Journal of the Royal Society of Western Australia, 86(4), December 2003 
least three years to replenish sufficient resources to fruit 
successfully again (Mawson 1995; Johnstone & Kirkby 
1999). Therefore, the cockatoos can't simply feed from 
the same trees each year; they must somehow assess the 
energy yield of the fruits from individual trees each time 
they fruit. Either they must reassess a tree each time it 
fruits, or remember trees with a high yield from the 
previous fruiting (if indeed high seed yield trees remain 
so in successive fruitings). 
It is important for the future management of these 
cockatoos that we are able to predict which are feed or 
non-feed Marri trees. Therefore we examined the 
possibility of categorising feed and non-feed trees from 
fruit and seed characteristics, independent of site and 
year influences. As we measured 6 fruits per tree, we 
used the mean value for each characteristic for each tree 
to avoid pseudo-replication. There were no differences 
(by t-test) between feed and non-feed trees with respect 
to fruit wet mass (p = 0.710), length (p = 0.781), or 
diameter (p = 0.652). However, there were highly 
significant differences (by t-test) between feed and non¬ 
feed trees with respect to seed number (p = 0.006) and 
total seed mass (p = 0.001). The individual seed mass did 
not differ significantly (p = 0.155), indicating that the 
differences mainly reflected the number of seeds rather 
than their individual mass. Although it is possible on a 
statistical basis to distinguish samples of feed and non¬ 
feed trees, it is more difficult to categorise a single tree as 
feed or non-feed because there is considerable overlap in 
both seed number and total seed mass between feed and 
non-feed trees (Fig 1). A discriminant analysis for 
categorising fruits (using averages of 6 fruits for each 
tree), based on number of seeds (n) and total seed mass 
(m, in grams), since these were the two most useful 
characteristics for discrimination, yielded the following 
two discriminate functions; 
F n = 1.992 n +17.454 m -7.912 
F f = 2.040 n +8.622 m -5.096 
These functions can be used to classify samples for a 
further tree by substituting its values for n and m 
(averages for a sample of 6 fruits); which ever function 
value is higher categorises the tree as feed (F p ) or non¬ 
feed (F n ). The functions correctly classified 71% of feed 
trees and 68% of non-feed trees. However, the practical 
application of the identification of potential feed trees is 
limited by the variability in Marri fruiting on a year to 
year basis (Mawson 1995). Therefore, habitat trees can't 
be selected due to their status as feed trees in a particular 
year. Further study is required to determine patterns of 
feed tree use over a longer period of time. 
This study suggests that Forest Red-tailed Black 
Cockatoos select trees for a higher energy yield per fruit, 
and provides a method for prediction of whether a Marri 
is a feed or non-feed tree in any particular year. 
However, it is important for further studies to determine 
how frequently individual trees are used, and how 
patterns of feeding from particular trees reflect the three 
to five year Marri fruiting cycle. Further investigation is 
also required to determine the number of different feed 
trees used in a particular area over a longer time period, 
and the total number of trees used by individual flocks of 
Forest Red-tailed Black Cockatoos. 
Acknowledgments: We thank Kim Sarti, Bungendore Park Management 
Committee, for access to the park. All Marri fruits were collected under 
licence from the Department of Conservation and Land Management. 
References 
Abbott 11984 Emergence, early survival and growth of seedlings 
of six tree species in Mediterranean forest of Western 
Australia. Forest Ecology and Management 9:51-66. 
Carr SGM, Carr DT & Ross FL 1971 Male flowers in eucalypts. 
Australian Journal of Botany 19:73-83. 
Clout MN 1989 Foraging behaviour of Glossy Black-cockatoos. 
Australian Wildlife Research 16:467-473. 
Cooper CE 1999 The metabolic ecology of cockatoos in the south¬ 
west of Western Australia. Honours Thesis, Zoology 
Department, University of Western Australia, Perth. 
Cooper CE 2000 Food manipulation by southwest Australian 
Cockatoos. Eclectus 8:3-9. 
Cooper CE, Withers PC, Mawson PR, Bradshaw SD, Prince J & 
Robertson H 2002 Metabolic ecology of cockatoos in the 
south-west of Western Australia. Australian Journal of 
Zoology 50: 67-76. 
Garnett SG & Crowley GM 2000 Action Plan for Australian Birds 
2000. Environment Australia, Department of Environment 
and Heritage and Birds Australia. 
Gill AM, Brooker MIH & Moore PH 1992 Seed weight and 
numbers as a function of fruit size and subgenus in some 
eucalypt species from southwestern Australia. Australian 
Journal of Botany 40:103-111. 
Johnstone RE & Kirkby T 1999 Food of the Forest Red-tailed 
Black Cockatoo Calyptorhynchus banksii ttaso in south-west 
Western Australia. Western Australian Naturalist 23:167-177. 
Johnstone RE & Storr GM 1998 Handbook of Western Australian 
Birds Vol 1 Non-passerines (Emu to Dollarbird). Western 
Australian Museum, Perth. 
Mawson PR 1995 The Red-capped Parrot Purpureiceplmlus spurius 
(Kuhl 1820): a pest by nature or necessity? PhD Thesis, 
Department of Geography, University of Western Australia, 
Perth. 
Pepper JW, Male TD & Roberts GE 2000. Foraging ecology of the 
South Australian Glossy Black-cockatoo (Calyptorhynchus 
lathami halmaturinus). Austral Ecology 25:12-24. 
Robinson A 1960 The importance of the Marri as a food source 
to south-western Australian birds. Western Australian 
Naturalist 7:109-115. 
Saunders D & Ingram J 1995 Birds of Southwestern Austialia: 
An Atlas of Changes in the Distribution and Abundance of 
the Wheatbelt Fauna. Surrey Beatty, Chipping Norton, NSW. 
Saunders DA, Rowley I & Smith GT 1985 The effects of clearing 
for agriculture on the distribution of cockatoos in the 
southwest of Western Australia. In: Birds of Eucalypt Forests 
and Woodlands: Ecology, Conservation, Management (eds. 
A Keast, D McFarland, D Milledge & J Trompp). Surrey 
Beatty, Chipping Norton, NSW, 309-321. 
142 
