Home => fruit
Notes on Avocado Pollination
by Laurie Meadows

Avocado flower floral partsMale phase avocado flower. The female parts of the flower are obscured by upright inner whorl of stamens. In this photograph, the white bulbous anthers at the top of the stamen filaments have not yet shed pollen. Only the stigmatic surface - at the very top of the style - can be seen. Notice that one stigma is white and receptive, and the other brown and unreceptive.

The style, and ovary at the base of the style, are obscured by the upright innermost stamens.

The yellow staminodes function as nectar secreting organs, and there is also a true nectary at the base of each of the 3 inner stamens.

Both the petals and the 'tepals' in the avocado are almost identical in size, shape and color. The tepal is the equivalent of the sepal in a flower. Sepals are usually small relative to the petals, and a different shape and color (green, usually). When the sepal is nearly identical in form and color to the petal - as it is in avocado - it is termed a 'tepal'.

Modified from Wikipedia image

avocado flower, valves open, pollen
              sheddingCloseup of avocado cultivar 'Sharwil' anthers shedding pollen. The 'tear tabs' (valves) on the side of the anthers have peeled upward, releasing the pollen. The pollen is moist at first and remains in a clump. It soon dries enough to disperse, but some pollen grains remain loosely attached to the underside of  the valves (just visible in the photo).

There are 4  valves on each anther, but not all valves necessarily open, or open fully.             



Hass is in a class of its own. It sets fruit when other cultivars either don't set (edranol), or set erratically (Pinkerton).

What makes Hass so reliable?

The basics of avocado sex
Each flowering panicle has a large number of flower buds. Every day new flowers open, and two-day-old flowers are shed. 

On its first ever opening, the flower opens as a female; all 9 male stamens do not shed pollen, and lay down flat and tight against the green tepals, leaving the the female ovary, style and stigma totally exposed. The stigma is white and receptive to pollen. After some hours, the flower closes.

The flower remains closed until the following day, when it opens again in the pollen shedding 'male' phase, although the female parts might still be receptive. If the stigmatic surface has turned brown and somewhat shriveled, it is considered unreceptive to pollen. After some time, the flower closes for the second, and last, time. It it hasn't successfully been fertilized, it withers and falls off. If it has been fertilized, it begins to form a fruit.

The flowers synchronize their sex phases - the tree slow motion switches from all flowers on the tree female flowers only, to all flowers on the tree closed, to male flowers only, to all flowers closed and so on. But the neat time-of-day synchronization of floral gender only happens in consistently warm temperatures.

In New Zealand, there is often overlap between the genders. Some flowers are late to finish their cycle, and some flowers of the opposite gender-phase open before the last gender flowers of the previous gender-phase have all closed. In the main flowering period (when it is warmer), the Hass female flowers open around about midday. If it has been cool for some days, female flowers might not open until the late afternoon. (Recent research in New Zealand has shown that periods below 7oC push Hass female flower opening later and later in the afternoon, and then into the night.) Conversely, if it has been very warm for a few days, female flowers might even open in the morning and close at midday (as it does in hotter countries). The flower's female phase ends when the flower closes 1 or 2 hours after opening. The same flower re-opens the next day, but as a male, shedding pollen.

The Hass flowers male phase (pollen shedding phase) typically starts when the flower opens in the afternoon or evening. The flowers stay open all night, and are finally exhausted by late the next morning. Sometimes the male phase flowers open early, about mid morning.

Either sex-phase of the flower can vary greatly in opening and closing time on any given day.

Indeed, I have noticed that early in the season, sometimes morning sunshine on the eastern side of some trees seems to promote flower opening on that side, whereas the flowers on the southern and/or western side (shaded from the morning sun) remain closed. Interestingly, researchers in New Zealand have noticed the opposite phenomenon, albeit with much larger trees. Either way, variation in the timing of flower opening
'within-tree' might possibly be useful in allowing overlap of the sex phases.

The variation in opening and closing times of the two flower phases is not particularly important. What is (somewhat) important is that there is often a little bit of overlap between the end of one flower gender phase, and the start of the other gender phase.

That said, in New Zealand's humid conditions it may well be that the
male phase flower may still have viable female parts when it opens, and may be able to set by itself, given a little assistance from pollinating insects (or even gravity). There is mounting evidence that the male phase flower is the one that sets the bulk of the crop (contrary to what was thought).

But for physiological reasons, the avocado flowers simply won't set until it is warm enough for long enough. And once it is warm enough, set of the main summer and autumn cultivars (Hass and Reed) is virtually assured.

The bottom line
Whatever the intricacies of flower pattern, in New Zealand at least, Hass sets well in the backyard without any need for pollen from a different cultivar - as does does the late-setting Reed variety.
As long as only these varieties are planted, they will set fruit. (The 'up and coming' Hass-like green-skinned 'Gwen' variety also has a reputation overseas for similar easy reliability, but in New Zealand, not so much.)

But Hass and Reed varieties are more or less summer and autumn varieties (respectively). Hass fruits can be (selectively) harvested early, from around September, and Hass continues to fruit until January, at least. Hass trees in cooler parts of a garden might flower later, and hold their fruit on the tree as late as May. Reed is ripe late summer and autumn. So the 'gap' in fruit production is June, July, and August. (The winter varieties 'Bacon' and 'Zutano' ripen in about July to August, but their flavor is very poor, so we can ignore them).

The only reason to even think about pollination is if you are trying to establish good quality varieties that might fruit in winter. If you are happy with summer and autumn fruit, then read no further!

Winter fruiting good-quality Avocados

The only good-quality avocado varieties available that are 'supposed' to fruit in winter are Fuerte, Pinkerton, and, to an extent, Hashimoto.

Fuerte, like Pinkerton, flowers over a very long time, from early winter to late spring. In a warm spring its early flowers may set fruit, which are good quality by about July.

In a 'normal' year (north of Auckland, at least) its flowers won't set anything until about October, the same as Hass. Fuerte will produce female flowers at lower temperatures than most other 'B' types, but, ironically, needs warm days (best at 25oC or over) to set seed and hold the fruit.

