IN FUCHSIA GLOBOSA. 419 
deal with such cases in which transformation is subordinate, and 
the coalescence itself is the main point to be considered. We 
may bring the cases of coalescence under two groups—the first con- 
taining the growing together of whole flowers, and of flowers with 
other parts of the plants; whereas the second includes different 
parts of the same flower grown together. 
(a) Coalescence of whole flowers [synanthy], de. — As is generally 
known, not seldom two flowers are developed in the axils of the 
leaves instead of one. In this ease the flowers must have a 
strong tendency to coalesce. In fact such a growing together is 
by no means rare and may be met with in all possible stages. 
The slightest degree is a simple connection of the peduncles, which 
still show the separate parts by a distinct furrow. In a subse- 
quent stage this furrow may disappear, and consequently a perfect 
union arise. Pl. LIX. fig. 33 illustrates this phase, and further- 
more a cohesion of the tubes and the bases of the adjacent sepals. 
In another example—one of the specimens of the collection of 
Prof. de Vries, which abounds in modifications of this sort—the 
sepals of both flowers were disposed in one whorl. Two of the 
sepals had joined each other so completely as to betray a binary 
character by a small incision at the top only. Asregards the petals 
and stamens of the two flowers, they appeared arranged separately 
round their own styles, so that the whole gave the impression of 
two flowers surrounded by a single calyx. 
About another flower we find in our notes : * two flowers grown 
together, one pentamerous, the other trimerous. In the trimerous 
one a supernumerary stamen is present. Ovaries cohering, so are 
the calyx-tubes ; a sepal of the former coalesces back to back with 
asepal of the latter, but their midribs do not exactly correspond. 
Again, we examined a flower belonging to Prof. de Vries with 
the formula S (8) P 8 St8+8 O (8). In this case two flowers had 
completely coalesced. We observed, however, that the stigma 
was divided into eleven slight lobes and the style flat. Though 
eight cells were present in the ovary, still one could see in the 
lower part of the ovary two separate placentas, whioh tended to 
converge higher up, but did not unite altogether. 
A very surprising instance of coalescence is shown in figs. 34 
and 35, drawn from specimens in the de Vries collection. Two 
opposite peduneles have grown together some way with the 
internode between them! The drawing illustrates one peduncle 
continued almost straight, but the other strongly bent. In both, 
LINN. JOURN.—BOTANY, VOL. XXV. 2G 
