 253 



upon this question, united the two groups into one and the same 

 family of Ternsircemiacece, the history and the general organiza- 

 tion of which he made known. The genera before established 

 which lie admitted 1 were thirteen in number : — Coehlospermum, Tern- 

 strcemia, Eroteum {Freziera, Cleyera), Eurya, Sauravja, Stewartia, 

 Gordonia, Ar chit a a, Mahurea, Marila, Kielmeyera, Caraipa, Thea ; he 

 added Laplacea and Bomietia, but he left there wrongly one of the 

 Pixacece — Ventenatia? In 1855, Choisy, continuing the study of this 

 family in a detailed monograph, 3 found there as newly-established 

 genera, 4 Adinandra, Jack ; b Anneslea, Wallich ; 6 Pyrenaria, Blume, 7 

 and Schima, Reinwardt. 8 He admitted besides the genus Penta- 

 phylax, Champion, which has been finally rejected. Bentham and 

 J. Hooker, in their Genera, 9 comprise in this family Rhizobolece, that 

 is to say, the genera Caryocar and Anthodiscus ; 10 Marcgramce, with 

 the three old genera, Marcgravia, Norantea, and Ruyschia ; Actinidia 

 and StacJiyurus, genera belonging to other groups, 11 and which they 

 united in the same tribe with Saurauja ; the Omp/ialocarpmi of 

 Palisot de Beauvois, which is one of the African Sapotacecs ; the 

 Microsemma of Labillardiere, whose place among the Ternstrce- 

 miacece has also been contested ; 12 also Pelliceria, whose existence in 

 Columbia 13 Triana had just indicated, and Haploclathra, detached 

 by Bentham 14 from the ancient genus Caraipa. In 1865, Beddome 15 

 added Poeciloneuron to the preceding genera ; which, besides the 

 doubtful types, 16 makes the number of genera which we can preserve 

 as autonomous twenty-eight. 



1 Only counting those which we have pre- 12 Tr. & Pl., in Ann. Sc. Nat., ser. 4, xvii. 

 served as distinct. 359. 



2 He did not include Visnea (Mocanera) 13 Ex B. H., Gen., 186 (1862) ; in Ami. Sc. 

 which JussiEtr (Gen., 318) had placed, we know Nat., loc. cit., 3S0 (1863). 



not why, among the Onagracece. H In Joum. Linn. Soc., v. (1861). 



a Mem. sur les Fam. des Ternstreemiacees et lo In Joum. Linn. Soc., viii. (1864). 



Camelliacees (in Mem. Soc. Bkys. et Hist. 16 These are: 1. Michoxia Velloz. (Fl. 



Nat. de Gen., xiv. 94). Flum.,v.t. 103), ascribed doubtfully by Be> t tham 



4 Without citing those which had not been & Hookek (Gen. 438) to the genus Ternstroe- 

 kept as autonomous. mia, but finally rejected from this genus by the 



5 In Comp. to Bot. Mag., i. (1835). same authors on account of its 2-fid calyx and its 



6 PI. As. Ear., i. (1830). cucullate nectary. 2. Hexadica Loue. (Fl. 



7 Bijdr., 1119 (1826). Cock., 562), which Mueller d'Argovie ascribes 



8 In Bl. Bijdr., 129 (1825). (in DC, Prodr., xv. p. ii. 1259) to Ternstree- 



9 I. (1862-1867), 177, 981, Ord. 28. miacea, or to Clusiacem; an opinion which 



10 G. F. Mey., Prim. Fl. Fssequeb. (1818). Bentham & Hooker find inadmissible. 3. Cato. 



11 The former has been studied among the stemma Benth. (in Hook. Lond. Joum., ii. 365 ; 

 Dilleniacem (vol. i. 114, 134) ; the latter was ii. 365). In this tree of British Guiana, 

 formally placed among the Piltosporece (Endl., which was first ascrihed to the abnormal Tern- 

 Gen., n. 5699), and appeared nearly allied to slroemiacea>, then to equally abnormal genera 

 certain Ericaceae. of Myrtacece, the flowers are hermaphrodite and 



