342 PHANEROGAMIA. 



thickish, glandular-dotted, deciduous. I have seen only fertile flowers, 

 which appear as if hermaphrodite ; but the anthers contain no good 

 pollen. Stamens 8, shorter than the petals : filaments dilated, linear- 

 lanceolate, glabrous: anthers sagittate, mucronate, adnate-introrse. 

 Hypogynous dish gynobasic, very short, annular, entire. Ovary gla- 

 brous, depressed-globular, four-lobed, four-celled, the four carpels united 

 to the umbilicate apex; which bears a columnar central style that 

 exceeds the ovary in length : stigmas 4, thick and short. Ovules 2 in 

 each cell, probably collateral at an early period, ascending, hemi- 

 tropous; the micropyle superior. Capsule tetracoccous, rather deeply 

 four-lobed, with the lobes obtuse and rounded, diverging so as to be 

 star-shaped, half an inch or rather more in diameter, locidicidal, so 

 that the cocci become two-valved; the exocarp thick and coriaceous 

 when dry, somewhat wrinkled or reticulated; the thin and papeni 

 endocarp separating after dehiscence, glabrous within. Seeds 2 in 

 each cell, or sometimes by abortion solitary, black and shining like 

 those of Zanthoxylum, ovoid, li to 2 lines long; the integument 

 drupaceous; the exterior pellicle being thin and fragile, and connected 

 by a sparing dry pulp with the inner very thick and crustaceous 

 portion : hilum ventral, linear. Embryo straight, nearly as long as 

 the fleshy albumen: cotyledons oval, flat: radicle superior, slender. 



The variety (3., of which there is only an imperfect specimen, with 

 ripe fruit, appears to differ merely in the much larger capsxde (an 

 inch in diameter), with very thick and Ugncscent walls. 



The variety y., from Mouna Kea, is distinguished from the other 

 Hawaiian specimens (of which the foliage is mostly greener than in 

 those from Oahu) only by the longer and narrower leaves (3 or 4 

 inches in length), with a more tapering base, and of thinner texture. 

 It probably grew in a closer forest. 



All the specimens I have seen are in fruit, save one from Macrae's 

 collection ; from which the (fertile) flowers are here described and 

 illustrated. 



Plate 35. — Pelea cluslefolia : in flower and in fruit. Fig. 1. A 

 flower-bud. 2. A petal, from the same, seen from within. 3. A 

 stamen, from the same, seen from within. 4. An expanded (fertile) 

 flower. 5. Vertical section of the same, showing the ovules, &c. 6. 

 Pistil, showing the disk, &c. 7. A seed. 8. Vertical section of the 

 same. 9. Embryo, detached. — The analyses all variously magnified. 



