safford: a new genus of annonaceae 17 



nized by Baillon, who placed them in the genus Xylopia, in a 

 section by themselves, which he called Pseudanona (see Adan- 

 sonia 4: 142. 1864); but in studying the genus Xylopia the 

 present writer has become convinced that these two Mauritian 

 plants cannot possibly be regarded as congeneric with the West 

 Indian Xylopia muricata, the type of the genus as established 

 by Linnaeus,^ and it is therefore proposed to segregate them by 

 raising Baillon's section to generic rank under the name of Pseu- 

 dannona. The genus may be described as follows. 



Pseudannona (Baillon) gen. nov. 



Calyx gamosepalous, subtriangular or 3-lobed. Corolla com- 

 posed of 6 petals in 2 series, all of which are valvate, the outer 

 3 longer and broader than the inner and subspatulate in form, 

 the inner more or less triquetrous especially near the apex. Recep- 

 tacle (torus) domeshaped, with a crater-like depression at the 

 center from which issue the styles of the sunken ovaries. Ovaries 

 distinct, each terminating in a slender style which is recurved near 

 the extremity where it is provided with numerous stigmatic 

 papillae. Ovules several, biseriate, not superimposed as in the 

 genus Xylopia. Stamens numerous borne on the outer slope of 

 the torus and forming a crowded mass surrounding the central 

 cluster of carpels; filaments stout, bearing on their back a pair 

 of linear parallel pollen sacs opening extrorsely by a longitudinal 

 slit and surmounted by the broadened connective. Fruit a clus- 

 ter of distinct several-seeded carpels borne on the hardened recep- 

 tacle, or torus, not a syncarpium composed of many 1-seeded 

 carpels fused together as in the genus Annona. Climbing or 

 erect shrubs with solitary 1-fiowered peduncles indigenous to the 

 islands of Mauritius and Madagascar. 



Thus far but two species can with certainty be referred to this 

 genus, a third species, from the forests of Mauritius is like P. 

 grandi flora, a climbing shrub and has been described ^s Anona 

 pyriformis Bojer. Tho the latter cannot possibly belong to the 

 genus Annona it is doubtful whether it should be assigned to 

 the present genus, owing to its broad inner petals which resemble 



2 Linnaeus. Syst. Nat. 2: 1250, 10 ed. 1759. 



