70 
E. A. CHESTERFIELD ET AL. 
temperatures could be experienced naturally by 
the species involved in the trial. 
The root medium was washed with water and 
the plants fertilized to excess every second or 
third day. initially with 60 mL of nutrient sol¬ 
ution, then with 120 mL as they increased in 
size. Glasshouse temperature and humidity 
were monitored by a continuous recording 
thermo-hydrograph. Occasional roots extending 
beyond the growth tubes were clipped and stored 
in an alcohol and formalin mixture. Fallen 
leaves were collected for dry weight determi¬ 
nation. 
The trial ran for 100 days after which plants 
were measured for diameter at the harvest point 
(Dia) and total height (Ht). After harvest, the 
roots were placed on a fine wire mesh and 
washed free of sand by a gentle spray and hand 
teasing, with final cleaning of attached sand 
grains by forceps. They were stored in an 
alcohol/formalin mixture for root length deter¬ 
mination. Total root length (RL) was deter¬ 
mined using a Comair Rootlength Scanner 
(Richards et al. 1979). Root lengths estimated to 
be greater than 100 m were partitioned for sep¬ 
arate measurement. Roots and shoots were held 
at 65°C for five days and dry weights of root and 
shoot (Dwt(r), Dwt(s)) were determined after 
stabilising for 1 hour at room temperature. 
RESULTS 
Mean monthly air temperature and relative 
humidity in the glasshouse during the trial 
(Table 3) were uniform, with a diurnal fluctu¬ 
ation in mean maximum and minimum temper¬ 
ature of 10°C and in relative humidity of 
approximately 20-30%. Absolute maximum 
and minimum temperatures of 37.7°C and 
1 1.2°C were recorded. 
Increment in growth was obtained by sub¬ 
tracting initial values for each plant in the case of 
height, and means of an initial sample for root 
length and dry weight of roots and shoots. An 
analysis of variance showed all variables to be 
significantly affected by temperature (Table 4), 
with growth reduced at the lower soil tempera¬ 
ture, as expected. 
Height was the only variable for which a tem¬ 
perature x species interaction was significant 
and this was highly significant (Table 4). At the 
higher temperature, height growth of E. regnans 
and E.fastigata was significantly greater than E. 
nitens and E. delegatensis (p < 0,01) which were 
not significantly different from each other. At 
the lower root temperature, height growth of E. 
nitens was significantly greater than that of £ 
delegatensis or E. fastigata (p < 0.001). At 5°C, 
height growth of both E. delegatensis and £. 
fastigata was ca. 60% less than that at 10°C. In 
contrast the height reduction in E. regnans and 
E. nitens was ca. 30% or less (Table 5), Eucalyp¬ 
tus regnans achieved much better height growth 
at both temperatures at this age than all other 
species (Table 5). At both temperatures, the sou¬ 
thern New South Wales form of E. nitens grew 
significantly better than the West Gippsland 
form (p < 0.01), and within forms there was no 
significant difference between provenances. 
A comparison of shoot to root ratios at the two 
temperatures (Table 6) suggests that E. nitens 
and E. regnans differed from E. delegatensis and 
E.fastigata in their ability to maintain stronger 
shoot growth relative to root growth at the lower 
Mean Temperature (°C) 
Mean Relative Humidity % 
Period 
Maximum 
Minimum 
Maximum 
Minimum 
May-June 
26.5 
16.5 
67.0 
43.1 
July 
27.2 
17.6 
67.3 
46.4 
Aug-Sept 
29.1 
15.1 
72.7 
40.6 
Table 3. Mean monthly air temperatures and relative humidities in the glasshouse. 
Variable 
Height 
Diameter 
Root Length 
Dry Weight 
Shoot 
Dry Weight 
Root 
Dry Weight 
Total 
Temp. 
0.007 
0.023 
<0.001 
0.011 
0.005 
0.009 
Temp, x Species 
<0.001 
0.353 
0.433 
0.893 
0.342 
0.822 
Table 4. Probability values from analysis of variance testing the effect of soil temperature and temperature/ 
species interactions. The main effect of species was significant at p < 0.001 for all variables. 
