Geniiiimlion of submergcd seeils froiii Jloodplainx 
Table 2. Resuits of statistical analysis (ANOVA) of the differences of germination 
defined as radicle emergence between non-submerged and submerged seeds of 
the twelve study species (P > 0.05). 
F-ratio 
P 
Aldina latifolia 
13.500 
0.002 
Campsicmdni coiiiosa 
3.857 
0.065 
Cecropia latiloba 
9.000 
0.008 
Crateva bentiianii 
3.857 
0.065 
Criidia amazônica 
9.000 
0.008 
Mora paraensis 
21.000 
0.0001 
Nectandra ainazoninn 
6.081 
0.024 
Senna reticidala 
81.000 
0.0001 
Swartzia polyphylla 
100 
0.0001 
Tabebuia barbata 
3.857 
0.065 
Vatairea guianensis 
36.000 
0.0001 
Vitex cyniosa 
13.500 
0.002 
floocling regime on such sites is usually shorterandshallowerthan in 
Amazonian floodplains. Hence, germination under water may have 
some ecological signifieance (DuBairy 1963). 
As long as the seeds of the chosen species were in the water, they 
remained viable. Seeds which were kept in the air dried or decomposed 
within a few days (e.g. Tahehiiia Ixirhata, Nectandra aiuazoinim) or 
weeks (e.g. Sewui reticiihita, Aldiua laíifolia), whereas submerged 
seeds of ten of the 1 2 species remained viable for at least 60 days if the 
water was changed frcqiicntiy (Parolin 2001). This stands in contrast 
to the majority of land plants, whose seeds losc their viability if 
submerged for prolongcd periods (1 look 1 984). Seeds that opened under 
water had a highcr viability than seeds that remained closcd; a highcr 
percentage germinated, and they germinated faster after water rcccdcd, 
comparcd with the non-swollcn seeds of all species. 
327 
cm 
SciELO 
10 11 12 13 14 15 