Fuerte fruit set in a normal year aren't good quality until about August. Even then, once picked, they can take 9 to 14 days to ripen on the kitchen bench, so the fruit is not ready to eat until late August/early September. Normal season fruit size-picked in July usually taste like soap.

Fuerte fruit set is erratic from year to year, depending on winter/early spring temperatures, and so season of acceptable ripeness can also shift - earlier or later than 'normal'.

Pinkerton has about the same flowering period as Fuerte (maybe slightly later). In San Diego USA it is called 'winter Hass', because its earliest crop is in mid winter. My experience is that flowers will not set until mid October - once again, the same as Hass. The fruit from this set matures about September, but can be picked as early as late July. The fruit are oily and very good, but don't have the depth of flavor that Fuerte has. They are better than early picked Hass. I assume that in unusually warm years it will set earlier, and be harvestable in August, like Fuerte. My experience is that this variety sets very poorly, possibly unless it has a suitable pollinator (when it sets very well, and is very reliable year to year).

Hashimoto matures fruit from roughly april to july (but can also be picked much earlier and still be good). I suspect that it is like Reed, a late spring flowering tree whose fruit take 18 months to mature. I am guessing the fruit hold on the tree into the early winter months. The trees here were only planted in spring 2012, so it will be a few years before I find out. (Late set Reed fruit will also hold into july, but late flowers in any given year can't be guaranteed. ) 

Although Hashimoto is a 'B' type, it fruits well. Presumably it sets well on its own pollen. If Hashimoto holds on the tree into winter, then it may turn out to be the best choice for a good-quality winter fruiting avocado. The ability to pick fruit in late autumn would be a bonus. Hashimoto fruit have very good oil content, albeit the flesh can seem a little dry, not as 'smooth' as Hass. It also has obvious brown fibers in the flesh, but these are not noticeable when eating the fruit. It has a thick skin, like many of the large round Guatemalan avocados, making a little hard to tell when the fruit has ripened. In the hot conditions of California, it has a reputation for going 'bacony' and 'off' if left on the tree too long. I doubt that will be a problem in our cooler climate.

Barriers to fruit set

The most important barrier is the flowers dependence on warm temperatures to set fruit.

The avocado is a subtropical fruit. Its natural habitat in the subtropics does extend up the side of mountains, into cooler climate zones. But there are limits. Its foliage is frost sensitive, and trees can be killed by severe freezes. If the tree allows flowers to set at cold times of year, then it has more chance of a night, or nights, cold enough to freeze the developing fruit and kill it. The investment in initial fruit growth would then be lost. So the tree behaves as if it has a 'built-in thermostat' that stops flowers from setting when the general air temperature is relatively low. The tree may flower in low temperatures, but it drops them all, even if the flower is pollinated.

Flowers of a number of 'B' types (e.g. Sharwil) will not overlap their own male and female flowers well unless temperatures are at least 25oC by day and at least 20oC at night (Whiley 1987). This is irrelevant for New Zealand because 'A types' such as Hass are available to act as pollinators. More to the point, while some 'B' types produce prodigious numbers of male flowers (for example Sharwil and Fuerte), they may produce very few female flowers anyway. (Some 'B' type varieties such as Edranol seem to have an additional temperature mechanism - the female phase simply doesn't open if the temperatures are not high enough for long enough. And in this part of New Zealand, they evidently are not!)

of 'A' type avocados are most likely to set from self pollination (i.e. there is at least some overlap between male and female phasing) when day temperatures are between 23C and 27C, and when night temperatures are 10C or higher. Natural daily variation above or below these temperatures can also help the overlap.

Recent research shows a combination of high relative humidity and temperatures over 20oC are needed for maximum pollination adhesion to the stigmatic surface and maximum pollen germination.

(Conversely, when temperatures exceed 25oC, pollen germination decreases in the male stage flower, probably due to the female stigma drying out - but if relative humidity is high, the stigmatic surface remains receptive, and pollen germination remains high, even at these higher temperatures.)

Pollen takes between between 24 and 48 hours to grow down the female flowers stigma, reach the ovary, and fertilise the ovule. Cool conditions slow the growth of the pollen tube. If it is too cold, the pollen tube stops its journey altogether. But there's a deadline. If the ovule isn't fertilised within 48 hours, ethylene is produced at the base of the ovary, and this causes the ovule to abort and the flower to drop off.

In addition, there is some evidence that the ovule itself does not mature properly in cool weather. In this case, even if the pollen tube reached the ovule in time, no embryo would form.

Young fruitlets are retained only when air temperatures are warm enough for long enough. Research suggests a minimum night temperature over 15.5oC, and a daily maximum of at least 21oC. These minimum temperatures probably need to continue for at least 2 consecutive days. The 15.5oC night temperature is almost certainly incorrect under our conditions, as such high night-time temperatures are not often reached in spring (when fruit set) in New Zealand.

In summary, temperature seems to be the primary determinant of when avocado panicles will start to set and hold fruit under our climatic conditions. The avocados here either don't set at all, or don't hold the 'apparently set' fruitlets until extended periods of warmer conditions arrive - usually in very late october and early to mid november.

Every other factor - pollinating insects, overlapping male and female floral stages, cross pollenizing varieties - appears to be secondary to prolonged periods of warmer weather.

Male and female flower parts receptivity
Overlap between the separate male and female flowering phases
Here in Helensville, at least, different Pinkerton trees can be 'out of synch' with each other by an hour or so, resulting in a slight overlap of male flower phase and female flower phase between different trees. The trees are at some distance from each other, so pollen is not likely to be carried to the female phase tree. Nevertheless, small differences in timing do exist between trees, perhaps reflecting different micro-climates, or even different rootstock effect.

The earliest flowers of one sex-phase's floral opening often overlap with the last flowers to close of the previous sex phase. Again, the result is overlap of male and female flowers, but this time within the same tree.

A small amount of overlap between phases is fairly common in our climatic conditions. Under cool conditions the time of opening of female flowers is delayed. Here in Helensville, flowering at the warmest part of year is fairly synchronous, with relatively little overlap, but if there are unusually cool nights or days, opening and closing phases shift all over the place. Indeed, the 'A' pattern can temporarily become the 'B' pattern.

