Australia has largely depended upon the importation of cultivars for the horticultural industry which have been developed overseas for climatic conditions that are quite different from our own. This has been a short-sighted policy in that these imported cultivars generally do not perform as well in Australia as in their country of origin. If Australia is to remain competitive in export markets of primary produce, it will be necessary for Government to liaise with and encourage private breeders, provide ancillary services at cost and embark on breeding programs at Government level where required.
The Macadamia Industry in terms of other fruit and nut tree crops is a very young industry. The edible species are indigenous to the subtropical rainforests of north eastern New South Wales and south east Queensland. Macadamias were found growing naturally as far south as the Clarence River. (Cheel & Morrison, 1935). The discovery of the first botanical specimen of Macadamia is credited to Leichhardt in 1843 (Story,1959). However in 1828, a botanist and explorer, Allan Cunningham discovered M. integrifolia at the base of Mt Tambourine in southern Queensland. (A.M.S. News Bulletin, 1988). The first recorded commercial planting was at Rous Mill in N.S.W. in the 1880's. At about the same time macadamias were introduced into Hawaii by William Purvis. Although a fledgling industry began in both countries, it never really developed until vegetative propagation problems were overcome. The Hawaiian Agricultural Experimental Station began vegetative propagation in 1922 and was successful in growing cuttings, air layering, inarches and grafting. As early as 1930 the then Queensland Department of Agriculture were experimenting in vegetative methods of propagation with moderate success and also other enthusiasts, including W.R. Petrie were doing likewise (Cheel & Morrison, 1935). Petrie was one of the pioneers of the Australian Macadamia industry and some of the Hidden Valley selections and two of our varieties can be traced back to one of his selections.
In 1936 a variety selection program began in Hawaii, although 60,000 trees were planted in Hawaii at that time, the initial selections were made from 20,000 trees which were bearing. Five varieties were named and two of these, HAES 246 and HAES 508 are still used to this day. In 1937 the Hawaiian industry reached another milestone when it was reported that the causal agent for a successful grafting of macadamia is an accumulation of starch above a girdled branch, which serves as a reserve food to sustain growth after the union is completed.(Shigeura & Ooka,1984).
In Australia, the macadamia didn't really start to develop as a potential food crop until the early 1960's, when C.S.R. Ltd initiated the first large scale planting in Southern Queensland. They were able to utilise the research done in Hawaii in girdling branches prior to grafting and used a modified side graft, which was the method used by Norm Greber, whom they employed as an adviser at that time. C.S.R. Ltd were of the opinion that to initiate a tree breeding program would be a waste of time and money and relied entirely on varieties that were developed in Hawaii for Hawaiian conditions. This was a very short sighted policy and we have paid the price, none of the Hawaiian varieties imported at that time have performed as well here as they do in Hawaii, which is to be expected as it is a totally different environment. I am sure that parallels can be found in other branches of horticulture where we have relied too heavily on overseas expertise.
