What’s this site all about?

This blog is a notebook of thoughts on gene expression in avocado, avocado germplasm, avocado climatic adaptation, and the future of avocado in its native forests, orchards, and home gardens.

This format helps me to understand the operation of genes in avocado, and, to a limited extent gives me an insight into the fast evolving techniques for locating genes and inferring their purpose and interactions.

Wild and domestic genetic resources are disappearing at the same time as ‘gene jockeying’ makes important advances. Can gene editing and insertion of genes from other species simply ‘re-mould’ avocado without recourse to conserved germplasm?

Or is this concept flawed by presumptive thinking?

As this blog is written by an amateur, it is bound to have errors. Errors of fact, errors of understanding, errors of ommission.

Corrections are welcomed!

Opinions are my own, and therefore eclectic and idiosyncratic.

Laurie Meadows, 29 August 2019.

Seedling # 2

‘Nice’ but major faults…

This fruit taste good (for the most part), but the tree is unproductive and the fruit has bitterness under the skin and intrusive fibers.

The fruit illustrated above fell naturally from the tree on the 3rd of December 2019. It was ripe on the 8th. So roughly a week to ripen.

This fruit is nearly as good as Hass at this time of year (early Summer), but is in a different flavor class. It has a “nice” flavor, creamy tasting, oil is there but it doesn’t “taste” oily. There is no ‘nutty’ note, unlike Hass. It has a lingering creamy taste in the mouth after eating. The flesh just under the skin is distinctly bitter. This flesh has to be removed for a pleasant eating experience.

This alone disqualifies it from further interest.

The flesh is pale in color, firm, smooth, and the leathery skin peels easily.

It has relatively few fibers, but they are long and unacceptably intrusive.

This fruit is 220 grams. The seed size I rate about medium.

Its parentage is unknown.

The tree is upright, with deep green foliage.

It has no flowers at all this year, and will probably be top-worked.

ref: Hill sdlg

Homologous, Orthologous or Paralogous?

Arabidopsis thaliana, not much more than a ‘steekin little weed’, has proved to be a towering giant in revealing gene structure, function and location in plants.

Genes tend to be conserved over time, even as speciation takes place. It’s been shown to be reasonable to suppose that if a gene found in Arabidopsis does a certain thing with an observable effect, then the same effect observed in a different species may be caused by a gene very similar to that already described in Arabidopsis.

Genes that are common to two (or more) species, and assumed to be from a lineage continuous (up each fork) from the branching point of a common ancestor of the two species were/are termed ‘orthologs’ of each other. But if the gene functioned differently in the 2 species (influenced by, for example, the ‘epigenetic’ effects of other mutated genes in one or both the species) it wasn’t/isn’t of primary importance.

The same gene appearing in two different species, but with no ‘line of descent’, were/are called ‘paralogs’. (In other words, a gene duplicated ‘anew’ in the two species, independently.)

When genomes of Arabidopsis, Vitis, Populus, Solanum lycopericon and now Persea are compared, you can see same or related genes across genera/species. Are they retained from a common ancestor in the distant past and thus orthologs? Arisen anew and thus paralogs? Who knows?

Who the hell cares? What do the genes do? How does the expression of these genes differ (if they do)? What other genes affect (or even effect) the expression of the gene in the taxon of interest (in this case, Persea)? This is what is important and interesting.

So I am going with Roy Jensen’s concept (outlined below) and use the term ‘homolog’ and the appropriate adjective (if determinable), unless referring to speciation events and whole genome duplication events in deep time.

“Genomic biology needs to get beyond semantic issues. It needs to focus on defining those sequence-structure-function relationships that are necessary for understanding both the structural origins of biological function and the molecular bases for the divergence of biological function. So, those of us who study the relationships among sequence, structure, and function should discontinue the use of ‘ortholog’ and ‘paralog’, unless we want to focus on the speciation and gene duplication events that produced functional diversity in homologs.

But, unlike Petsko [2], we believe that genomic biologists need to describe, compare, and contrast sequence-structure-function relationships not only for a complete group of homologs but also for subsets of homologs that share particular attributes. Based on our experiences, genomic biologists need words to describe ‘homologs encoded by different genomes’ and ‘homologs that have different functions’.

To accomplish these needs, we suggest the following adjectives to describe homologs: ‘Isofunctional’ homologs exhibit the same function(s); ‘heterofunctional’ homologs exhibit different functions; ‘isospecic’ homologs are found in the same species; and ‘heterospecic’ homologs are in different species.

Jensen RA. Orthologs and paralogs – we need to get it right. Genome Biol. 2001;2(8):INTERACTIONS1002. doi:10.1186/gb-2001-2-8-interactions1002

Avocado Flower Mutations

Avocado flower ‘petals’ are comprised of 2 whorls, each of three similar-sized petal-like structures called tepals. No one can decide if avocado ‘petals’ are actually sepals or really sepal-like petals – thus the compromise name ‘tepal’.

According to a University of California avocado expert, in some small-fruited species in the genus Persea the subtending tepal whorl is quite a bit smaller than the overlaying whorl. As a result, the flowers in these species look like they have 3, not 6 tepals.

But the avocado almost always has 6.

