Data I’m never going to publish – factors affecting sycamore flowering and fruiting patterns

As a teenager I used to have a favourite thinking place, underneath a large beech tree half-way down the school drive.  I used to watch the activities of my school mates, while contemplatively chewing beech nuts (my school friends found this mildly disgusting).

Some years beech nuts were much easier to find than others; although I didn’t realise it at the time, this was my introduction to the phenomenon of masting.  At this point I had better fill you in on the basics of tree reproduction. Like most plants, trees reproduce by producing flowers that are pollinated, depending on the species, by vertebrates, insects or the wind. The fertilised flowers then produce seeds that are housed in what we term fruit or cones, and which in many cases aid their dispersal. Reproduction is energetically a costly process, reserves channelled to reproduction cannot be use for growth and defence.  Trees have evolved three different approaches to this problem. Some trees produce a moderate number of seeds in most years, others have an Irregular fruiting pattern and some, such as beech and oak, have strongly periodic fruiting patterns, “mast” years.  Interestingly (my wife hates me starting sentences off like this), trees that mast are wind pollinated.

Beech (Fagus sylvatica) mast production over a sixteen year period in England. Data from Hilton & Packham (1997

You might wonder why, if reproduction is costly, that some trees are ‘willing’ to expend so much energy in one go.  There are two schools of thought regarding this. One, which I find fairly convincing, is the “predator satiation” hypothesis (Janzen, 1971).  This basically says that the trees, by having on and off years, starve their specialist seed predators in the off years, thus reducing predator pressure by killing lots of them off. In the mast years, there are enough seeds to feed the surviving predators and produce another crop of trees.  A more recent, and less exciting suggestion (to me anyway), is that if the trees have a mass synchronised flowering effort, i.e. a mast year, then the chances of being pollinated are greatly increased (Moreira et al., 2014).

People tend to associate masting with trees that produce heavy fruit, acorns, hazel nuts and beech nuts for example, and I was no exception, so it wasn’t until a couple of years (1995) after I started my mega-sycamore study at Silwood Park that I had a bit of a revelation. I realised that not all of the trees flowered and that there seemed to be a lot fewer seeds that year than I remembered there being the year before. Sycamore seeds come equipped with two little wings (they are wing dispersed) and occur in little bunches (infructescences) so are quite noticeable.

Winged sycamore seed and ‘bunch’ of sycamore fruit

My sycamore study was one of my many side projects set up to satisfy my’ satiable curiosity’ and I had, at the time thought that I had made sure I was measuring everything that could possibly interact with the aphids feeding on the trees. I had, however, somehow overlooked sycamore flower production 🙂 I had taken into account that in some years the sycamore aphid can be present in huge numbers and and I was well aware from the work of my PhD supervisor

Sycamore aphids emerging in spring – some years you can see even more on the newly flushing buds

Tony Dixon, that the aphids can cause substantial losses to tree growth (Dixon, 1971), so had included tree girth and height measurements into my massive data collection list. Strangely, however, despite knowing from my work with

The effect of the sycamore aphid, Drepanosiphum platanoidis, on leaf area of two sycamore, Acer pseudoplatanus, trees over an eight year period (Dixon 1971).

the bird cherry-oat aphid Rhopalosiphum padi, that even quite low numbers of aphids could have substantial negative effects on cherry production (Leather, 1988), I had totally overlooked sycamore flowering and seed production. I am just thankful, that I only missed three years of flowering data 🙂

The effects of bird cherry aphid infestation on reproductive success of the bird cherry, Prunus padus (Leather, 1988)

Unlike the rest of my sycamore data set, the flowering data collection was actually set up to test a hypothesis; i.e. that aphid numbers affected flowering and seed set. Sycamore is in some ways similar to the well-known masting species such as oak and beech in that it is (jargon coming up) heterodichogamous. All flowers are functionally unisexual and appear sequentially on a single inflorescence. The inflorescences can however be either protandrous, i.e. male anthesis takes place before the stigmas become receptive, or protogynous where the reverse sequence takes place. Where it differs from the typical masting species is that is produces wind dispersed seeds and is wind and insect pollinated; oak, beech and hazel are entirely wind pollinated.  Pierre Binggelli, then based at the Unibersity of Ulster, hypothesised that protandrous trees may suffer less herbivore damage than protogynous trees (Binggeli, 1992). He suggested that protogynous trees, having less energy available to invest in defensive chemistry, are more attractive to insect herbivores, particularly chewers. On the other hand, sycamore trees that have been subject to previous insect infestation have fewer resources available to produce female flowers, become protandrous and avoid infestation by herbivores the following year. Presumably the next year, having escaped insect attack by being protandrous they should become protogynous again. So, if I wanted to test this hypothesis, I needed to learn how to sex sycamore flowers. Despite a handy guide that I came across (Binggeli, 1990), ) I found it almost impossible, to do, so