Perhaps before I go any further I should give you a little information on my background in the Macadamia Industry and how I became involved in tree breeding. My interest in macadamias dates back over thirty years to when we planted our first trees. After spending ten years prior to that growing fruit that had a limited shelf life and harvesting time span, I was attracted to macadamias with their hard shell and long term keeping qualities. We have been processing our own crop since 1976 and we started our present breeding program over twenty years ago. In 1989 we became the first plant breeders in Australia to be granted Australian Plant Variety Rights. It became apparent to me very early in the piece that we were not achieving the yields that we should with the Hawaiian varieties. In those days we used to graft in situ in the field and we always allowed a sucker to come away below the grafted area to see how it performed, we observed that very often the sucker from the rootstock bore heavier crop than the grafted variety. We also noted the heaviest crops were usually on seedling trees of hybrid origin grown in our locality. In the course of over twenty years of the selection process, we have screened more than 10,000 trees from which we have made 428 selections most of which have since been discarded, we have released three varieties to date. The Hawaiians in their original screening released five cultivars from 20,000 trees, so the percentage is not very different. However we believe the Hawaiian genetic base is much narrower, the 20,000 trees in the Keauhou orchard from which the five selections came from were M. integrifolia seedlings from the original Purvis trees and seedlings from the next importation into Hawaii by the Jordan brothers in 1892. It has been stated that it seems likely that those first introductions came from south of Brisbane or from the southern end of the species M. integrifolia, also that the general average quality of the trees and the nuts produced were not as good as those from trees growing in the wild in the Amamoor and Imbil areas. (Beaumont, 1956.) We have here a situation where a small number of seed nuts of doubtful quality formed the basis of an industry, later introductions were made in the 1930's by seed obtained from Walter Petrie and since then by Dr Beaumont in 1954 (Beaumont, 1958). However in my opinion the damage was already done and mathematically it was too far down the track for the later introductions used for poly crosses to make much difference. I believe the early Hawaiian progeny populations were more homozygous than the Australian equivalent which has a much wider genetic base and more variation in seedling populations. Our own experience has been limited to using some of the earlier Hawaiian cultivars as F parents for open pollination and we have found that a large proportion of the progeny are morphologically very similar to the parent trees, so from a breeding point of view they are not very useful because you have to screen far greater numbers. This may not apply with some of the later Hawaiian varieties and selections, particularly if used for cross breeding with Australian varieties. We have found that some local varieties or selections, particularly if of hybrid origin are more heterozygous and produce more variability in the progeny. However there are parallels to the Hawaiian family tree, the Australian variety Hinde or H2 is an example, perhaps the original seedling came from an isolated pocket of rainforest where there was inbreeding depression. Anyway the fact remains that as an open pollinated F parent there is insufficient variability in the progeny and many of these look like H2 grafted trees. Some of these seedlings were planted out in a progeny block next to A4 self seedlings of the F2 generation which have a high degree of variability compared with the H2 open pollinated seedlings. It has been stated 'all the selected and named horticultural varieties now being grown in Hawaii (1956) were derived from the progeny of one introduction of only a few seeds of one of the two important species. A detailed study of the variability within the two species and of their natural ranges within Australia confirms the statement that each has certain desirable horticultural characteristics not fully revealed in the progenies grown in Hawaii.'. (Beaumont,1956). Fig 1 shows the approximate distribution of macadamia at the time of European settlement and Fig 2 shows the distribution at the time that Dr Beaumont made his collection for the Hawaiians.
Early varietal selection in Australia of macadamias was based primarily on screening for kernel quality first and then looking at other characteristics. We believe this is the wrong approach when screening large numbers of progeny. There is no point in selecting a seedling because it has excellent kernel and later having to reject it because it only has a few nuts. It is also extremely costly and time consuming to test entire populations on this basis. The same would apply in other fruit and nut crops. We believe the best approach in screening a progeny population is to select in the first year of screening for precocity and in the second year for yield, trees which have not produced crop within three years of screening for yield are eliminated, the best yielding selections are then assessed for all the other characteristics which we look at. It is true that using this methodology one is probably throwing out the baby with the bath water in some instances and it can be argued that a selection that does not perform well at Hidden Valley Plantations may well perform well somewhere else, which is correct. However it can also be argued that trees that produce very heavy crop in our environment have the potential to produce very heavy crop elsewhere, if yield is a multigenic characteristic, which I believe it is. One could say that in order to achieve the ultimate in yield, it would be necessary to breed different varieties for every region, apart from being impractical the costs would be enormous.
I will now move on to the methods which we employ which I believe can easily be adapted to other crops that are propagated vegetatively. Most of you here today are probably well aware that breeding seed populations is a very different ball game and involves huge numbers, it is doubtful whether say a grain farmer would have the time or the resources to embark on such a project. However in vegetatively propagated tree crops we are talking of the selection of individuals, not populations. In stud sheep or cattle breeding, you are looking at maintaining a certain type in a population and the number of characteristics being assessed are much smaller but the principles are the same and farmers are very successful at it. In breeding race horses, you are looking at individuals and the risks are much higher, a mare only has one foal while in tree progeny populations you are selecting from thousands and have a much greater chance of success, yet there are plenty of private breeders in the racing industry. There are many scientific aids that can and are used in tree selection and breeding, I will move on to that area later. However in the final analysis when choosing a parent, or a selection for future testing after assessing all the characteristics, gets down to a gut feeling and close observations in the field, much the same as breeding animals.
