Mango and Lychee - manipulating the flowering by Don Batten


I worked for 17 years in research at the Alstonville Research Station, Dept of Agriculture NSW and my responsibility was basically everything other than bananas, avocadoes and macadamias, which gave me a pretty fair spectrum for doing all sorts of things and I was responsible for bringing in and testing new fruit crops that might possibly be commercial.

I’ll be talking about flower initiation in both lychees and mangoes because they turned out to be very similar in the way they form flowers, but it also applies to longans and virtually any terminal flowering sub-tropical / tropical type fruit.

The myth of winter dormancy

There’s a myth that for a lot of these species you need a winter dormancy to get them to flower. It’s bollocks.  The myth has been around since the 1920’s when a fellow by the name of George Groff, an American missionary in China, got it from the Chinese. Researchers around the world took on this myth that you needed this dormancy to get regular flowering. Consequently, all sorts of experiments were done—I could write a book about all the experiments I did to get winter dormancy to get them to flower better and every experiment failed.  In fact, the more dormant they were, the less they flowered but DPI in Qld and NSW advised growers not to water or fertilise because that was going to make them flower less.  I did experiments to get that to work, but nothing worked.  I sat under a tree one day, despondent, thinking I’d give up, when I made a significant observation—probably not significant to you—but it was a bit like the apple falling from the tree for Newton (if that really happened).  This was an inflorescence with leaves on the bottom part: a shoot with leaves near the bottom and flowers towards the tip.  I thought that if flower initiation occurs in dormant buds, then that can’t happen. You can’t get leaves produced and then switch to flowers unless the bud is actually growing while this happens – it cannot happen while it’s dormant. You see this in a number of lychee varieties—Bengal, Kwai Mai Pink (‘Bos 3’), Salathiel.

Also, in SE Qld and northern NSW in any given year there will be some trees that flower and bear a decent crop. In other words, there is always enough chilling to get them to flower.  There is never a lack of winter chilling; that is just not an explanation for failure to flower in this area. So the question was why does one tree flower and not the next?

Testing the idea that flower initiation occurs in growing buds

So I had the idea that flowering induction occurs in young growing buds, not dormant buds.  That was pretty radical at the time, but the question was how do you prove it?  Well, the experiment was this:  Begin buds growing at a high temperature, e.g. in a glasshouse with a temperature of 25ºC constant, (they won’t flower at this temperature; it’s non-inductive), and then transfer them into the inductive temperatures like a cold glasshouse eg 12/18º (night/day) or outside in winter (late July, early August) and see what happens. And what happens if buds are different sizes?  If my idea was correct, small buds should produce just flowers, but longer ones at transfer, which will already have produced some leaf initials, will produce leaves on the bottom and flowers on the tip.  Bigger ones again should only produce leaves.

The experiments:  I cut the tips off to get back to auxiliary dormant buds, forced them to start growing at the high temperature, and transferred them to the cold conditions at a whole range of different size buds (having measured the sizes of the buds).  At ‘zero’ size, we just pruned the plants and put them straight into the cold conditions so they didn’t have a chance to grow. With one experiment with Salathiel, half of them grew out and produced inflorescences but the other half produced leaves only.  If we look at buds that had grown a bit, e.g. 1–1.8mm long, which is not very long when transferred into the cold, we got about 70% that produced pure inflorescences and 10% that produced mixed leaves with flowers at the top—so over 80% produced flowers—and about 15% produced only leaves (vegetative).  Then if you go bigger, you start to get mainly mixed ones—in other words they have produced some leaves and changed the flowers and an increase in the number of vegetative ones. Then as you get bigger again, you get even more that are just leaves and fewer with flowers and finally if they’re bigger you just get leaves.  So you can see if the buds are growing, you actually get better flowering than if they’re dormant but of course what happens here is that the ones that are actually dormant, they actually start growing in the cool room and then they flower but if they actually don’t start growing they won’t flower so the idea of winter dormancy is really crazy when you think about it. Tai So, which is an early variety and tends to be cold sensitive, in this experiment, didn’t work well and the reason I think was because they were pot bound but when they were transferred at 18º constant day temperature and 12º constant night temperature, a lot of the buds that had actually started to grow, stopped and then aborted.

With Tai So we got no flowers whatsoever when they are dormant when they were transferred to the cold room, but at least we got some flowers on the buds that were growing when transferred.  With Wai Chee, the ones that were dormant, about 60% gave flowers but if they were growing, all of the buds flowered.

We also did a set of experiments where we grew them through a cycle of growth at between 20–27ºC before we did the experiment so there was no way in the world that they could have been pre-induced to flower (just to cover this possibility).  The effect was even clearer, if anything.  With fresh, vigorous plants, it worked really well.

We found the same thing in mangoes—flower initiation occurred on young growing buds. 

Where did the ‘winter dormancy’ idea come from?