A. Protogynous inflorescence (female II flowers of Mode G are male II in Mode B). B. Protogynous infructescence, Mode B. C. Protogynous infructescence, Mode G. D. Protandrous inflorescence.
E. Protandrous infructescence. F. Vegetative shoot, G. Flowering shoot (Mode E).
H. Fruiting shoot (Flowering Modes B,C,D & G). (From Binggeli, 1990)

contacted Pierre, who very kindly agreed to check some of my ‘guesses’ for me.  Despite this help, I still found it very difficult so opted (very unwisely as it turned out) to collect fruit samples from each tree, put them in paper bags, and bring them back to the lab for sexing at a later date.  As you have probably guessed, I ended up with lots of paper bags which I then, not very cleverly, stored in plastic bin bags.  This went on for several years as I kept putting off the day when I would have to sit down and sex several thousand bunches of sycamore fruit. Then came the happy disastrous day when I came back from holiday to find out that the cleaners had disposed of my bin bags. To tell the truth I was not that upset as it gave me an excuse to stop collecting the fruit samples and reduced my feelings of guilt about having huge piles of unsexed sycamore fruit bunches cluttering up the lab 🙂 I did, of course, carry on counting the number of flowers on the trees, which was much easier data to collect and analyse.

I reluctantly ended my study in 2012 when I left Silwood Park for pastures new, but despite this I still haven’t analysed all my sycamore data, although I was very happy a couple of years ago when a PhD student from the University of Sheffield (Vicki Senior) volunteered to analyse some of my sycamore aphid data which was published last year (Senior et al., 2020). The winter moth data and orange ladybird data are also being analysed by a couple of my former students and hopefully will also be published by next year.

So what does the sycamore fruiting data show? Well, first, despite sycamore being reproductively somewhat atypical of other masting trees species, I would contend that my 17-year data set of sycamore fruit production looks remarkably similar to the Hilton and Packham beech masting data set. I am thus confident in stating that sycamore is a masting species.

Mean sycamore fruit production at Silwood Park, averaged from 52 trees 1996-2012,

Am I able to link sycamore seed production with aphid abundance, is the fruiting pattern a result of herbivory?  I can’t test Pierre Binggeli’s hypothesis about sex changing trees, because I lost the data, but I can try and see if aphid infestation affects fruit production. The two most common aphid species on the Silwood Park sycamore trees are the sycamore aphid Drepanosphum platanoidis and the maple aphid, Periphyllus acericola.  

Mean sycamore aphid and mean maple aphid loads (average annual counts per 40 leaves from all trees) 1996-2012.

They can both occur in high numbers, but in general, the average numbers of P. acericola are much higher than D. platanoidis. The reason why P. acericola has much higher numbers is a result of its over-summering strategy.

Over-summering morphs of the sycamore and maple aphid. Images from

While the sycamore aphid spends the summer aestivating (basically a summer version of hibernation in that metabolism is reduced and reproduction ceases), the maple aphid produces a huge number of nymphs, known as dimorphs, which over-summer in dense, immobile aestivating colonies.  The sycamore aphid can escape predators by flying off the leaves if disturbed, the maple aphid dimorphs on the other hand, rely on their huge numbers to ensure survival of some of them over the summer to resume development and reproduce as autumn approaches, a form of predator satiation. They thus suffer a huge reduction in numbers compared with the sycamore aphid. (I must publish that one day). This makes drawing conclusions about the of herbivory (aphid feeding) on the trees a bit difficult.

Mean combined aphid load, showing how the number of dimorphs of the maple aphid skew the perceived aphid load.

Given that Tony Dixon showed that sycamore aphids cause a significant reduction in tree growth (Dixon, 1971), I

Relationship between mean combined aphid load (sycamore and maple aphid) and mean sycamore fruit production.

expected to see a negative relationship between aphid numbers and fruit production. What I did find was that there was a significant positive relationship between sycamore aphid numbers and fruit production, i.e. the more sycamore aphids, the more fruit produced, whereas with the maple aphid it was the other way round, more maple aphids, fewer fruit. If I combined the aphid loads, then the relationship becomes significantly positive, the more aphids you get the


Relationship between mean combined aphid load and the number of sycamore fruit produced the following year.

significantly negative relationship between aphid numbers and sycamore fruit production, but as I pointed out earlier this is driven by the preponderance of maple aphid dimorphs in the summer. You might also argue, that rather than looking at aphid numbers and sycamore fruit production in the same year, I should be comparing aphid numbers with fruit production the following year, i.e. a lag effect. I did indeed think of this, and found that there was, for both aphid species, no significant relationship between aphid numbers the previous year and fruit produced the following year. In fact, if I was an undergraduate student I would point out that there was a positive trend between aphid numbers and fruit production 🙂  If I do the same analysis using the combined aphid load, then the relationship becomes significantly positive, the more aphids you get the more sycamore fruit you get the following year which although counter-intuitive fits with the idea that stressed trees tend to produce more offspring (seeds) (Burt & Bell, 1991) and given that we know from Tony Dixon that the sycamore aphid causes a significant reduction in growth (Dixon, 1971) which is an indication of plant stress (Grime, 1979) makes perfect sense. 

