IN PASSIFLORA KERMESINA. 5 
That the.tube of the calyx is ulterior. 
That the processes or cili? are ulterior, neither appear- 
ing until the ovule have commenced being coated ; and the 
anthers so far perfected as to present parent cells, with 4 
ovary granules and proper coats. à 
Mr. Lindley says, because normal metamorphosis is cen- 
tripetal, the ciliz are petaloid rather than staminal But 
they are between petals and stamens, consequently of cen- 
tripetal metamorphosis; but normally they belong to the 
stamens. ' 
Observation, which does not rely on lucky chances, and 
which argues from things (seen by the eye,) not things ima- 
gined (seen by the mind,) proves that they are neither one nor 
the other. 
IV. 1. Flower bud, sepals removed. 
2. Same, petals and two stamina removed, shewing 
the first outlining of the pistillum. 
3. Stamina in front. 
V. 4. Similar view of a more advanced pistillum. 
5. Vertical view of pistillum. 
VII. 6. Flower bud. 
7. Ditto sepals removed. 
8. Ditto and petals removed. 
9. Pistillum and two stamina. | . 
10. Pistillum long section carried through the base 
of the calyx. 
11. Transverse of ovarium. 
VIII. 1. Base of calyx. Stamina in situ and pistillum. 
2. Long section through pistilla, etc., one stamen 
remaining. 
3. Ovula of same. 
4. Double long section of base now tube of calyx, 
stamina and base of pistillum: this is rather - 
earlier than 1; it shews the filaments to be 
cellular processes from the inner lining of the 
tube of the calyx, and therefore analogous to 
* 