Choice of parents
In a new crop such as ours, very little is known as regards inheritance of characteristics. The results from a controlled cross are a gamble, partly because of the heterozygosity of the species, particularly in Australia, also because we know very little about which genes are dominant, recessive or multigenic. What we do know, has often been learnt the hard way. As an example, when we first began a breeding program, we observed that seedling trees in our locality that had the heaviest yields were of hybrid origin. We happened to select parent trees that had thick shells and out of the first thousand seedlings, I think only one of the progeny had a thin shell, we then knew that thickness of shell was dominant and had to start all over again. However in the choice of parents, it still gets down to the old maxim of marrying best with best in my opinion and basing the choice of parents on one or two major characteristics only. If both parents happen to have say good field characteristics, all the better, but it is a mistake to expect that the bulk of the progeny will perform in a similar manner to the parents.
Progeny Blocks
Each tree in a progeny block is marked on a computer map (Fig. 3) which is taken into the field during preliminary assessment. Each tree has a cell in which notes can be written, prior to being transferred to the Assessment sheets, that is if it is good enough to be given a coded number.
Progeny Assessment
It is important when setting up a breeding program for fruit or nut trees, to consider first those factors which are likely to influence the end product and build those characteristics into your assessment program. The characteristics that you choose can be weighted according to their importance and ranked with the aid of a computer. It is often possible to pick up a selection that you would have otherwise missed by ranking different areas such as Field Characteristics, Kernel or Fruit Characteristics, All Characteristics, Weighted and Unweighted. Needless to say the weightings given depend upon the importance of that Characteristic. In visual assessments we use Industry Standards to measure against. We have been using a computer in tree breeding since 1983 and were probably among the first in vegetatively propagated tree crops to do so. Fig. 4 shows a data recording sheet which I take into the field when assessing macadamias, something along these lines could easily be adapted for other crops.
Basically, we believe the selection process should be along the following lines:
Fig 1 : Distribution of Native Stands of Macadamia at time of settlement
Fig 3 : Sample Progeny Block Map
Fig 4. Sample Data Recording Sheet
Selection Aids
There are many aids which can be used in tree breeding and the selection process, one I have already mentioned - the computer. It is after all only an aid in the selection process, however very often it gives a fresh look at say a selection that you previously discarded. This actually happened with one of our varieties, Hidden Valley A38. We had previously discarded it as a selection some years before on account of several negative field characteristics, however we later decided to include % Whole Kernel in our criteria. A38 turned up on the computer with very high % Whole Kernel and the field characteristics had improved in the interim period, it was then released as a new variety.
Some years ago we found that there was direct correlation between Canopy Area or the photosynthetic area and Yield (Bell & Bell, 1984), we improved upon the original method (Bell, 1986). The technique was further refined and improved and it was found that Canopy base m2 using an average of two measurements across the base of the canopy correlated equally well with yield and related to the maximum return per hectare that could be achieved (Trochoulias, 1992). I am sure this would work well in a lot of other tree crops. The yields and measurements of a known variety, preferably with a variation in size can be graphed (Fig. 5) and used as datum trees to measure against. Top selections can then be put on the same graph to see how they measure up against the datum trees. A projection can also be made as to what yield one can expect when the trees reach a certain size. It is the photosynthetic area that is being measured and I believe deciduous trees would produce similar results if graphed, using the same methods.
Mutation Breeding
The quest is for dwarf varieties suitable for high density planting, we have collaborated with the Queensland Dept. of Primary Industries in irradiating budwood with Gamma rays and also treating seed nuts with colchicine. We were unsuccessful with the Gamma ray treatment, I believe partially because the dosage rate was too low, the only previous experience was in avocado wood which is much softer.
In the colchicine seed treatment there did appear to be slightly more variability in the progeny, but not enough to warrant continuation at this stage. U.V. short wave light used for irradiating pollen which is then used for controlled cross pollination has produced the best results to date. Several polyploids appear to have been produced and also dwarf types from the one pollen source and a higher percentage of mutants than one would expect in a normal progeny population. We have also used a natural mutation for crossing purposes and at this stage it appears we may have several dwarf types which may be useful in the future.