A Hashimoto flower with a size-reduced tepal whorl

On 9th November 2018, towards the end of the avocado flowering (for most cultivars) I noticed a few flowers in a Hashimoto tree that had one whorl so greatly reduced in size that at first glance I mistook it for a three-tepal abnormality.

Left, normal flower, right 4 tepal mutant

Checking other cultivars, I found a few flowers with 4 tepals. These were mainly in Hass seedlings, Pinkerton, and Gwen. Today, a year later, I noticed several 4 tepaled flowers in a Maluma tree.

4 tepals, equal size, viewed from under the flower

In all cases, the 4 – tepaled flowers have a single whorl. There are no subtended tepals of any size.

8 tepaled avocado flowers – left a ‘Mexican’ type, right, Maluma

Several other Maluma flowers have 8 tepals, as does a seedling expressing many characters typical of the Mexican ecotype.

I haven’t seen an explanation of these phenomena in avocado.

In the case of the reduced whorl resulting in an apparent 3 – tepal flower, maybe a point mutation resulting in a loss-of-function meristem mutation (ptl) in PETAL LOSS (PTL) gene or affectors as discussed by Quon et al 2017 in Arabidopsis?


The 4 tepal flower could perhaps be the result of a different gene action. Quon et al 2017 cite Lampugnani et al., 2013 to note that “growth-induced distortions to auxin accumulation internal to the sepal whorl” may disrupt petal initiation. Several genes control auxin transport and accumulation and when these are compromised auxin levels in meristem tissues may be abnormally low. Assuming, of course, that Persea has homologous genes.

The AUXIN1 gene (AUX1) has variants in Arabidopsis whose partial- loss-of-function mutant form arose as a point mutation (e.g. aux1-7). If there is an homolog in avocado, and a point mutation arose (the most likely explanation), it might result in sepal loss. The loss of 2 tepals in a whorl may result in the creation of a 4 tepal whorl.

The 8 tepal flower is a bit more difficult to explain.

Global Warming & Avocado Breeding

Note: I am grateful to Chris Sayer, Californian avocado grower, for documenting climate effects on avocado cropping via Twitter.

We are on track to retain more and more energy in our climate system

Increasing carbon dioxide emissions from burning fossil fuels will increase the energy retained in our climate system by at least an additional 4.5 Watts per square meter by 2100. This suggests a rise in global average temperature of around 2.5 to 2.7o C.

The consequences

As ocean surface temperatures rise they release more water vapor. As air temperatures rise water vapor levels rise (every 1 degree centigrade allows air to hold about 7% more moisture). This results in heavier downpours. In large land masses such as South America and USA, coastal areas and high mountain chains are hit hardest. Smaller land masses surrounded by oceans are also affected.

In large continental masses the difference between high temperatures inland and lower temperatures on the coast can cause strong, drying winds; and as inland temperatures rise with climate change, so do wind speeds.

Strong winds can snap off or shred mature avocado leaves. Chris Sayer’s rough estimate is a loss of around 5% on these 6 year old Lamb Hass trees.

Rising temperatures perturb high atmospheric circulation patterns, causing large pulses of cold air to extend out from Arctic and Antarctic poles. These can bring unusually intense cold periods and even affect subtropical jet stream patterns, causing warmer and drier conditions.

These are examples of some of the more obvious effects. What other knock-on effects that may occur in the interconnected complexity of the weather system are harder to predict.

What does it mean for avocado growers?

In some areas, higher temperatures may affect the ability of avocado flowers to set. In other areas cultivars that do poorly due to a higher heat requirement for flower set may begin to bear well. If marine coastal fogs decrease in spring, avocado pollination may be enhanced. Initial trends suggest reduction in summer, but it is highly complex and uncertain. Winners and losers. Growers in the more challenging dry Mediterranean climate areas will likely to struggle with sunburn damage to foliage and trunks.

Damage to tender new avocado foliage caused by dry anabatic winds in California, USA.

As mentioned, strong drying winds are likely to increase in strength. Autumn winds can break branches carrying heavy fruit loads. Stronger winter winds can blow off both leaves and immature fruit.

Increases in climate variablity may mean successive challenges within a year – frosts, damaging storms, waterlogged soils, then drought and dessicating winds.

Increased evapotranspiration may ultimately lead to salination, especially where irrigation water already contains undesirable salt levels. Extended periods of drought may reduce the level of natural aquifers.

Managing the challenges

Most weather-related problems associated with avocado culture are already being managed.

Heavy terminal clusters of avocado fruit in Lamb Hass may snap branches in wind storms – but higher productivity per tree might more than offset the loss.
Staking, hedgerowing, and other innovations cost time and materials, but long term these investments may pay off in both monetary and peace of mind terms. (LH = ‘Lamb Hass’).

Wind damage can be mitigated with shelterbelts, staking, hedgerowing.

Frost damage can be mitigated with sprinklers.

Appropriate mulching conserves moisture, builds organic content of the soil and improves surface drainage.

Drought can be managed with irrigation and mulching.

Salinity can be managed with salt-tolerant rootstocks and irrigation leaching. New technology can help guide appropriate decisions.

This photo illustrates management mitigation techniques – kaolin sprays, staking, mulching. These newly planted Lamb Hass trees are also irrigated.

Sunburn can be dealt with via talc/kaolin clay sprays.