Relationship between mean combined aphid load and the number of sycamore fruit produced the following year.

Instead of mean aphid load, perhaps we ought to be thinking about aphid occurrence at crucial times of the year for the tree, for example budburst. If you go back to the top of the page and look at the photograph of the infested buds you can see that there can be a huge number of aphids present at this time of year just when the trees are starting to wake up and put on new growth. Any interference to the uptake of nutrients at this phase of their life cycle could be detrimental to fruit production.  One way to measure this is by looking at the date the first aphids appear on the buds in the expectation that the earlier the aphids start to feed, the bigger their impact on the trees. Sure enough, the earlier the aphids start feeding, the lower the number of fruit produced.

Significant negative relationship between date of first appearance of aphids on the buds and number of fruit produced in spring.

Although all the relationships I have discussed and shown are significant, the amount of variation is explained is pretty low (over 20% but less than 30%). The relationship that explains most of the variation in any one year is the size of the tree, the bigger the tree the more fruit it produces.

Relationship between size of sycamore tree and number of fruit produced (2009).

As a rule of thumb, the bigger a tree the older it is and older trees have more resources and can afford to produce more offspring than younger smaller trees.

In conclusion, what I can say with confidence is that there is significant variability in sycamore fruit production between years and this is, in my opinion, evidence of masting events, and may be linked to the size and timing of aphid load but is moderated by the size and age of the trees. If you have any other suggestions please feel free to add them in the comments.

If anyone is interested in delving into the data in more depth I will be very happy to share the raw data and also the local weather data for the site.


Binggeli P. (1990) Detection of protandry and protogyny in sycamore (Acer pseudoplatanus L.) from infructescences. Watsonia,18, 17-20.

Binggeli P. (1992) Patterns of invasion of sycamore (Acer pseudoplatanus L.) in relation to species and ecosystem attributes. D.Phil. Thesis, The University of Ulster.

Burt, A. & Bell, G. (1991) Seed production is associated with a transient escape from parasite damage in American beech.  Oikos, 61,145–148.

Dixon, A.F.G. (1971) The role of aphids in wood formation. 1. The effect of the sycamore aphid, Drepanosiphum platanoides (Schr.) (Aphididae) on the growth of sycamore. Journal of Applied Ecology, 8, 165-179.

Hilton, G.M. & Packham, J.R. (1997) A sixteen-year record of regional and temporal variation in the fruiting of beech (Fagus sylvatica L.) in England (1980-1995). Forestry, 70, 7-16.

Hilton, G.M. & Packam, J.R. (2003) Variation in the masting of common beech (Fagus sylvatica L.) in northern Europe over two centuries (1800-2001). Forestry, 76, 319-328.

Janzen, D. H. (1971) Seed predation by animals. Annual Review of Ecology and Systematics, 2,465–492.

Leather, S.R. (1988) Consumers and plant fitness: coevolution or competition? Oikos, 53, 285-288.

Leather, S.R. (2000) Herbivory, phenology, morphology and the expression of sex in trees: who is in the driver’s seat? Oikos, 90, 194-196.

Moreira, X., L. Abdala-Roberts, Y. B. Linhart, and K. A. Mooney. (2014_. Masting promotes individual- and population-level reproduction by increasing pollination efficiency.Ecology, 95, 801–807.

Grime ., J.P. (1979) Primary strategies in plants, Transactions of the Botanical Society of Edinburgh, 43,2, 151-160.

Senior, V.L., Evans, L.C., Leather, S.R., Oliver, T.H. & Evans, K.L. (2020) Phenological responses in a sycamore-aphid-parasitoid system and consequences for aphid population dynamics; A 20 year case study. Global Change Biology, 26, 2814-2828.


Filed under Aphidology, EntoNotes

4 responses to “Data I’m never going to publish – factors affecting sycamore flowering and fruiting patterns

  1. Humphrey Crick

    That’s an interesting set of findings from a neat side project! The question that came to me is how climate warming might affect the interactions between aphids and sycamores. If greater aphid numbers result in lower fruit production; and earlier aphids lead to reduced fruit production; then with warming, which presumably enhances aphid over-winter survival + earlier aphid emergence, the poor sycamore tree will be hit by a double whammy. However, I guess that sycamores also bud-burst earlier, so it would be important to know if there is an effect of relative earliness (both aphids and tree advance with warming but at different rates) or absolute earliness (which ignores any differences between the advancement of tree and aphid – which I think is what you have shown in your penultimate figure). Have you got any measurements of timing of aphid appearance on the buds, so you can calculate some sort of relative timing?

    Just a thought!


    Liked by 1 person

  2. Indeed, I have dates for everything 🙂 There is a trend for earlier bud burst in the sycamores


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.