VOYAGE FROM INDIA TO SIAM AND MALLACCA. 79 



the base when the flower opens, then the two parts are patent 

 ovate, with marked veins, somewhat fleshy, concave, one inch 

 long and persistent Hg-ht green. The real flower consists of 

 four oval, concave white petals red at the edges, they are a 

 little longer than the calyx. 



The stamens are very numerous, they are joined together 

 at the base, a fleshy skin surrounding them, as m the Eugenias, 

 they are like thin threads, smooth, shiny, milky white as far as 

 the middle and towards the end beautifully pink, they are twice 

 as long as the flower and fall off. 



The anther oblong four plicate yellow. The ovary is 

 situated underneath the flower and renders the stalk impercep- 

 tibly thicker and square. The style is threadlike, thicker than 

 the stamen, smooth, shiny rose coloured, persistent, and three 

 times longer than the stamen. The stigma is flat, perforated 

 and white. The receptacle is umbilical, and rough. 



The fruit is a square pyramid, a little longer than it is 

 broad, at its end stands the divided calyx straight up; they are 

 rounded, the surface is smooth and shiny ; greenish white in the 

 beginning, changing when ripe into a brown leather colour; in- 

 side it is filled with stiff, flat viscous fibres, surrounding an oval 

 seed bigg'er than a pigeon's egg, which has a hard, thin shell, 

 the taste whereof is very bitter. 



In all probability this tree is the Mammea asiatica more 

 especially that which Mr. Osbeck has seen near Java, only it has 

 escaped his notice that the blossoms are borne on the ovary 

 that the stamens are all grown together at the base and drop off 

 independently of the flower, the stigma is flat and perforate, the 

 fruit is described as a pear, and as being divided in to four 

 parts. This partition is caused by the circumstance that they 

 are crossed by the in the inferior ovary time of bloom and short- 

 ly afterwards, and so it presents itself in the shape of four rose 

 coloured partitions, oneof themgenerally containing a white round 

 seed as big as a mustard seed. The three remaining parti- 

 tions in all the specimens I have seen, were either quite empty 

 or contained only rudiments of seed. It is very rare that 

 more than one partition contains any seed, and if so, these dis- 

 appear soon, and only one of them generally ripens and fills the 

 whole fruit, I have cut many of these fruits and always found 

 them as I described them. The fruit is a real Drupe. The sea 