Rainstorms can be managed with improved subsoil drainage, planting on mounds, ditching, and soil structural improvements via deep rooting cover crops.

At this point, it seems likely that future shifts in duration and intensity of adverse climatic events can continue to be dealt with by existing and new management practices. High temperatures in Spring that exceed a critical threshold for fruit set, and summer temperatures sufficient to damage fruit seem to be the only exception.

Is there really any need to genetically ‘improve’ avocado as a crop to meet higher wind energy, higher global average temperatures, and more intense rainfall effects?

Breeding avocado is more expensive than breeding other crops

Avocado is very difficult to ‘breed’ in the classical sense of controlled pollinations. Worse, time from seed to fruit ranges from about 5 years to 15 years. It is an expensive and difficult crop to breed, even with the aid of new genomic techniques. New characters could be added to meet climate change (wood tensile strength was mentioned, but characters affecting tree size, trunk caliper, and so on could be considered). Each new character added requires generations of breeding cycles, and sometimes very very large populations to find rare genetic recombinations.

Solving problems with management techniques is much quicker than waiting for plant breeding results.

And it seems to me the only problem that can’t be solved by management is heat stress at flowering and summer fruit maturity.

The answers to the heat stress problem

Heat stress at flowering may prevent fruit set. Summer fruit can be severely damaged in inland heatwaves. Regular and prolonged whole-tree heat stress is likely to damage important physiological processes. It is ‘game over’ at that point. Growers must choose:

  • relocate to a suitable climate
  • grow a different crop
  • fund a cross-species genetic modification or gene editing program that allows flower set under high heat conditions AND confers physiological resistance to whole-tree heat stress
If climate and soils are unsuitable in one country, large vertically integrated growing/shipping corporations can develop large orcharding operations in countries with a more suitable climate

The only growers that are likely to relocate (within a country, or to a different country entirely) are corporate growers. Relocating is expensive. Only corporate growers have the capital (or can attract it). Inevitably there is a delay of 4 or so years before the first useful crop volumes flow from new plantings. Corporates can wait years for a return on investment.

Smaller family-based farmers have less financial flexibility. In some hotter inland areas away from ameliorating coastal ocean effects, avocado orcharding will have to be abandoned.

Strawberries crop the year of planting, cilantro in 6 weeks or so, but avocado takes 5 or 6 years for good crop levels. But flipping’ between annual crops is also challenging. Different annual crops require different equipment, different skills.

Most medium to smaller growing enterprises are far more likely to change crop, give up growing (especially if their land is being encroached by suburbia), or grow a different (probably short-cycle) crop.

In short, moving countries is an effective heat stress mitigating measure, and requires no further plant breeding effort.

Avocado nodal groups already carry climate resilience

Avocados of the tropical group (so-called ‘West Indian) grow and fruit well in spite of tropical temperatures. ‘Guatemalan’ avocado nodal group generally fruit well in the warm temperate+ to subtropical zone of the South American Andes. Many of the more northerly ‘Mexican’ node tolerate cooler conditions, including some frost resistance.

The driver of this wide band of climate tolerance seems to me to be a once contiguous wild avocado population experiencing wide temperature differences with both latitude and altitude. (More complex local climate effects within parts of the broad range may also have played a part.)

Localised population genetic diversity has probably been fragmented and thus ‘pruned’ by human deforestation from shifting ‘slash and burn’ agriculture, and more recently by forest clearing for coffee growing. The case of the cultivar Nabal illustrates the point:

“The parent tree was accidently destroyed in June 1917, by a laborer who was planting coffee. It stood among coffee bushes in the Finca ‘Santa Lucia,’ 7a Calle Poniente, near the Alameda de Santa Lucia, Antigua, Guatemala. “

Wilson Popenoe, 1917

If Popenoe had not collected budwood and shipped to the USA over a hundred years ago, the Nabal cultivar would not exist today. And neither would cv. Reed, as Reed is a seedling of Nabal.

‘Guatemala’ is a relatively small geographic area, and possibly has greater depletion of diversity than its much larger neighbour. It seems to me, the greatest amount of avocado genetic diversity is likely to be in local trees in isolated villages, farmlets, and wild places throughout Mexico. And as some parts of Mexico experience dry heat, perhaps some genetic adaptation has occurred.

Genes for heat tolerance can’t be seen

Heat tolerance is probably mediated by a number of physiological processes regulated by a number of interacting genes. ‘Classical’ avocado breeding is primarily guided by easily measurable physical attributes – fruit size, flesh quality, fruit storage quality, crop load, tree size and so on. Avocados historically have been selected in climates well suited to avocado culture and production. The ability of the plant to tolerate heat isn’t revealed until plantings are tried in marginal climatic zones.

Problems in breeding for heat tolerance

If tolerance to heat has evolved in the hottest part of avocado’s natural range, then that is where germplasm should be sought. Not in commercial orchards, but farmlets, villages, towns, wild places.

Historically, the primary selection filter was fruit quality, coupled with reliable productivity. Tolerance to heat and cold was a bonus. But as the ‘pedigree’ of avocado in commercial use is narrow (excluding tropical avocados) there may not be any marked heat tolerance within this rather small subset of avocado diversity. (This is not to say that there is no useful selectable variability in existing germplasm collections. Avocado cultivars are very heterogenous, and their flowering system tends to maintain that heterogeneity.)