It is ironic that the cool weather in the early part of the avocado flowering season causes the greatest overlap  in the female and male flowering phases (sometimes up to 4 hours), but the same low temperatures mean that fruit are most unlikely to set!

Overlap between male and female parts within the same flower
Florida research showed that after the female phase flower closes, very often the female stigma remains white and receptive the next day, when the flower re-opens as a functioning male. In fact, their data showed that anything from 25% to 85% of male phase flowers are successfully fertilised (lower in dry conditions, higher in moist, humid conditions). This effect may be peculiar to
their humid tropical climate.
The success of flowers self or cross fertilising in the male phase is perhaps supported by the fact that Israeli research showed that Hass male phase flowers have over twice the amount of sugar that female phase flowers have.

A lot of the information on avocado pollination comes from California and Israel. These are Mediterranean climates, with much less relative humidity in spring. San Diego, for example, averages about 60% to 70% relative humidity - low by our standards - even although it is in the supposedly 'humid' coastal zone (Chilean conditions are very similar - dry, with low rainfall). And both California and Israel can have 'mini heat waves' when temperatures skyrocket, even in spring. Under these conditions, it is much more likely that the female stigma and style on the male phase flower will have dried out, making male-flower pollination less likely. No wonder they focus on the importance of 'B' type varieties as a pollen source for the commercially important 'A' type Hass variety.

Recent (2001) research in California has shown that the Florida situation is operative in California. Whether bees were present or not, only about 3% of available stage 1 female flowers received 1 or more pollen grains (there was no overlap of male and female phases at the time of this study, so the pollen likely came from nearby Zutano trees.).

And again,
whether bees were present or not, around 18% of the open stage 2 male phase flowers had 1 or more pollen grains on the stigma. This pollen was from the trees own flowers (or from adjacent Hass trees). In the more humid coastal areas, over 95% of the male phase flowers had receptive white stigma during the 2001 flowering season (hotter inland areas only had around 50% receptive stigmas in the male phase). 

Although a relatively low number of the total avocado flowers open ever received any pollen at all (~ 3% in the first opening and ~ 18 % in the second opening), it means that even in California, about 80% to 85% of Hass flowers actually pollinated were from self-pollination in the male phase flowers.

For New Zealand, with its roughly 90% relative humidity in springtime, the female structures in the male phase flower are probably in excellent condition to receive pollen from the immediately adjacent anthers - as is the case in humid and semi tropical Southern Florida and in Coastal California.

Of course, planting other varieties nearby does allow cross pollination between varieties. The 2001 Californian work showed that in the absence of overlap between the two gender phases, wind borne pollen from other adjacent varieties would contribute 15% more flowers with one or more pollen grains to the stigma. That does not translate to a 15% increase in total crop (by weight) however!

For a start, South Florida researchers showed that pollen from other varieties did not set more fruit than a varieties own pollen.
Isozyme analysis of the set fruit of 3 different cultivars showed that over 85% of the harvested fruit were from self pollination. In other words, a given varieties own pollen is usually as 'potent' as that of another unrelated variety.

It had previously been shown that Hass seemed to preferentially retain fruit that had resulted from out-crossing with another variety. However, it has since been shown that the fruit set right at the end of the flowering period are the ones most likely to drop (due to competition for resources from the flush of new vegetative growth and the rapidly growing earlier-set fruit). Late set fruit are formed when the main cross pollinator varieties have finished flowering, so late set fruit are all self fertilized. This was interpreted as a general disposition of self fertilized fruit to drop, and out-crossed fruit to hang on.

Further work has demonstrated that in humid climates there is no significant difference in ultimate crop weight between Hass orchards with pollenizing varieties nearby, and orchards with no pollenizing trees at all. (However, Pinkerton, specifically, is an exception to this rule.) In contrast, in arid Mediterranean climates with dry air (Chile, large parts of inland California), 'B' varieties for cross pollination are essential.

If it is the case in New Zealand that the male flowering stage is the stage most likely to set fruit - as seems likely - then there are only two other possible barriers to within flower pollination worth noting: lack of insects to transfer pollen from anther to stigma (a distance of only millimetres); or some physical barrier to a flower pollinating itself in the male flower stage.

A further twist to the tale of avocado flower set was the discovery by Spanish researchers that the avocado flowers that hold on the tree and form fruit are disproportionately those flowers with high starch levels. But only a few of the thousands of flowers on a tree have very high starch levels. The researchers showed that whether a Hass tree is in a heavy cropping 'on' year or a low cropping 'off' year, the proportion of flowers in a raceme that go on to form and hold a fruit are the same. The 'on' year fruit are more numerous only because the sheer number of flowers is much higher in that year.

Flowers, fruit, roots and leaves compete for resources
Hass tends to have heavy crops one year and light crops the next. A heavy crop on the tree is still maturing its fruit in spring, just as the flower panicles are demanding stored resources to expand and produce their innumerable flowers. New leaf growth starts pushing through the flower panicles at the same time. And it is said that root growth more or less coincides with shoot growth. So there is a very heavy call on the trees stored resources in spring, just after a long period of unfavorable wet and cool conditions in winter.

You could argue that a tree with a heavy load of fruit to ripen 'self regulates' by devoting more stored resources to ripening its existing near-mature fruit than to producing lots of flowers for next years 'possible' crop. And it is a forest tree, so if it is to successfully compete for light, it has to extend growth every year, no matter what the fruit load. So any one years flowering is, to a degree, expendable for the long-term greater good of the tree.

Physical barriers to pollination
The male flower stage is illustrated in the photo at the top of the page. One characteristic is that the innermost group of 3 stamens stand tight and vertical against the style. In most avocado cultivars the tip of the style, bearing the receptive stigmatic surface, extends slightly higher than the enclosing inner stamens, and so is still exposed to pollen. Pinkerton has a long style with a distinctive 'kink' in it. Perhaps it is bent because it is so long. Whether this delays the transit of Guatemalan race pollen to the ovule (it is claimed to set heavily in the presence of Mexican race pollenizers) or not is speculation, but could be considered as an explanation for Pinkerton's reluctance to set fruit. Pinkerton is a seedling of the variety Rincon, and Rincon is sometimes considered self-infertile. Possibly it suffers from the same physical barrier.

