109 



fruit dehiscent by three sutures. The nut having been taken 

 out appeared to be perfectly normal, only flattened in the 

 places of meeting and covered there with an especially thick 

 layer of mace. The diverging valves are double and deeply 

 grooved at their backs and tending to split up their tops. In this 

 way there are created six valves, the cohering parts belonging 

 to different carpels. We have consequently before us a three- 

 carpelled fruit in its very perfect development. 



For completeness' sake we may add that the length of the 

 fruit amounts to 68 mm. and the circumference, apart from 

 the splits, measures 244 mm. The perusal of Prof. Janse's 

 paper, cited above, has wholly confirmed my opinion about 

 the plurality of carpels in the said cases, thus rejecting the 

 accidental development of ovules dying in normal circumstan- 

 ces as Mr. Warbiu-g seeks to advocate. Were the latter conception 

 perhaps admissible in the case of two ovules, the motive loses 

 its conclusive force altogether when applied to triple and 

 quadruple nutmegs. 



In conclusion I wish to draw the attention to the remark- 

 able fact, also confirmed by Prof. Janse on the strength of his 

 own observations and the assurances of the cultivators, that 

 the compound fruits are produced only by trees of mixed sex, 

 occurring not so very rarely between the, for the rest, dioeci- 

 ous specimens. 



RUTACEAE, a AURANTIEAE. 



Aegle Marnielos Correa. 



Coll. Dr. Elbert, Sept. 1907 on the Kendeng. 



Hab. Java, India and Further India. 



Leaves trifoliate as usual. A deviation occurs only on the 

 terminal leaflet and consists of a little cup. This cup, showing 

 various degrees of development, is nearly always supported by 

 a threadlike stalk, and does not spring from the apex of the 

 foliole but a few^ m. m. lower ; on that spot the midrib leaves 

 the foliole and passes to the stalk of the little pitcher. 



Pitchers may, as is generally known, be born in three difife- 