I don’t suppose anyone has ever attempted to characterize what ‘heat tolerance’ in avocado is composed of (in terms of which genes affecting which physiological processes, and in what manner). It is certainly possible to do this, and a similar process has been carried out by Prof. Neena Mitter at the University of Queensland to identify genes involved in Phytopthora resilience. Prof. Mitter’s work identified 17 genes involved in a variety of physiological processes apparently responsible for the Phytopthora resilience phenomenon. ‘Heat tolerance’ may not involve as many genes, but is likely to be multigenic.

If genes are identified, then the process of constructing and patenting gene-markers can begin. An expensive process, and probably an expensive product.

Plant exploration is also expensive, and in the case of invisible characters, difficult. I also doubt anyone has gone in search of ‘heat tolerant’ avocado plants in Mexico.

If an apparently heat tolerant seedling can be found, the fruit will likely be inferior – small, big seed, probably thin skinned. Transferring the wild genes from putative ‘heat tolerant’ individuals into commercial cultivars would be a long job.

Why breed for heat tolerance?

Why bother? Who would do it? Who would benefit?

Outside the tropical avocados, at least, this question implies that breeding for heat tolerance, whether by conventional or genetic manipulation techniques, would only happen if Hass variety was threatened on a relatively wide scale.

In such an ’emergency’, I would hope ‘science diplomacy’ would steer the task, with a cooperative effort for the common good, and the tab picked up by everyone.

Temperatures that would threaten Hass avocado

Temperatures above 35 °C (95 °F) have been quoted as adversely affecting fruit set. I am not sure if this is due to warm dry air (perhaps plus wind) desiccating the flowers in the crucial late spring period, or if this is a temperature level that triggers abiotic stress gene up-regulation. This might result in flower or fruitlet drop, or maybe early-stage pollen or ovule damage or death.

Will temperatures rise above 35 °C in the critical spring period?

If so, will such effects be localised to more inland and lower latitude areas of the Mediterranean-climate avocado producers (USA, Israel, Spain, North Africa)?

Will the speed of climate warming outpace the speed of breeding more heat tolerant avocado cultivars?

We don’t have answers. Yet.

Increasing general resilience

Avocado is not an important food in the well-fed West. Objectively, at only 2% protein and about 20% oil, it hardly stands out nutritionally. In contrast, for example, low THC hemp seed has much more protein (around 25%) and contains about 30% oil (which contains exceptionally high levels of the 2 essential fatty acids linoleic acid and alpha-linolenic acid), and unlike avocado, hemp is relatively drought tolerant. But avocado – with corn and beans – is an important food for poor Mexican folk. The trees barely need cultivation, are long-lived and fairly reliable producers.

But avocado will always have a place in Western domestic economies, even in times of economic constraint.

Therefore, the avocado ‘resilience system’ should be strengthened as a matter of general food security. Security is the essence of a social good, and its delivery is the responsibility of government.

In the case of securing maximal resilience in the avocado system all governments should cooperate in funding avocado germplasm conservation everywhere (resilience depends on multiple redundancies), but most particularly in the natural range.

The accent should shift from selecting existing cultivars with ‘good’ fruit or fruiting to collecting a range of diversity, including diverse tree forms. Mutants, particularly dwarfing mutations, should also be curated widely.

Universities and other research organisations should continue to cooperate to better understand and explore the gene-environment nature of avocado.

Breeders, both private and public, should generally share information to aid insights into avocado breeding. Non-commercial cultivars should also be shared (and phytosanitary requirements rigorously observed).

If avocado fruit are a public good, then the consuming public should contribute a small amount of each avocado purchase price to an international, consensus-driven Trust tasked with maintaining avocado diversity in perpetuity for the common good.

Ironically, a somewhat similar arrangement has been put in place for conserving germplasm of coffee, a tree whose industrial culture is responsible for destroying a part of avocado diversity forever.

Perhaps coffee drinkers should pay ‘reparations’ for the damage they have done. The money could then go to a community fund to enduringly avocado germplasm for the common good.

Further reading

David Putland, 2011. Climate change and climate policy implications for the Australian avocado industry

Seedling # 1

Seedling #1 of a series

One of the better ‘home garden advantage’ avocados. From time to time a home garden seedling will be superior to Hass in the early part of the Hass season. And that is the case here.

Picked 18 September, it was ripe on the 25th. So roughly a week.

This fruit is better than Hass at this time of year (early Spring), oily, rich, but not as nutty as Hass.

The flesh is between soft/firm, and the very thin skin peels OK (ish).

It has some fibers, albeit they are not particularly intrusive. Like many thin skinned-green fruit, it develops black spotting of the exterior skin when ripe. These spots don’t intrude into the flesh, at least in this season, they are cosmetic.

The fruit (so far) vary quite a bit in size, from about 150 grams to about 290 grams. The seed size varies from small (as illustrated above) to large in larger fruit sizes (below).

It is a Gwen seedling. The pollen parent, as usual, is unknown.

Like Gwen itself, the tree is ‘temperamental’, soaking up nitrogen without greening up satisfactorily. To be fair, I can’t discount the beginning of Phytopthora infection. But it does have the virtue of not being excessively vigorous.

It seems to be an ‘A’ flowering type, and at this point it has a long flowering period, starting with a modest winter flowering, and, like Carmen Hass, setting a few winter-set fruit.

It remains to be seen whether or not it will fruit well in the longer term.

ref: 42

Author’s Motives

Laurie Meadows, the author, Pinkerton, the avocado.

Whether we are consciously aware of it or not, I believe that we act in what we believe to be our own best interest, all the time.

So it must be in my best interest to allocate time write on this subject.

I am a ‘non-scientist’, retired, a home gardener on a ‘lifestyle block’, and my interest in avocados is a subset of a life-long general interest in food plants, particularly fruit and nuts.

I started out with the idea of growing and selling fruit of some of the ‘old school’ and some of the new avocado varieties (to cover some of the costs of my hobby when on a pension income), but in the end I decided it was too much work for too little reward – especially as the few good areas for avos here have increasingly been affected with phytopthora.

Anyway, ‘gardening’ for me is pretty much planting the tree and doing very little else thereafter. The results – or lack thereof – reflect my negligent style. Not a good attitude for a commercial avocado grower!

The non-monetary reward, really, is that it is quite interesting fiddling around with edible plants, and because I am a home gardener I can grow any ‘uneconomic’ plant I want.

As a result, we have a smattering of pecans, bags of macadamias, a variety of cherimoya, and so on. Usually bird pecked or possum-damaged, but it doesn’t matter that much, especially when my partner is a compulsive bottler and jam-maker.

I came to realize that ‘avocados’ are more than ‘just Hass’, so they excite my ‘huh, that’s interesting’ gland. Which, as my off-sider points out, tends to bouts of oversecretion.

Over the years a small number of avocado trees resulting from random seeds originally planted as rootstocks have fruited. Self sown seedlings have also popped up and fruited. It is always fun to see what the fruit are like, because the home gardener is not shackled by the tight requirements of commercial cultivars.

The home gardener doesn’t give a toss if the fruit are small, big, thin skinned, subject to scarring, doesn’t keep, is a funny shape, the wrong color, crops only modestly etc.

But now that the average size home garden is not much bigger than a sheet of A4 paper, the home gardener does care a great deal about how big the tree is.

So, in retirement, I am throwing in a few seeds of whatever interesting (non restricted) avocado cultivars I can find in the hope – and it is a very faint hope indeed – that a small avocado tree with a small ‘footprint’ can be found for todays micro-garden. But where to find interesting seed-parent trees?

Sadly, the ‘official’ New Zealand avocado variety collection (curated with 50% taxpayer funding) was mostly chainsawed in the late 1990’s. Wish I knew their plan before the deed was done. Still, as my mom-in-law used to say “if wishes were horses, beggars would ride”.

I am not an avocado breeder. Avocados are all but impossible to ‘breed’ in the conventional sense of controlled pollination, large numbers of progeny, re-selecting and re-crossing and so forth. And as seedlings take 5 to 20+ years to fruit (if they fruit at all) planting a few seeds late in life seems a little futile.

But it is interesting and fun.

And in the unlikely event a small tree was found, I would be very happy to have contributed to the dietary quality of the average householder. And if the tree was really small, it might fruit in a large pot, and become part of a mobile orchard. After all, dwarf apples and peaches will fruit in pots, so why not a dwarf avocado?

Will a dwarf avocado come from commercial programs? Doesn’t seem like it right now. Dwarf avocados were once bred in Mexico, but now seem to have fallen off the radar. Perhaps they weren’t ‘commercial’.

There is some hope. First, a small side trip into the driver of modern fruit breeding and the introduction of contract law instruments is required.

A major problem for commercial growers is that avocados are now reaching the point of being a commodity. Commercial success comes from being a low-cost producer, and not too far from your market. These conditions are found in only a limited number of countries with ideal climates, cheap labor, political stability, plentiful water, and good geographic location. So what’s the ‘fix’ for the rest of us who don’t have these advantages?

Well, in the kiwifruit, stonefruit, and pipfruit industries the unstoppable trend is to breed unique and desirable fruit varieties that are exclusively licensed to a small ‘club’ of limited number of growers worldwide. The idea is to limit supply and keep up the prices for these elite varieties. And they are branded and promoted as a being ‘a cut above the rest’.

The trees don’t usually make it to the home gardener unless they fall dramatically from commercial favor. If the ‘club’ no longer uses them, they may (or may not) allow them to be released under plant patent law.

Before the so-called ‘club fruit’ concept caught on, home gardeners used to receive the commercial ‘cast-off’ varieties of the horticultural industry. That now seems to be coming to an end.

As a result we only have the ‘old school’ avocado varieties that were available decades ago – if they still exist. That seedling growing from the compost in your back yard may be the only source of a new avocado variety for the home garden – as long as it has some fault that disqualifies it from becoming a commercial cultivar! And chance seedlings do have a small history of making ‘the big time’.

After all, today’s commercially dominant Hass variety grew from an ungrafted rootstock tree grown from seed of the Lyon cultivar. Lyon, very likely the seed parent of Hass, was in turn a home garden tree growing in Mrs.Lyon’s front yard in – yes – Hollywood!

The irony is that as commercial cultivars improve – in part maybe due to application of gene technology – every home gardener who plants an avocado seed has a better and better chance of coming across a good quality avocado. Whether it is small enough for the backyard is a different matter.