The female stage flower has no problem receiving pollen because all the male stamens are 'out of the way', laying down flat, totally exposing the the style and stigma. But in the female stage, no pollen is shed from it's horizontal male parts. So if Pinkerton is physiologically self infertile, or if the male phase flowers have 'weak' pollen that takes too long to grow down the long style, then Pinkerton is almost obliged to have another variety nearby that is shedding pollen when Pinkerton is in its female stage. And as a result, Pinkerton is very reliant on insects moving to it from an adjacent pollen bearing tree.

Wind pollen transfer
South Florida research suggests that in humid conditions wind and gravity are effective agents of pollen transfer.
Indeed, in Southern Florida, shaking branches at the time of pollen shed increased fruit set and retention over unshaken branches! Researchers further showed that the moist pollen clumps dried out within 30 minutes or so, and were soon dispersed by insects, wind, or gravity. The amount of pollen in the air in the tree is not really important - it is the amount of pollen that blows the few millimetres to the female stigmatic surface of the same (male phase) flower. An avocado flower needs 20 grains of pollen on the stigma to be able to complete fertilization (even although only 1 grain ultimately makes it into the ovule to form the embryo).

Some suggest effective fertilization in the male stage is peculiar to the tropical ('West Indian') types that Florida grows. Interestingly, many tropical avocado cultivars show little or no self overlap between the sexes.

No research has been carried out in New Zealand into the effectiveness or otherwise of within-flower wind or gravity assisted 'self' pollination. So that leaves insect pollination.

Insect pollen transfer

Once again, under low relative humidity Israeli and Californian conditions, it has been shown that almost no fruit set if pollen-transferring insects are excluded from the tree. And when a beehive is included, fruit set normally. But these facts may reflect the dependence on flower stage (female stage vs male stage) overlap, where, under Mediterranean climatic conditions, only female stage flowers can set fruit, as the later male stage flowers have dried-out female parts (due to the low humidity). The male stage, under these conditions, can only act as potential pollen donors, given they 'hang on' long enough to overlap the earliest flowers of the next tranche of female flowers to open.

Under New Zealand conditions, flowers may set fruit at the second floral opening from pollen released immediately adjacent to the still-functional female stigma. This process may be enhanced, or overtaken, by insects knocking pollen onto the stigmatic surface.

In its native habitat in Guatemala and Mexico, avocados are said to be pollinated by stingless bees.

My experience here at Helensville can be summarized as follows:
1. In general, there are either relatively few or no larger insects on the flowers at any given time, except;
2. When the male phase flowers are producing nectar there will almost always be 1 to 6 larger insects on the flowers
3. There are more insects on the flowers when it is a consistently warm day, and
4. There are more insects on the flowers later in the season when the weather is warmer and the days longer
5. Some varieties clearly produce more nectar than others, and are therefore more attractive to insects than others.
6. Set seems to happen in late october, very early november, even when in some trees there are few flowers remaining on the panicles (relative to early and mid october)

Measured by pollination effectiveness (taking into account the number of pollen grains deposited on stigmas, the frequency of stigmatic contact, the number of flowers visited per minute and the number of individual of a species around at any point of time per number of open flowers at that time), research by Rader et al (2012) on pollination of a brassica field crop in the South Island indicates that the natural diverse wild pollinator insect population pollinates the crop about as effectively as managed hives of bees.

Most of the pollination was done by 8 species of insects. The 'number one' pollinator was the honey bee, second in importance was the bumblebee, third was the drone fly (Eristalis tenax),  fourth was a native bee, Leioproctus sp., fifth was the hover fly Melangyna novae-zelandiae, sixth the blossom fly (also called the March fly or St. Mark's fly) Dilophus nigrostigma, seventh was the hover fly Melanostoma fasciatum, and eighth was another native bee, Lasioglossum sordidum.

Results from an annual crop grown in a different geographic and climatic region are not necessarily comparable, but the effectiveness of the various pollinating insects is perhaps indicative (even considering the differences between brassica and avocado pollen).

The effectiveness of the honeybee and the bumblebee are not surprising - both work when avocados are flowering in late September, and October to November. Both touch the stigmatic surface of almost every female flower they visit, both carry greater than the minimum 20 grains of pollen needed to effect pollination. What is a little surprising is that the drone fly just about matches the top 2, in these attributes; it is just not as abundant. Bumblebee numbers peak in mid spring to early summer - exactly in the main flowering period.

The drone fly is attracted to yellow and green flowers - which happens to match the avocado flower color tone. Male drone fly numbers peak in summer, and female in winter (Irvin et al. 1999). Summer is a little late, and winter is a little early!

The Brassica crops fourth most effective pollinator, Leioproctus sp does not appear in Helensville until avocado flowering is all but over (in Hass, at least). The blossom fly also doesn't appear in numbers until the flowering is all but done. That leaves the hover flies.

Melanostoma fasciatum is abundant here, and while it is low on the ranking of the 'top 8', it does appear as the avocado blossom season starts. 

The Blossom fly, a Tachinid fly ('bristle fly') is said to have its peak of activity in November and February (Harrison 1990), a time of peak insect activity and peak species diversity. It has been recorded by New Zealand researchers on avocado flowers, with avocado pollen observable. The pollen load is on the underside or the body, most likely to contact the stigma, and the amount is considerable - almost as much as bees (Rader et al. 2009).


Many flowers have a resident population of thrips. They are active little insects, and I have seen them climb all over the shedding anthers, across the stigmatic
surface, down to the base of the flower, and back again. Whether they knock the required 20 grains of pollen onto the stigma, or whether they can carry a decent load of pollen on the underside of their bodies, I don't know.