High Density trials
We believe there will be increasing emphasis in any tree crop in raising production per hectare as the price of land goes up and the population increases, particularly on the east coast of Australia. It has happened in tree crops in Europe and it will happen here. We have a trial going at present based on truncating alternate rows when the canopy becomes too dense, then when the truncated trees have grown up, the adjacent rows are truncated. The theory being that the adjacent rows would always be at different heights, higher light levels, easier spraying and less disease problems and erosion because grass would remain in the inter row space. (Bell and Bell, 1993) and (Bell et al., 1994). Fig. 6. We believe suitable dwarf varieties may be the long term answer.
We are also looking at methods for converting existing orchards to new varieties. The approach which we are investigating at present is to skeletonise every second row for Top Working over to improved higher yielding varieties and interplanting in the same row with new varieties (Fig 7). This strategy, which also results in a high density orchard, would mean you would get income from the remaining old trees until the top worked trees and the young trees started to cover costs. The remaining old trees would then be top worked. Cutting back every second old tree should raise the light levels high enough to interplant between the trees and at the same time provide shelter for the young trees.
Fig 5 : Sample Yield vs Canopy Area Graph
Fig 6 : Possible Growth Cycle for a High Density Orchard
Fig 7 : Strategy for Converting Existing Orchards to New Varieties
Disease and Insect Resistance
We have 64 named varieties and selections included in the trial, four replicates of each with guard rows round the outside and the cultivars and selections randomised for statistical analysis. The intention is to introduce diseases and insects when the trees reach bearing age and see which show the most resistance, needless to say the block is in an isolated area. It is hoped that the Entomology and Plant Pathology branches of the Queensland Dept. of Primary Industries will become involved at a later date when the trees reach bearing age.
Propagation from Cuttings
In macadamia, vegetative propagation is done by grafting or punch budding because we are told that you can't produce a good tree from a cutting, which is not correct. This line of thought dates back to the thirties, cuttings were then grown outside in cutting beds from large scionwood sticks several feet long, needless to say you finished up with a lot of bench roots. It is interesting to note that in South Africa the majority of macadamias are grown on clonal rootstocks (cuttings). In most other orchard crops, grafting is done for a reason such as disease resistance, dwarfing, precosity. Disadvantages in grafting or budding include excessive suckering below the graft and rootstock - scion incompatibility. Using modern techniques, it is possible to produce good trees from cuttings and we have been doing that for a number of years. We believe the success of high density plantings in Macadamia will partly depend upon the ability of producing a cheaper tree at nursery level, which we hope to be able to demonstrate.
Overseas Trials and Breeding for Climate Variation
We currently have trials in Hawaii, Costa Rica, South Africa, Malawi, Kenya, New Zealand, Cook Islands, Israel, Argentina and Ecuador. Overseas trials in different environments serve several purposes: It is a way of extending our regional trials at no cost. We receive meteorological information and performance data from areas which would be difficult to duplicate in Australia at the present time, because there are at present no trees growing in similar climatic conditions. As the population increases along the coastal belt of Eastern Australia, fruit and nut tree crops are going to have to extend their present range. In order to do this it will be necessary to breed drought, cold and heat resistant varieties and ones that perform well at high altitude such as on the Atherton Tableland. Many of the countries where we have established trials of Macadamia varieties and selections under Government control, have reciprocated by sending us their varieties to trial out here. Although I don't expect them to perform well under our climatic conditions, they have great value for breeding purposes if we are going to extend our present boundaries in tree crops.
The present change in climate with higher temperatures and less rainfall in recent years has caused more stress to tree crops and this will increase with global warming, so in breeding new varieties we will have to look more towards this factor in the future. In macadamias we have symptoms called 'dieback' and 'sticktights', both of which can be associated with stress, both can be related to higher temperatures and lack of water. The heavier the crop the tree is carrying, the greater the stress.
Summing up, future tree breeding will be aimed more in the direction of smaller precocious high yielding varieties, with the accent on pest and disease resistance and adaptable to wider climatic variation.