If you get a flush in April–May (autumn) they often don’t flower so that’s where the myth came that they have to be dormant in winter—people rightly observed that if they did this in April-May before winter (so the flush would be still hardening off during winter), they don’t flower.  But they don’t flower because they can’t grow—because they are dormant!

Lychees, mangoes, avocadoes, longans and all sorts of terminal flowering sub-tropical and tropical tree crops grow by flushing and when they flush, they actually have to have a period when they harden off the flush before they start to grow again.  So you cannot make them grow from the tip until the terminal bud hardens off or matures.  When the bud finally matures it will grow again, but there’s nothing you can do to make it grow until it is mature enough.  You can prune it off, and then you’ll get growth from lower auxiliary already-mature buds.  In summer in SE Queensland, it can be about 50 days (7 weeks) from the beginning of a flush to the start of the next one, but in winter this stretches out to more than four months so if a flush starts in May, it can be September before the terminal buds can grow and so they don’t flower because it is too warm then for floral induction.

How long does the cold spell need to be?

When we transferred Irwin plants with buds just starting to grow (in a hot house) outside in winter (average daily min 10°C and max 20), after only 30 days you could see flowers developing—the induction obviously occurs quite quickly.  Lychees took a little bit longer, but it happens very quickly. The idea that you need months of cold is rubbish. We have strong evidence that less than 3 days of low temperatures is fine to get them to switch to flower.

The buds in Irwin can range from 0.7 mm up to 10 mm long and you’ll still get flowering, so there’s a big window of opportunity to get the cold snap while the buds are growing.
If you look at the weather data for Cairns (13°S), for example, they usually get 1 or 2 cold snaps in a year (temperature drops down to 12 or 13 degrees at night) for a couple of days.  If a lychee happens to have just started to grow when they get that cold snap, they get a good crop.  If it’s not cold during that window of opportunity when the buds are small, they don’t flower.  That’s why cropping is somewhat erratic at Cairns. 

In that environment, growers could tip prune trees several weeks (timing would have to be worked out by experimenting) before a cool snap is expected to maximize the likelihood of a crop (although just when it comes is a bit unpredictable).

George: What’s the minimum and maximum temperature for floral initiation?
Don:    Nobody knows really precisely, but a daily temperature range of about 13–20 degrees will produce flowering.  Daily maximum temperatures above 20 inhibit flower formation. It could also be a drop in temperature that is more efficient than absolute temperature so if it’s fairly warm and it’s just starting to grow, then the temperature drops, it could be the drop in temperature that is more important than the absolute temperature. 

The optimum size for the mango buds was 1.5–2mm long for getting pure inflorescences. Irwin is a Florida variety and is known for fairly reliable productivity—not so good to eat but reliable productivity but what is weird is that when Irwin’s buds were bigger (about 10 mm long), it actually produced a flush of vegetative growth and just kept on growing and produced an inflorescence on the tip.  So it didn’t actually have a dormancy period when they stopped growing and hardened off. The cold actually stimulated it and it kept on growing and it produced inflorescences on the top of the vegetative shoot. Mangoes are actually easier to deal with than lychees; I have not seen lychees do that but not all mangoes will do it either.  I think it’s probably a characteristic of some of the more productive Florida varieties and that’s one of the reasons they are more productive.

How do we use this knowledge?

Lychee and mango grow in flushes. Following a recent flush, bud growth won’t occur because of maturation of buds so if buds can’t grow during inductive conditions (e.g., winter), flower initiation won’t occur so we need to make them grow during winter—it’s all to do with timing. In SE Qld a flush beginning in May (autumn) will prevent flowering because that flush will take months to mature and they won’t be able to grow again until spring at which time they’ve gone through flower induction period so you won’t get flowering.  This applies for lychee and most mango varieties. If the post harvest flush starts (2mm long buds) in early March (first week of autumn) that will in most circumstances result in having nice mature buds coming into winter without starting to flush in May (late autumn) and it will be ready to begin to grow in June/July (winter) giving good flower initiation giving you the foundation for a good crop.

In Cairns, Nth Qld they get two flushes between harvest and flowering but between there and here in Brisbane there can be a no man’s zone where you have to delay the post harvest flush so you don’t get a second flush starting before the cool period—this can be tricky.

In this part of the world, we need to get the flush away pretty well straight after harvest and that means that your trees should be fertilised / watered and do everything to them to make them grow at that time.  Once the post harvest crop is underway; that is, it’s growing away nicely, don’t fertilise/don’t water—just let them settle down and hopefully they won’t grow again until winter arrives. So in June (winter) you should fertilise them lightly and water the trees to stimulate them to grow. This is the opposite of the advice that has been handed out in the past—it’s the complete opposite, but it works!

Timing for longans will be different because they harvest later then lychee.&nbsp%

This article was compiled by Sheryl Backhouse

Authored by: 
Don Batten
Sourced from: 
STFC Newsletter April-May 2007
Date sourced: