Sarah Bone, The University of Melbourne, Burnley College


Eight Spotted Gum types (6 provenances of Corymbia maculata and one each of the species Corymbia variegata and Corymbia henryi) were assessed in a field grown  provenance trial undertaken in the grounds of Burnley College, Melbourne Victoria.  Differences in growth rate, stem structure, tree form and health of the provenances were recorded over an eight month period (Nov 2000 – July 2001) from the planting of glasshouse raised four month old seedlings to harvest at twelve months of age.  While preliminary data analysis has identified significant differences in provenances for measures of survival, growth rate and stress tolerance (p £ 0.05) significant differences in stem structure and other measures of tree form have not been recorded.  This paper summarises methods used to assess and measure growth traits of the 160 trees studied and presents a preliminary view of the results (project completion date March 2002).


Urban tree improvement through selection of superior ecotypes (within tree species) is an under-utilised tool in Australian amenity horticulture. Our native trees have received relatively little attention in street tree improvement programs to date.  With the growing trend of municipalities selecting alternatives to eucalypts for street plantings, the need for reliable data on the performance of our eucalypts and identification and selection of ecotypes better suited to urban situations is now greater than ever. Corymbia maculata (formerly Eucalyptus maculata Hook.) has been the subject of a number of provenance studies over the years (Larsen, 1965; Andrew 1969; Darrow 1985 ; Forestry Commission of NSW 1985; Tibbits 1999; Mazanec 1999) none of which have compared ecotypes using horticultural criteria.  The provenance trial described in this paper was established in order to identify and recommend superior ecotypes (if any) of the Spotted Gum group using urban tree selection criteria.

Figure 1 Seed lot numbers and origins for Spotted Gum field trial (left) Map of locations of seed source.


Materials and Method

Eight Spotted Gum provenance seed lots were purchased from the Australian Tree Seed Centre – a division of CSIRO following assessment of a total of ten natural forest stands of C. maculata in NSW and Victoria in June -July 2000.  From available seed lots, eight ecotypes were selected (listed in Figure 1) including one each of two other Spotted Gum species previously known as Corymbia maculataCorymbia henryi  and Corymbia variegata.

Trial Design & Establishment

Germination was undertaken in August 2000 and seedlings were raised in a glasshouse for a period of 4 months before field planting.  Of an average of 120 plants reared from each provenance 30 relatively uniform (in height) seedlings were selected. To offset any tendency to pick the tall poppies (although this was carefully considered during tagging of plants) of the 30 selected, the ten largest specimens were removed to arrive at the final 20 for planting.

An 8 x 35m plot of sandy loam soil in the grounds of Burnley College was cultivated and covered with weed mat prior to planting. 160 trees (20 replicates of each provenance) were planted at 1 x 1.3m spacing within the plot.

Trees were planted following a randomised block design consisting of four blocks of forty trees.  Each block contained five replicates of each provenance, occurring as a randomly planted row.  A buffer strip was planted on the northern and southern perimeters of the plot but not on the eastern or western plot boundaries due to space restrictions.

The trees remained in the ground for a period of eight months.  After an establishment phase of three months, measurements commenced and continued until mid July 2001 when trees were harvested.


Trees were assessed on numerous traits of growth and early structural development – categorised as either growth rate, stem form, tree form or health and stress response (See Table 1). Both numerical and categorical data were collected.  The trees were subjected to several biotic and abiotic stresses during the course of the trial.  These included insect grazing and bird damage during the establishment phase, presence of a fungal pathogen as a secondary stress (isolated yet not identified), cold temperature injury and soil nutrient imbalances. Originally planned to run until the end of August, the trial was harvested in mid July 2001 due to declining growth rates and large levels of dieback and foliage distortion exhibited.  The unexpected stresses did however allow for further comparison of performance of the eight provenances prompting additional measures of health and structure to be recorded.