Dependence on Ancillary Services provided by Government
It is very important that Government Departments should continue to have input into any private breeding venture, if varieties are later going to be released from that venture into the Industry. One reason for this is that it removes any suggestion of bias in favour of a new variety being promoted by the breeder. In our situation, we have collaborated closely over the years with the following Government Departments:
Queensland Dept. of Primary Industries who have provided numerical kernel data for all coded number selections, they have also assisted in visual assessment of kernel samples and collected yield and kernel data from regional trials in Queensland. In collaborative research areas, a number of joint papers have been published.
The International Food Institute of Queensland have tested our varieties for shelf life and roasting quality.
The N.S.W. Dept of Agriculture have supplied data from their regional trials and have collaborated with the Queensland Dept. of Primary Industries and ourselves in this area. There is also a progeny block of seedling trees from selected parents, mainly of H.V.P. origin which is being assessed on a collaborative basis by the N.S.W. Dept. of Agriculture, Queensland Dept. of Primary Industries and myself.
C.S.I.R.O. and the University of Queensland have also been involved with us for many years and we have provided research facilities and accommodation for several postgraduate students as well as free exchange of research information in areas which are of joint interest.
Technical information and assistance has been made available by the C.S.I.R.O. and Queensland Dept. of Primary Industries when requested and this has greatly assisted us in areas such as microscopy and mutation breeding in our search for dwarf types in high density planting. Government participation in Fruit and Nut Tree breeding programs is imperative if Australia is to export at competitive prices. Increasing yield by breeding better varieties is the most effective way in the long term of reducing costs per hectare of crop and bringing higher return at farm level. Nutrition, pollination , irrigation, insect and disease control all effect the final yield of a variety. However if you want to raise the yield beyond that point, the only option is to breed a better variety.
In recent years politicians have systematically dismantled Agricultural Departments and research institutions in various States, not replaced staff who have been forced into early retirement and adapted a policy of 'the User Pays' and accountability for every dollar spent, to the detriment of Primary Industry in general. Well it is about time they started putting those departments together again, I am not a civil servant so I can say these things. The User will only pay if he is getting value for money and if it comes to that, he will go to a private consultant if he is going to receive better value for services. The one area where we have a lead against many competitors for export markets is in technology and if we can't maintain that lead, then our chances are not good.
The system of Accountability which has been adopted whole heartedly by Australian Government, originated in the United States where they are now having second thoughts about it I understand. Australia and New Zealand are the only countries silly enough to have a tariff free policy, the rest of the world pays lip service. Countries like the U.S.A. can establish secondary or primary industries in third world countries and export product to Australia and at the same time receive preferential treatment as product from a third world country. A good example is where we exported our expertise to Thailand in growing pineapple, the Thais implying that it was for local consumption only, which it was not. Product now comes back into Australia and has sent many local growers to the wall. How can we compete against that sort of thing? In order to reduce Departmental debts and overheads we have even cut staff and services at Quarantine level, sooner or later we will pay for that mistake. Present Government policy is holding back development in more ways than one. A signature on a phytosanitary certificate costs eighty dollars, why?
At Industry level the Australian Macadamia Society Ltd. are in a position where they can deal directly with Government. One of the sub committees of the A.M.S. is the Research & Development Committee. Their task is to liaise and initiate research projects funded by the A.M.S. It is important that in any fruit or nut tree industry that direction for research comes from the grass roots and not simply applications for funding from Government bodies and research institutions, which are then considered on their merit. Priorities should be established by the Industry.
It is noteworthy that while H.R.D.C. provides government funding for industry research projects, it also has the power to dictate which projects are supported. Cutting back on H.R.D.C. funding would be a retrograde step, at the present time the horticultural industry needs to spend more on research if it is to remain competitive against imported produce and in the overseas market. If Government input of funds drop below a certain level, it would not be in the interest of the industries concerned to apply for H.R.D.C. funding, when H.R.D.C. dictate where those funds are spent.
Government should also take a hard look at the A.H.C. policy where the Horticultural Industry not only pays the A.H.C. for developing new markets and sales promotion, but also for a Government administration charge. They are after all competing in areas covered by sections of Industry who are selling their own product in markets that they have developed themselves.