Avocado flowers often have 2 or 7 ants crawling around the base of the flowers when the nectaries are producing sugar. The ants tend to remain at the base of the flower, but even so, may knock the pollen from the inner anther whorl onto the female stigma. Their bodies lack much in the way of hairs that might trap pollen. I have looked at a couple of ants close up, and they only have 3 or 4 pollen grains on them, usually on their legs. One ant had a single grain on its head. So while they might help distribute pollen within the flower, the chances of transferring pollen between flowers seems pretty small. Ants are known to secrete an antibiotic which kills pollen, so they are unhelpful in pollen transfer between flowers.

Miscellaneous beetle, weevils, and other random insects

avocado pollen eating beetle excretaFrom time to time avocado flowers are visited by an eclectic mix of insects, from beetles to weevils. I have no idea what they are. One visitor, a small metallic grey-green beetle, had a few pollen grains stuck to the underside of its body, but nothing on its rather hairy legs. Its excreta was largely pollen 'shells', which are excreted after the nutritious contents have been enzymatically extracted.

Fungus eating ladybird on Pinkerton female
                  flowerThe steel blue ladybird, Halmus chalbeus, is fairly frequently seen on avocado flowers, and I have watched it eat pollen. It, too, may accidentally knock pollen onto the receptive female stigma, but is unlikely to be hairy enough to transport pollen between flowers.

In late april 2013 I observed a fungus-eating ladybird, Illeis galbula, 'nose down' in an ultra-early Pinkerton flower. It was in the female phase, so no pollen was available. Observation with a hand-lens as it moved about the panicle suggested it was as likely eating the very fine hairs on the avocado, or perhaps detritus trapped on the hairs. Again, an unlikely pollinator.

Adult click beetles sometimes feed on nectar in late spring.

                  beetle on avocado flowerI have seen the odd flower longhorn beetle (Zorion sp.) amongst the flowers on the odd occasion, but only as a relatively rare sighting (photo left - also, note the open valves on the anthers) . There is a claim that a "high majority" of the fruit set in a particular western Bay of Plenty organic orchard is "facilitated" by these native beetles. This beetle was first observed at that orchard in October, so the timing for pollination is right. A team from Plant and Food Research catalogued the pollinating insects (other than honey bee) on 4 avocado orchards, including this orchard, and how much avocado pollen Zorion and other insects carry.

See below for the results.


Sturgis McKeever, Georgia Southern University,
                  Bugwood.orgThe drone fly, Eristalis tenax, turns up from time to time fairly early in the flowering season. This insect looks for all the world like a bee, but isn't. It has a hairy body, and can transfer as much, or more, pollen as a bee.

Drone fly.
Sturgis McKeever, Georgia Southern University, Bugwood.org

Other, smaller flies commonly visit avocado flowers in spring, especially in sunny weather. Flies typically wander all about adjacent flowers in between nectar feeding and they have hairy bodies, so they may be useful agents for pollen transfer. Blowflies are common in late September when the weather has warmed, but while they visit flowers, the visits are relatively infrequent, and most of their time is spent basking in the sun on the mulch at the base of the trees.

The only fly I have seen on the autumn flowers (so far) is the Australian leafroller Tachinid (Trigonospila brevifacies). This stripy little fly lays an egg on leafroller caterpillars, and when the maggot hatches it eats the caterpillar. The avocados have never been sprayed, so the trees do get leafroller, but not that many. Perhaps the fly is the reason why. Anyway, the fly observed fed diligently at the nectary of the flower it was on for some time.

Around 10 species of solitary bees are one of the primary pollinating insects in the humid subtropical forest that is the native habitat of the avocado.

New Zealand has several species of native solitary bee, but they are not active until early summer, by which time most avocado flowering is over (except for Reed). Around
about the 1st of December 2012 I observed both the most common native bee (Leioproctus fulvescens), and the second most common native bee (Lasioglossum sordidum) on very late Hass avocado flowers. A Reed tree, which is later flowering than Hass, was also attended by L. fulvescens at this time. While Lasioglossum sordidum is not very hairy, Leioproctus fulvescens is. The females of both species carry pollen in pollen sacs on their hind legs. Leioproctus fulvescens also accumulates pollen on its hairy thorax, and, to a lesser extent, on the underside of the thorax. Both bees nest in holes in the ground, and prefer areas of bare earth. I suspect these solitary bees may be useful pollinators for late flowering avocado cultivars, such as Reed and Fujikawa, but there is no evidence either way.

The native bees of the Guatemalan and Mexican subtropics that pollinate avocado are adapted to the humid climate there. When a similar species was introduced to Israeli avocado orchards to improve pollination, the colonies quickly died out due to periods of intense dry heat. So the domesticated colony-forming bee, Apis mellifera, is the most common substitute we have for the native pollinator.

Here in Helensville, bees visit avocado flowers more frequently than any other insect, with the exception of the bumblebee, Bombus sp., the German wasp, Vespula germanica, and possibly the Australian paper wasp, Polistes humilis. But there is a big qualifier: not in cold weather.

Bees 'pack it in' when there is a run of cool weather (once temperatures drop below 12oC, and sun radiant energy falls below 300 watts per square metre - 800 watts/M2 is a bright cloudless day). Even so, I am not infrequently surprised at the odd intrepid bee visitor out and about in gusty or overcast conditions. Bees are good pollinators because they are more diligent than many other insects, they tend to go from flower to flower methodically, rather than spending time flying from one area of the tree to another. Like flies, they tend to clamber over adjacent flowers, and could easily knock pollen onto the stigma, as well as brush the stigma with pollen laden body hairs. Israeli researchers found that a bees head will almost always have to push against the inner whorl of anthers when it is collecting nectar. Not only does this make dislodging pollen more likely, but the pollen tends to collect in clumps just between the antennae on the bees head, ideally positioned to transfer to other flowers.

Nectar collecting bees visiting flowers
that are shedding pollen usually carry large amounts of pollen on their bodies - several thousand grains. But, (for the Hass variety) there are only an average of 25 pollen grains on the hairy underside of bees visiting the tree at a time when it is doing its female phase flowering. This points once again to the possible importance of the male phase flowers, in that bees moving between and within them are carrying a very large pollen load, so if the female stigma is still receptive in the second-day flower, it is almost certain to receive pollen. But the female flower will be visited by bees carrying very little pollen - unless there is a little overlap between the close of the male phase and opening of the female phase; or unless an 'opposite phase' avocado is flowering close by.