Table 1.  Growth and development traits recorded in Spotted Gum provenance trial

Growth rate Fresh weight  (above ground portion of the tree) kg (lopped 5mm from base) At harvest
  Fresh weight  (root systems) kg (root ball reduced to 20mm radius) At harvest
  Stem diameter (basal) mm (measured immediately above lignotuber) At harvest
  Stem diameter (at 5mm from base) mm (measured 5mm from base or ground level) At harvest
  Height mm (measured from base to growing tip) 2 month intervals
  Canopy width mm (measure on north – south axis) 2 month intervals
  Lignotuber diameter mm (measured at widest point) At harvest
  Number of laterals Count (all laterals originating from main stem or a dominant stem) Final week
  Root mass Count of lateral and plunging roots based on orientation At harvest

Tree form

Crown shape Categorical 1(round) 2 (oval) 3 (conical) 4 (rounded to conical) Final weed
  Canopy balance Scale 1 (balanced) 2 (unbalanced) Final week
  Branching pattern Scale 1 (good radial balance) 2 (slight imbalance) 3 (severe imbalance) Final week
  Angle of branch attachment (º) Average of  3 angles taken from 1st 2nd and 3rd portions of the tree Final week

Stem form

Habit Categorical 1 (single), 2 (bifurcated) , 3 (multi-stemmed) 2 month intervals
  Number of forks 0 (none) 1 (one fork) 2 (two or more) Final week
  Angle of stem attach

(forked trees only)

(if bifurcated) degrees –where? Final week
  Clear Leader or shared dominance? Count of forked or multi-trunked specimens without a clear leader Final week


  Stem Straightness

(single stems only)

Scale of 1 (very poor) to 6 (excellent) Final week


Condition Scale 1 (dead/dying) to 5 (no apparent problems) 2 week intervals
  Deaths – survival Count At harvest
  Foliage Density Scale 1 (light) to 3 (dense) Final week
  Dieback exhibited Yes/no (if yes in what region of the tree) Final Week
  Tip death Yes/no Week 20
  Branch abscission Yes/no Final week
  Change of leader Yes/no (if yes at what height) Final week
  Lignotuber shoot initiation Yes/no (if yes how many?) At harvest
  Foliar nutrient concentrations %w/w, mg/kg Week 16

Data Analysis

Only preliminary data analysis has been conducted to date.  Two way analysis of variance has been undertaken to establish significant difference (p £ 0.05) between provenances and the necessity for further analysis (log transformation of data, pair-wise comparisons and least significant difference tests).  Categorical data appearing mostly as percentages was subject to arcsine square root transformation prior to initial statistical analysis. Three deaths were recorded during the trial and were removed from the data set prior to analysis. Tables 2-5 display the means for each provenance and the relative ranking of each (if applicable). P-values that have been generated in analysis of variance are given for most of the assessments reported in this paper.  Readers are warned to view results with caution as neither variation within the provenances nor variation between blocks (accounting for site differences and nuisance variables) have been taken considered in this early stage of data analysis.

Results and Discussion

Growth Rate

Table 2. Growth rate. Preliminary results

Table 2. Growth rate. Preliminary results

Although the findings recorded in Table 2 are at a preliminary stage of statistical analysis, the most vigorous provenances are clearly identifiable.  Significant differences between provenances have been found for stem fresh weight (P < 0.05:P = 0.005), fresh weight of roots (P < 0.05:P = 0.000), stem diameter 1 (P < 0.05:P = 0.006), height (P < 0.05:P = 0.000), lignotuber diameter (P < 0.05:P = 0.004) and the number of lateral branches present (P < 0.05:P = 0.001).  Provenances 4 (C. maculata Bodalla SF site 2), 8 (C. maculata Wingello) and 6 (C. variegata – Richmond Range S F) appear to exhibit stronger rates of growth than the other provenances.  These three ecotypes are ranked between first and fourth for every measure.  Conversely, provenance number 3 (C. maculata Mumbula SF) is ranked seventh or eighth for 87.5% of measurements – demonstrating its weakness in comparison to the other ecotypes.

