598 DR. M. T. MASTERS ON THE PASSIFLORACEZ. 
observer to conclude that the bell-shaped organ is a calyx, the outer segments being in 
that case referable to bracts; but in the male flower, where the perianthial whorls have 
the same form as in the female flower, the fertile stamens are alternate with the lobes of 
the corolla, and the glands are opposite to them, showing that the glands are to be 
regarded as the representatives of a second row of stamens. 
In Ceratiosicyos there are five awl-shaped staminodes in the situation of the tubercles 
ot Acharia, the five fertile stamens being alternate with the petals. On the whole, then, 
the evidence is in favour of considering the outer envelope in these two genera a calyx, 
the inner a corolla, though the latter is gamopetalous, an exceptional feature in the 
order, but one shared by the male flowers of the Papayacee. 
A, L. de Jussieu, who was the first to propose the establishment of Passifloreæ as a 
separate order, held the opinion that true Passion-flowers (Passiflora) had no true petals, 
but rather a double calycine whorl, the inner segments of which are petaloid. He based 
his opinion on the circumstance that the sepals and petals are confounded at the base, that 
they wither together, and are not separated one from the other*. But this opinion is 
quite opposed to the analogy with other genera described since the time of Jussieu, and 
is also not in accordance with organogeny, which shows that the sepals are developed 
successively, while all five petals appear simultaneously after the production of the 
sepals. Nevertheless there are some species of Passiflora (tribe Cieca), in which the 
petals are constantly suppressed, as also in the genus Tryphostemma. 
Corona.— Under this general head is included all that series of rings, scales, or threads 
intervening between the petals and the stamens. From their almost universal presence 
in the species of the order, their variety of form, and the part they play in the physiolo- 
gical functions of the flower, considerable interest attaches to them. Sometimes there 
is but one ring ; at other times there are a great number, lining the whole of the interior 
of the flower-tube. The corona is met with in greatest complexity in some species of 
Passion-flower, and in the greatest simplicity in Malesherbie. It may therefore be of 
interest to trace the gradual advances in complexity in the various genera, beginning 
with those in which the structure is simplest, and passing on to those in which it is most - 
complex. 
In Malesherbie the corona exists as a thickened rim or, at most, as a series of 
tubercles projecting from the margin of the tube. In some species of Gynopleura there 
is a similar coronal rim, but of somewhat larger size, and partly deflexed. In Smeath- 
mannia the corona is but little more developed than it is in Malesherbie. In Paropsia 
the corona also consists of a single ring, emerging from the throat of the flower-tube 
(here very short), and divided into a number of ciliated laciniæ. In Crossostemma the 
corona consists of a single row of threads, encircling the base of the staminal tube. In 
Tacsonia the corona is for the most part twofold, consisting of an upper series of 
tubercles or threads emerging from the mouth of the tube, and of a lower membranous 
outgrowth projecting from near the base of the tube, and usually bent downwards and 
inwards. In Barteria the corona is also twofold, the two series being near together; 
the outermost is membranous, the inner one tubercular. In Zryphostemma and Basa. 
* Juss. Ann. Mus. Hist. Nat. vi. p. 100. omes aqu ucc se im -Hilaire in Mem. Mus. v. p. 304. 
P 