Nectar collecting bees don't collect pollen. Israeli researchers observe that after about every 4th flower visited, the nectar collecting bee very efficiently combs out the pollen while hanging from a leaf, or hovers and dumps it in mid air. A mid-air pollen dump causes a small cloud of thousands of pollen grains to disperse.

Bees are not always good pollinators; some bees can detect the freshly produced nectar in male flower buds that are just at the point of opening, and are adept at forcing the petal tips open far enough to suck up the sweet sugar while the bud is still closed. In this case, contact with pollen doesn't happen. It is an open question as to whether pollen masses are knocked onto the immediately adjacent stigma. I say this because the terminal and sub terminal buds of the avocado panicle are the largest and most mature, are probably are secreting the largest amounts of nectar, and possibly for that reason are favored by bees. My impression is that most avocado set is from these terminal and subterminal flowers.

Bumblebee feeding on avocado nectarBumblebees are a regular visitor to avocado flowers here. They don't always stay long, but if the nectar is flowing, there can be 2 or 3 on the same tree, and they spend quite some time collecting. Bumblebees have the ability to regulate their own body temperature, and can be the only insect working the flowers early on a cool morning, or late evening. On cool, windy, overcast days, when bees are almost absent, it can still be found at work on avocado flowers. The commonest species is Bombus terristris, and in Israel this species was noted as working faster than honeybees, visiting 20 flowers per minute against the honeybees 6 to 9 flowers. Israeli researchers found bumblebees were much less likely to concentrate solely on cultivars such as Ettinger which have high amounts of sugar. Bumblebees visited less attractive low sugar cultivars such as Pinkerton almost as frequently as the more rewarding Ettinger.

Bumblebees take some time to build up their new colony in spring. The colony starts with a single queen, and ultimately builds to maybe 200 or so workers. The small worker bumblebees appear around early to mid october here. Bee colonies, on the other hand, can, at full strength, have more than 50,000 workers - which makes them very valuable 'instant' full-strength big-number pollinators for commercial avocado orchards.

German wasps, Vespula germanica, are also common visitors to avocado flowers. Like bumblebees, they seem undeterred by cool temperatures and approaching nightfall. On a sunny day, there can be 4 or more wasps on an avocado cultivar with attractive nectar at the peak of nectar secretion. I haven't seen any information on the efficiency of German wasps as agents of pollen transfer. Likely they might knock the pollen of shedding anthers in the inner whorl onto the stigmatic surface in male phase flowers, but I doubt they carry much pollen on their bodies.

Chinese paper wasp Polistes chinensisPolistes humilis, the Australian paper wasp, and Polistes chinensis, the Chinese paper wasp (photo left), both regularly attend avocado flowers, P. humilis much more than P. chinensis. Again, their efficiency as pollinators can only be guessed at.
Microscopic examination of Polistes canadensis in Trinidad avocado orchards confirm that P. canadensis, at least, does carry avocado pollen on its body, but they carried less pollen on their bodies than bees. Presumably other Polistes are similar.
The Chinese paper wasp seems active earlier in the season, and these wasps are clearly attracted to avocado racemes even when only small numbers of flowers are open. They seem to locate honey sources with their antennae, rather than visually. This leads them to wander all over the fattest buds of the flowering raceme before they find an open flower. Sometimes they bite the yet-to-open petals of a very fat bud in a futile attempt to access the sugars within. When an open flower is found, they bring their antennae forward and parallel, 'checking the scent'. If ants are present in the flower (often the case), they will abandon it, and keep looking for a 'vacant' flower. This wandering, and aversion to ants, means that they visit relatively few flowers.

In early november 2012, I observed a European potter wasp (Ancistrocerus gazella) feeding on nectar (not pollen) in a Pinkerton tree.

Hover flies

The drone fly, a type of hover fly, has already been mentioned. While an effective pollinator, it is not particularly common.

 Both the 7-10 mm long 'large hover fly' Melangyna novae-zealandiae
(also called the 'dark hover fly', and 'the flower fly') and the 'small hover fly,'
Melanostoma fasciatum, visit avocado flowers from time to time. These 'flower flies' can be told apart by the orientation of their wings when at rest. The 'small hover fly' lays its wings parallel with its body; the 'large hover fly' holds them out at a 45 degree angle. Small hover flies seem more prevalent early in the flowering season, and are far more numerous than the 'large' hover fly. I have only seen the Large hover fly fairly late in the flowering season, when most fruit seems to set.

Again, there is little information on their efficiency as pollinators (but good information in relation to Brassicas), albeit my observations in late September 2013 indicate they quickly and accurately locate only open flowers.
Once on a nectar producing flower, however, they feed for a longer time than most other insects. Given their relatively small size, it is reasonable to suppose their energy needs will be relatively easily met, so the number of flowers visited might be relatively low. Rader et al (2012) found bumblebees visited about 17 - 55 Brassica flowers per minute, where Melanostoma fasciatum visited around 2 - 5 flowers per minute. Bumblebees deposited from ~ 50 to 500 Brassica pollen grains on the stigma. In difference, Melanostoma deposited about 1 to 14 grains.

Adult hover flies meet their protein needs by eating pollen of grasses, plantain, sow thistle, Ph
acelia, and certain other flowering plants. 'Small'  hover fly larvae eat small caterpillars (and aphids), so are 'dual purpose' insects around the avocado trees.

The 'Metallic Blue Hover fly' (Helophilus hochstetteri) visits avocado flowers when it is warm and sunny (its latin name 'Helophilus', means 'sun loving').

Sometimes a large striped hoverfly feeds on avocado nectar. It is probably the Threelined Hover Fly (Helophilus seelandicus).
Threelined Hover Fly Heliophilus sp.
Threelined Hover fly on an avocado leaf, basking and cleaning after a sweet meal.