As human induced climate change pumps more and more energy into the weather system we may have much higher energy winds. This may drive an effort to develop smaller trees capable of withstanding wind damage.

Marker assisted selection might accelerate the hunt for smaller trees with leaves that don’t snap or shred in high winds.

Eventually, such trees, even with higher-priced ‘elite’ fruit designations will be producing fruit for the retail market. Chance seedlings will be more likely to be be smaller – suiting the home gardener.

So there you are, my motive is fun. It’s a tremendous lot of fun cheer-leading from the sidelines of a commercially-funded ‘science-game’ in which the average person barely understands the rules, and can only ever be a spectator.

Trying to gain a level of understanding of the rules of the ‘molecular game’ is also a major reason for this blog, even when the molecular work is of no direct importance to my interests.

But from a home gardeners perspective it’s fun to throw a few seeds in the ground and see what you might get. E v e n t u a l l y…!


Authors’s Motives

Avocado genome, publication 2019

breeding, practical difficulties, overview

breeding, for heat resistance, is it worth it?

climate change, effect, avocado culture

cultivar, avozilla

cultivar, Lyon, origin

gene expression, fruit size
Cell Number Regulator/Fruit Weight
fruit elongation
cytochrome P450, 78A class

gene expression, tepal number

gene, AUXIN1

gene, PTL

gene models, Persea americana var. drymifolia. (Predicted and annotated)
gene models, Persea americana cv. Hass (Predicted and annotated)

germplasm, New Zealand, destruction of

Homologous, orthologous, paralogous

seedling avocados, worth

seedling avocados, home garden


The Rose and the Avocado

Sam McGredy’s ‘Oranges ‘n lemons

Today I learned famous rose breeder Sam McGredy recently died.

I met Sam when he brought some rose scions into the New Zealand Government MAF office I was working at. He came to arrange the International Phytosanitary Certificate necessary for shipping them overseas. Sam was a genial man, and naturally the conversation turned to his work as a plant breeder.

He was a great storyteller, and he told me a story relating to the nitty-gritty of the life of a working plant breeder. He was working in the heat of his greenhouse, sweat dripping from his brow, engrossed in transferring pollen from the chosen paternal variety onto the stigmas of the maternal parent, when he became aware he had an unexpected visitor.

A polite and very earnest young Japanese man introduced himself as a beginning rose breeder interested in picking up practical tips from the master. Noticing Sam was using a finger to transfer pollen, he asked Sam if he used alcohol to sterilise his finger between crosses. Sam had to explain that he simply wiped his finger clean on his profusely sweating brow!

At one point his operation was generating 60,000 seedlings per year, so speed in physically making the crosses is essential.

The results of his methods speaks for itself.

Sam was instrumental in setting up plant breeders rights in New Zealand, and his floribunda rose ‘Matangi’ was the first plant ever licenced under the new legislation.

According to Sam, some of his creations were so popular that some nurseries overseas illegally propagated them – or understated the numbers they were propagating – so avoiding the breeders rightful royalty.

Sam was familiar with the major nurseries, of course, and he told me he would hire a light aircraft to fly over the nurseries he suspected of cheating on him. It was relatively easy to spot blocks of his cultivars in flower from the air, and assess the size of the stock!

On the same day I read of Sam’s passing, I also learned the avocado genome has been sequenced.

The work was a collaboration between workers in Mexico, Sweden, USA, Canada, Belgium, Singapore, Australia, Spain, Denmark and France.

This achievement is the key to the door to finding and naming avocado genes that express traits of interest, from flower bud initiation to fruit skin color.

The effort to bring this project to reality is impressive. Multiple institutions across countries, complex models, sophisticated software, highly technical biochemical processes.

Why is it so important for avocado breeding?

Because conventional breeding of the sort that Sam McGredy did is all but impossible for avocado. Breeding for multiple traits – moderate tree size, self fertility, good production, high quality taste, no discernible fiber in the flesh, moderate seed size, skin that peels easily, skin that is not too thick and not too thin, fruit that ripens evenly from base to neck, fruit of ‘the right size’, fruit that cool stores well – requires high populations to select seedlings from, and it probably requires multiple unrelated families under selection at the same time.

Worst of all, it is prohibitively difficult to succeed in deliberate hand pollinations – which means difficult to produce seedlings whose parentage is known with certainty.

The bottom line is that very large hectareages need to be locked up in slow-to-flower trees for a long time, even while there is uncertainty of the genetic worth of one parent. And even when there is almost no information on the mode of inheritance of the characters of interest the parent (s) display.

Someone has to provide the land, the running costs, and the highly trained expertise. But not just the expertise, but the experience and insights that can only build up through years of interaction and observation with the test subjects. Given the difficulties in breeding, that’s not an easy situation to sell to anyone!

But now that the genome has been sequenced, it opens the way to use the genetic code for traits of interest that has been identified in other species (from tomatoes to poplar trees) to be compared to the raw code of the avocado genome. If it matches, even partially, some part of the avocado genome, then that section of the avocado genome may be coding for that trait.

