Statistical analysis. Data were tested for
analysis of variance using Genstat 5.2. Dunnett’s test (P = 0.95), comparing treatment
means to controls, is presented where
appropriate. The statistical designs were: 1)
Pruning experiment: 10 trees ´ 5 groups (control,
0.1, 0.5, 1.0, and 1.5 m pruning height) =
50 trees; trees were assigned treatments randomly
within the population. 2) Vase life of E.
globulus and E. cinerea: 24 stems of E. globulus
´ 3 treatments = 72 stems with two stems per
bottle = 36 bottles divided into three blocks
within a growth cabinet. For E. cinerea there
were 36 stems divided into 3 treatments ´ 3
blocks ´ 4 replications with one stem per
bottle. 3) Pulsing experiments: 3 pulsing solutions
´ 4 replications = 12 stems per genotype
with two stems per bottle; bottles divided into
two blocks within a growth cabinet. 4) Eucalyptus
albida pulsing experiment: stems were
pulsed with 4 pulsing solutions ´ 2 time/
temperature regimes ´ 4 replications = 32
stems with two stems per bottle. Bottles were
divided into two blocks within a growth cabinet