The importance of insects other than bees in avocado pollination
The results, published in 2017 (Read et al, 2017) reflect insects recorded on the orchards over 3 days in the first week of November. Recordings were made between 10:00 am and 15:00 pm. Temperatures happened to be relatively low, even although the days were fine and with little wind. Female flower opening had largely shifted outside the survey times, presumably due to low overnight temperatures. This meant that the record for flower visitations reflects visits to male flowers only (which offer pollen as well as nectar).

The most common non-bee visitor was Zorion guttigerum, with an average of 3.4% of total flower visitors recorded over the survey period. There were 126 sightings of this insect over the 3 days. Ironically, the highest sighting (79) was in a commercial orchard, with the organic orchard being second (29 sightings). The next highest non-bee flower visitor was the bumblebee, at 1.6% of total visitors. All but one of the bumblebee visitors were Bombus terristris (the other one was B. ruderatus). Calliphora stygia, the brown blowfly (or "blowie") was observed a total of 20 times. ( Calliphora vicina, the European blowfly, better known as the bluebottle, barely featured, at 4 observations, all on the same orchard.) There were also 4 observations of a plant bug in the family Myridae. Of the flower flies, Melanostoma fasciatum was observed 10 times, and Melangyna novaezelandiae was observed a total of 8 times. There were a few other miscellaneous insects in inconsequential numbers - a soldier fly (Exaireta spinigera), an unidentified fly (a tachinid of some species or other), a native bee (Lasioglossum spp.), and various species of unidentified Hymenoptera (bees, ants, wasps, sawflies etc). A single blossom fly (Dilophus nigrostigma) was also observed.

To put this in perspective, 3,674 individual insect visits were recorded in the survey - and 3,414 of those were bee visits. Bumblebees, the next highest visitor, had only 59 visits were recorded.

The organic orchard had no beehives on it, yet recorded 512 bee visits (the others had roughly between 3 and 5 hives per hectare). Interestingly, even although the organic orchard had diverse flowering plants to attract 'wild' pollinating insects, in general, the diversity of 'wild' insects present (and their absolute numbers) was not really different from the commercial orchards.

Nocturnal Pollination
Researchers have recently found that if temperatures fall below 8C overnight, then female flowers open up to 10 hours late - about late afternoon. But if the temperature drops to 4 to 6C overnight, then about half the female flowers will not only open late, they will remain open  during the night.

Most male phase flowers remain open all night anyway when nighttime temperatures are between 7C and 15C. This means that at times of a low prior nighttime temperature, both male and female flowers will be open the following night, and there is an opportunity for flowers to be pollinated by night flying insects. However, research by Pattemore et al show that insects in general are only likely to be found when the night temperature was mild (10C and over).

So the possibility for nighttime insect pollination depends on first, a cold night to trigger delay in female flowering the next night, then, in order for night flying insects to be active, a change in the weather conditions to mild by the time the female flowers do open (the following night).

Pattemore et al (2018) found 21 different species of insects either on or in proximity to open nocturnal Hass flowers (examined between 2100 hours and midnight) that had one or more pollen grains on their body. 11 records were of a single species, found only once. 5 species had between 3 and 10 individuals of that species recorded. There were 4 species where 10 to 15 individuals were recorded - a plant bug (Diomocorus sp.), grassgrub (Costelytra zealandica), a cranefly (tipulid species), and the Tasmanian Lacewing (Micromus tasmaniae). The average number of pollen grains carried per insect was Diomocorus sp. 3, grassgrub 10, cranefly 8, and Tasmanian Lacewing 5.

Given the low numbers of pollen carrying insects and their low pollen load, the most common nocturnal flower visiting insects are unlikely to be effective pollinators. I doubt their flower visit rates would be very high either.

The only insects with 'useful' amounts of pollen on their body were a bronze leaf-beetle (Diachus auratus) - 30 pollen grains, a blossom fly (Dilophus nigrostigma)
- 26 pollen grains, and a cranefly species - 19 grains. An unknown species of the blossom fly family (Bibionidae) was a real outlier, with 176 pollen grains on its body. But one insect can only do so much!

Combining numbers of insects of a species found, plus their ability to carry useful pollen, craneflies come out as 'interesting'. But of the 13 individual craneflies caught, only a bit over half carried pollen. Apparently, craneflies are most active at dusk and night, but not all craneflies eat, as their adult phase is short-lived, and mating is a priority.

Some notes and observations on flowering and insect activity
 These are small 'snapshots' of flowering behaviour and insect activity at various times of the flowering season, and under various weather conditions. Times after the end of September continue to be 'real' time, not daylight saving time.

Unless otherwise noted, all temperature records are from a maximum/minimum thermometer on the south side of a tree, in the shade of the foliage, and at head height.

January 01

Our 'newest' Reed tree flowered very well for the first time this year. It set myriads of tiny fruitlets, be-speaking a good first time crop. These have now all fallen. Hundreds of them. However, there are (literally) one or two tail end flowers that appear as if they are setting into tiny fruitlets. I don't expect them to hold. (April 2012 - they fell off.)

March 26

One heavily pruned Hass tree has several branches with 'out of season' flowers. The weather has been warm to hot (25oC maximum) for some time, with little rain, so it will be interesting to see if there is a set.

April 03

Temperatures on one day recently climbed briefly to 28oC. The coldest it has been at night is 12oC. It was cloudy but warm until middday. The temperature then was 22oC. The 'out of season' flowers on the Hass are female phase at midday. Only ants are present. The sun came out soon after, with temperatures reaching 25oC by 1.00 pm. The flowers are still in female phase, and well attended by ants, but no other insects were observed.

April 10

The first 'out of season' flower panicles on the Hass are fading a bit, but some new ones are developing. Days have been sunny, and I have seen a German wasp working these flowers. Ants were well present earlier, not so much now. ( august - nothing set.)

'out of season' flower panicle on the Gwen has now opened its first flower. (may 16th - It didn't set any fruit.)

Pinkerton, which has an extended flowering season, is, on average, in fat bud. One tree is already extending a few panicles.