I believe that Government could assist the horticultural industry in varietal improvement of fruit and nut tree crops in the following areas:
As an example, royalties from tree sales covered by Plant Breeders Rights which are sold by licensed nurseries, at the present time cannot be treated as part of the farm five year averaging system and is included in income from investments. The interpretation of the Tax Commissioner is that you have sold your expertise for financial gain. He does not take into account that it takes many years and a great deal of money before one variety gets to the stage of being released and that income from this source goes up and down like a yoyo, depending upon the state of the Industry.
Where urban development is encroaching on existing orchards and farm land, Government should legislate to put the onus on the Developer, not the Grower next door, to provide buffer zones where horticultural sprays are used. Biological control will become more and more a necessity and way of life, however until those controls become the norm, there is no reason why farmers should contribute land for a housing development next door. As long as urban sprawl is allowed to continue, the amount of useful horticultural land close to the cities is reduced and thus raises consumer costs at retail level.
Where there is no input from private breeders, or likely to be, initiate a breeding program. As an example, in custard apples Growers have had to rely on one back yard selection made over fifty years ago and one imported variety, needless to say neither perform well in many areas. No industry will grow unless it can achieve viable yields.
I believe it will be some years before genetic engineering in some of our tree crops becomes a reality. When it does there will still be room for the careful observer in the field, because he will make the initial selections. If for instance we splice a disease resistant gene into another variety which we know is susceptible to the same organism, the resistant selection still has to be found in the field and that can only be done by a trained observer.
Government and Growers should heed the lesson learnt in the macadamia Industry from what happens if you rely on imported varieties when establishing a new industry. They should also take note of the results of our regional varietal trials where no varieties perform in the same manner at different sites and under different climatic conditions. Optimum yields can only be achieved with the right varieties in the right place.
Australian Macadamia Society News Bulletin, 1988. (Author unknown). Vol 15. No 1. pp. 20-21.
Beaumont, J.H. 1956. New Name and Prospects for Macadamia in Hawaii. Calif. Mac. Soc. Yrbk. Vol 2. pp. 15 and 19-20.
Beaumont, J.H. 1958. The Macadamia in Australia and Hawaii. Calif. Mac. Soc. Yrbk. Vol 4. pp. 25-29.
Bell, H.F.D. 1994. What Comes First in Macadamia Varietal Selection? A.M.S. New Bulletin. Vol. 20. No 6. pp. 63-64.
Bell, D.J.D. and Bell, H.F.D. 1984. Relationship Between Leaf Canopy Area and Yield. Proc. First Aust. Mac. Res. Workshop. Eds. R.A. Stephenson and E.C. Gallagher. 14: No 2. pp 1-7.
Bell, D.J.D. 1987. Recent Improvements in Methods for Measuring the Canopy Silhouette Areas of Trees. Proc. Second Aust. Mac. Workshop. Eds T. Trochoulias and I. Skinner. 2: No 4. pp. 49-56.
Bell, H.F.D. and Bell, D.J.D. 1993. High Density Plantings of Cuttings in Macadamia. A.M.S. News Bulletin. Vol 20. No 4. 23-24.
Bell, H.F.D. , Bell, D.J.D. and Gallagher, E.C. 1994. High Density Plantings of Cuttings in Macadamia. 1994 Progress Report. A.M.S. News Bulletin. Vol 21. No 2. pp 57-61.
Cheel, E. and Morrison, F.R. 1935. The Cultivation and Exploitation of the Australian Nut. Technical Education Branch. Technological Museum, Sydney. Bulletin No 20. pp 5 and 13-16.
Shigeura, G.T. and Ooka, H. 1984. Macadamia Nuts in Hawaii : History and Production. Research Extension Services 039. pp. 17-19.
Story, W.B. 1959. History of the Systematic Botany of the Australian Species of Macadamia. Calif. Mac. Soc. Yrbk. Vol 5. pp. 69.
Trochoulias, T. 1992. Rootstock Type Affects Performance. Acta Hort. Tropical Fruits 296. pp 147- 152.