Table 3 Health. Preliminary Results

Of the 160 trees grown only three deaths were recorded all belonging to provenance number 1 (C. maculata Bodalla SF site 2) with 15% of the population not surviving, differing significantly from the other ecotypes (P < 0.05:P = 0.032). A number of low values were however recorded for fresh weights, heights and stem diameters suggesting that trees were struggling to survive in provenances 3 (C. maculata Mumbula SF) and 7 (C. maculata Curryall SF) also.   Provenance 4 (C. maculata Bodalla SF site 1) ranked highest for mean condition rating closely followed by Provenance 8 (C. maculata Wingello), with mean values of 4.05 and 3.8 respectively. Apical death was recorded as significantly different between provenances (P < 0.05:P = 0.001) with 90% of C. henryi (Provenance 5) specimens injured.   Significant differences were also found between provenances (P < 0.05: P = 0.019) for a change in leader as identified in the stem but were not recorded for foliage density, die back, or limb drop.

Stem structure

Table 4 Stem Structure. Preliminary results

The proportion of single stemmed trees was significant (P < 0.05:P = 0.000) with Provenance 6 (Corymbia variegata) recorded as a single stemmed tree 80% of the time.  Additionally, the same provenance was ranked above average to excellent (score 4 or above on a scale of 1 –6) in terms of stem straightness for 75% of assessments – 35% of the trees were recorded as exhibiting excellent stem form (6/6).  Significant differences were not recorded between provenances for percentages of bifurcated or multi-stemmed specimens, additional forking of stems or for shared dominance in forked or multi-stemmed trees.

Tree Form

Table 5. Tree Form. Preliminary results

Most measures of tree form did not reveal significant differences between provenances in two-way analysis of variance. This can most likely be attributed to the nature of the data (categorical) and its method of collection (subjective assessment).  The only differences in measures of tree form applied to trees with an upright growth habit (P < 0.05:P = 0.000). Corymbia variegata exhibited a strongly upright form in 90% of situations and specimens of Corymbia maculata (Provenance 8 Wingello) 80% upright in growth habit. Similarly there were no weeping trees recorded in provenances 6 (C. variegata) and 3 (C. maculata Mumbula SF).  Data concerning radial distribution of branches, canopy balance and mean angle of branch attachment did not uncover any significant differences between Spotted Gum ecotypes.


While preliminary results regarding stem structure and tree form do not suggest that variation between provenances is significant, promising results for vigour (tree growth) and vitality (tree health) have emerged.   Corymbia maculata provenances 4 (Bodalla SF site 2) and 8 (Wingello), and the species Corymbia variegata (provenance 6) have consistently ranked higher than other provenances in statistically significant traits measured.  Conversely, provenances 3 (Corymbia maculata Mumbula SF) and 7 (Corymbia maculata Curryall SF) have been the poorest performers in many aspects of this trial. Whether these patterns of performance will remain fixed after complete statistical analysis of data is currently unknown.  It is hoped however that by the completion date of these research project (March 2002) superior ecotypes of the Spotted Gum group can be recommended to the tree grower industry.


  • Andrew, I. A. 1969 Inter-provenance variation of Eucalyptus maculata Hook. grown in Zambia, Australian Forestry: 192-202.
  • Darrow, K. 1985 Provenance trials of Eucalyptus maculata in South Africa: tenth year results, South African Forestry Journal (133): 12-17.
  • Forestry Commission of NSW 1985 Notes on the silviculture of major NSW forest types, Sydney: Forestry Commission of NSW.
  • Larsen, E. 1965 A study of the variability of Eucalyptus maculata Hook. and Eucalyptus citriodora Hook., Canberra: Commonwealth of  Australia Department of National Development, Forestry and Timber Bureau, Forest  Research Institute.
  • Mazanec, R. 1999 Thirteen year results from a Spotted Gum provenance trial in the Wellington catchment of Western Australia Australian Forestry: 62 (4) 315-319.
  • Tibbits, J. 1999  An evaluation of the breeding potential of spotted gum for timber, Thesis (Bachelor of App. Sc Hons.), Creswick School of Forestry, Institute of Land and Food Resources, University of Melbourne.


Leave a Comment