Once a match is found and confirmed as being the gene controlling the trait of interest, genetic ‘probes’ based on these snippets of code can be artificially constructed (and presumably patented). These probes can be run on tissue samples from avocado seedlings to see if that particular seedling carries the desired quality. Any that don’t can be thrown out – saving very large amounts of time, space, and money.

The same check can be run over and over on the same seedlings, but using different traits of interest as the screen. For example, in principle, a probe could be developed for every trait of interest mentioned above. Of course, it’s not as simple as that. Someone, sometime, has to find and describe the gene of interest in another species and deposit that information in the communal database. But traits may be controlled by more than one gene. Or other, unknown genes may act as on/off switches to expression of the genes of interest.

All this takes time, effort, and money. And access to money always has competing interests.

For simple single-gene mediated traits it may be simpler to copy that gene from an avocado tree with a gene expressing that trait, and simply paste it into a seedling that is perfect in all respects – except one. Perhaps the seedling has everything the industry wishes for, but has green skin, and the industry considers a color change to indicate ripening is desirable from the consumer viewpoint. If purple skin color was controlled by a single gene (and no other gene interfered with its expression), than a copy (or copies) of the ‘purple skin’ gene could be inserted via gene editing technology.

So this is a significant advance. But it also signifies a shift. A shift of resources to those with the necessary institutional money and technical sophistication. It raises some very interesting questions about cooperation, patents, funder oversight, collaborator rights across international borders, affect of domestic foreign policy on projects (the domain of science diplomacy), and more.

The difference between the methods of Sam McGredy and these methods is obviously very great. But no-one should imagine that, in general, conventional methods have been made obsolete. Very far from it. But in the specific case of avocado, the door to a slow acceleration in progress is creaking open.

But nothing changes the question – what exactly are you hoping to achieve, and what facts and projections have you considered in deciding on your goal?

Avocado Seedlings

Avocado seedlings can pop up almost anywhere in the garden.

We had one grow near our incinerator, another came up in the compost heap, then there was the one by the deck, the seedling tree right outside the kitchen window. All these trees fruited (eventually). We named the fruiting seedlings ‘incinerator’, ‘compost’, ‘deck’ and ‘kitchen’ – thereby displaying a keen sense of geo-location, rather than a lack of imagination.

As we have a small rural ‘lifestyle block’ with some patches of soil suitable for avocado, we were able to plant 10 or so cultivars, as well as planting a few seedlings for later grafting (mostly from commercial fruit, mostly Reed, some Hashimoto, a few probably Fuerte).

My grafting skills are rather poor, and some of these few ‘rootstock seedlings’ grew on to produce fruit. I left a rootstock stem ungrafted on five of the rootstocks I did successfully graft. Although these trees are currently very badly affected by Phytopthora, several have had some flowers and even produced a pathetic sized fruit.

As anyone with a group of larger avocado trees will confirm, seedlings germinate endlessly from fruit fallen under the canopy. Like most people, I spray them out. But the odd one escapes attention.

Almost all self-sown seedlings are commercially useless – as are deliberately sown seedlings (whether sown by plant breeders or a home gardener.) Almost all, but not all seedling are useless. Self sown escapee seedlings in commercial avocado orchards have given us the commercial varieties Pinkerton, Maluma, and Reed.

A few avocado cultivars are more likely to give worthwhile seedlings – even if there is a fault (such as thin skin) which prevents them from being commercially acceptable. The best seed parent is Hass, and as it is almost the only cultivar grown (in Mediterranean and warm-temperate climates), most seedlings will have Hass as the female parent. Hass is self fertile – a big part of its commercial success – and so many chance Hass seedlings will be selfed.

Dr. Bob Bergh’s experience (Bergh 1976b) is that Hass is “probably the best of all progeny-tested parents”. Of 400 selfed seedlings, 18 were found worthy of selection for further study (1 in 22).

Since Bergh’s time at the University of California, the emphasis has shifted. Gwen, a seedling of Thille, has been found to be an outstanding maternal parent. Gwen and Gwen progeny (Harvest, Nobel, Marvel, Gem, 5 -552, Lamb Hass) have been used very extensively in the Californian breeding program.

But Gwen has never really gained much traction against Hass, the fruit are simply not available in New Zealand shops (and barely available in the USA) so the average random seedling is still likely to be a Hass seedling. And if it fruits it is likely to fruit at around the same time as Hass – and is highly likely to be inferior to Hass.

From a home garden perspective, the best seedling would be one that fruited in winter, and is good quality. Fuerte should be the number one choice for winter, but it is a large spreading tree, and is a very erratic bearer.

Fuerte seedlings, whether deliberate or accidental, are a waste of time and space. Selfed Fuerte seedlings were “markedly inferior” according to Bergh (1976b). He went on to comment that most [selfed] Fuerte seedlings “never set fruit, and most setters have fruits of deplorable size, shape and seed ratio”.

Mexican avocado cultivars have a much shorter time to maturity, are oily, and seedlings of Mexicola, in particular are quite precocious. Pity the tree is huge, and the cultivar no longer available (in New Zealand, at least).

In the long run, molecular techniques enabling selection for specific desirable characters will result in more and better commercial cultivars (hopefully, including season extension into winter).

As a result, in the distant future, home garden seedlings, whether chance or planted, will be much more likely to be desirable. But it’s a slow process, and it’s probably many decades away.

In the meantime, in the spirit of Dr. Berghs short reports on seedlings that ‘didn’t make it’, I will post photos and comments on random home garden seedling found here, both past and present.

Check the index under ‘seedlings’.

Bergh, B. O. 1976(b). Factors Affecting Avocado Fruitfulness. In: J.W. Sauls, R.L. Phillips and L.K. Jackson (eds.) Proceedings of the First International Tropical Fruit Short Course: The Avocado. Pages 83-88.
Gainesville: Fruit Crops Dept., Florida Cooperative Extension Service. Institute of Food and Agricultural Sciences, University of Florida

Bergh, B. O. 1967. Some late-maturing avocado seedlings of various parentage.
California Avocado Society 1967 Yearbook 51: 131-158

Bergh B.O., and Whitsell R. H. 1973. Self -Pollinated Hass Seedlings.
California Avocado Society 1973 Yearbook, 57: 118 -156

Bergh B. O., and Whitsell R. H. 1975. Self -Pollinated Fuerte Seedlings.
California Avocado Society 1974 Yearbook, 58: 128-134

Avocado Fruit Size

Some avocado fruit are very big. Wild Persea species tend to have fruit that are quite big relative to other wild fruit. But relative to cultivated avocado fruit, wild Persea fruit are quite small.

In turn, ‘normal’ commercial avocados are rather small relative to some of the largest cultivated avocado seedling variants!

A 2014 paper by Monforte et al ‘The genetic basis of fruit morphology in horticultural crops: lessons from tomato and melon’ found genes for fruit weight (likely a proxy for size) “co-localized frequently with members of the [Cell Number Regulator/Fruit Weight] CNR/FW2.2 and KLUH/FW3.2 families, as well as co-localizations between [Ovate Family Proteins] OFP family members”. My expansions are [bracketed].

KLUH gene may affect cell division, possibly increasing the number of cell layers in the fruit pericarp, thus affecting ultimate fruit size.

The CLAVATA (CLV) gene in tomato has a signal-receptor complex affecting meristems, wherein a mutation in CLV3 (whose secreted peptide signal binds to the CLV1 receptor) has resulted in larger fruit size in the course of domestication. (Xu et al 2015)

Other genes possibly implicated in regulating cell division are SUN (regulating fruit elongation), and CYP78A (cytochrome P450 of the 78A class).

These gene families were frequently found located on the chromosomes near both OFP family members and areas on the chromosomes associated with fruit shape.

The authors found two genomic regions in melon that had genes affecting fruit weight.

Recent work on the avocado genome has identified some similar regions when compared with tomato (the only genome I bothered to look at – there are others, such as grape, and Arabidopsis in the spreadsheets).

In the following list the first number is the reference number in the first column of the spreadsheet, followed by the function or type of protein, and whether the match between avocado and tomato is partial or complete. There are separate spreadsheets for ‘Mexican‘ and the ‘nodal hybrid’ cultivar Hass.

‘Mexican’ avocado nodal group :

  • Cell number regulator
    • 492 CNR 2 (complete)
    • 14532 CNR 2 (complete)
    • 3683 CNR 2 (complete)
    • 19056 CNR 6-like (complete)
    • 14532 CNR 8 (complete)
    • 3787 CNR 8 (partial)
  • SUN
    • 9572 SUN domain-containing protein 3 – like (complete)
    • 14052 SUN domain-containing protein 3 – like (complete)
    • 17388 SUN domain-containing protein 3 – like (complete)
  • Cytochrome P450 78A class
    • 18625 Cyto P450 78A 7 (complete)
    • 19185 Cyto P450 78A 7 (partial)

‘Hass’ avocado nodal group hybrid :

  • Cell number regulator
    • 8056 CNR 2 (complete)
    • 1262 CNR 6-like (complete)
    • 15483 CNR 6-like (complete)
    • 18988 CNR 6-like (complete)
    • 4911 CNR 8 (complete)
    • 11178 CNR 8 (complete)
  • SUN
    • 2543 SUN domain-containing protein 3 – like (complete)
    • 11026 SUN domain-containing protein 3 – like (complete)
    • 17897 SUN domain-containing protein 3 – like (complete)
    • 20992 SUN domain-containing protein 3 – like (complete)
  • Cytochrome P450 78A class
    • 728 Cyto P450 78A 7 (complete)
    • 15548 Cyto P450 78A 7 (complete)
  • Ovate Family Proteins
    • 23546 transcription repressor OFP 6

Fruit heavier than the current dominant ‘Hass’ cultivar are probably undesirable (heavier fruit may be suitable for the restaurant trade, but not so much for today’s small family size).

In summary, very large avocado fruit may result from over-expression of homologs of some of these genes. Other genes may also be involved in the expression of such genes.

Further reading:
Rendón-Anaya,M et al. 2019. The avocado genome informs deep angiosperm phylogeny, highlights introgressive hybridization, and reveals pathogen-influenced gene space adaptation.
PNAS August 20, 2019 116 (34) 17081-17089; first published August 6, 2019 https://doi.org/10.1073/pnas.1822129116

Xu, C., Liberatore, K., MacAlister, C. et al. A cascade of arabinosyltransferases controls shoot meristem size in tomato. Nat Genet47, 784–792 (2015) doi:10.1038/ng.3309