April 28

 'Out of season' flower panicles on one Hass appear to have set a few fruit. Another Hass 10 metres or so away is just flowering. At 1130 hrs it is female stage. No pollinating insects were noted (apart from ants), but the weather is humid, warm (21oC) and breezy. The Pinkerton trees are in 'fat bud', but low down on the shady side of one of the trees there are several fully expanded panicles. One panicle has some open flowers, which are female phase (at 1130 hrs). A one year old 'Carmen' Hass also has some panicles flowering. Other young Carmen trees elsewhere on the property are from the same purchase, and were planted on the same day, but have no flowers as yet.

May 10

It is astonishingly warm - 25oC in the shade at 1000 hours. It has been similarly warm for weeks. Night temperatures have been bracketed between 10oC and 16oC. Most days have been sunny or partly sunny. Wind has been almost non-existent or light. Conditions are ideal for fruit set.

There are very large numbers of Polistes around the place, but they are mainly interested in mating and 'swanning about' in the sun. There are bees, but there is very little avocado blossom around, so they concentrate on the few shrubs flowering at this time - big numbers on Lavendula dentata, small numbers on camellia. I didn't see any bumblebees.

At 1000 hrs our large but unproductive Fuerte tree is flowering well, almost exclusively on the warmer northern side. It is in male phase. There are large numbers of useless Polistes playing on it, some German wasps working hard, and a few bees, also working hard. Two of our 3 rather severely pruned Carmen Hass have some flowers, but rather not much. They are female phase. Apart from Polistes and the rare bee, there is little pollenizer activity. Maybe I should plant some Lavendula dentata nearby. And in-graft some Fuerte into Carmen.

May 16

The first flowers have appeared on one of the Pinkerton trees. It is only one panicle so far. The weather is windy, sometimes briefly sunny and warm, sometimes rainy and cold. No insects are about.

The branch carrying the first 'out of season' flower panicles on one of the Hass trees has snapped off in the wind. But other flowers panicles are developing.

May 17

Conditions are the same as yesterday, but the wind is both a bit less frequent, and a bit warmer.

At midday the Pinkerton flowers that are open seem at the 'moving between sexes stage', neither male nor female. Last year Pinkerton was often 'full on' female around 12.00 pm (standard time) in the afternoon, remaining female overnight and into the early part of the morning. In late spring, it was likely to be full female late in the morning.

At 2.30 pm the flowers are in the male phase, and pollen packets are clearly visible. Not a female is to be seen. However, it is, in theory, possible for the now 'aged' female stigma to still be receptive, if conditions are right.

The only other avocado close to flowering at this time is the Hass doing an 'off season flowering' (Fuerte would be, but it is miles away, and besides, it has been pruned heavily). Both Hass and Pinkerton are 'A' type flowerers, so there 'shouldn't' be any overlap of Hass's male flower phase with Pinkerton's female flower phase, and vice versa. But I wouldn't be surprised if their phasings were slightly out of synch, and therefore able to cross pollenize to a slight degree.

Pinkertons 'early flower flush' tends to open over an extended time, which means there are relatively few avocado flowers open on any given panicle.

To get an indication of the pollinating insects around at this time of year (and in this weather), I did a 10 minute observation of the flowers. I saw one honeybee - supposedly the most important pollinator for avocados - but all it did was a 'standoffish' survey of a few panicles, before accelerating off again at warp speed. There was a tiny 'gnat' looking insect in one of the flowers (where it stayed), but the most promising pollinator was a 'small hover fly'. The hover fly dutifully and thoroughly visited each panicle, feeding at several flowers as it went.

There are only a few avocado flowers open at this time, so the question arises, where does this insect go next? I had sprayed out most of the weeds in and around the trees, and there were only a few random sow thistles (Sonchus oleraceus) in flower. And the bright yellow flowers of the sow thistle was where the small hover fly was resting. Maybe it would be useful to have sow thistles flowering within the row in may and june, to attract more small hover flies.

It has been an unusually warm and wet autumn. Today is the first day of colder weather. One Pinkerton tree has commenced a light flowering. There is no sign of flowering on the main row of Pinkertons. Several of the Carmen Hass trees have had flowers for some weeks now.

May 18

Conditions are breezy to windy, warmer than the last few days, with the odd brief shower.

At 11:15 the small hover fly was found back on the same sowthistle flower as yesterday. Pinkerton is in female phase, but there are no 'residual' male phase flowers around to provide pollen.

It is not the case that conditions are too poor for bees and other large pollinators. I checked the dentata lavender (Lavandula dentata) plants in the garden, and I estimate there are 4 bees per square meter. A significant number of bumblebees were observed, and one wool carder bee. The wool carder bee seems like a very efficient forager - as good, if not better than, the honey bees. On the rare occasion the wool carder bee landed on a flower with a honey bee on it, it would bundle the honey bee off the flower, but did not pursue it. It ignored honey bees on adjacent flowers.

The question is whether or not growing dentata lavender adjacent to Pinkerton would attract pollinators to Pinkerton flowers, or whether they would attract pollinators away from the flowers. I have planted a Pinkerton hard up against the lavenders, so I may be able to answer this question in a couple of years time.

At 12:30 The wind is from the northwest, and it is warm enough that a cicada has burst (briefly) into song.

The Pinkerton flowers are still in the female phase. There are several aphid-sized 'gnatty' things on the flower, and one small hover fly. I stayed for 15 minutes to see who else would turn up. The only large bodied insect to show up was a honeybee. She did an exemplary job of visiting every panicle with open flowers, and a few with buds only. Indeed, she went around the flowers twice, feeding avidly, before leaving.

At 2:30 the flowers are still in female phase, and no males are present. A few flowers are either closed, or on their way to being closed. The only insect working the flowers is an ant.

May 19

Sunny and fairly warm this morning, the wind is in the south west this afternoon and it is cool to cold.

At 1:00 pm A bee and I arrived at the sole flowering Pinkerton tree at the same time. It was in female phase, but only half a dozen flowers were open. The bee searched the tree thoroughly, and after visiting all the open flowers, it then visited the panicles that were in bud. Here it selected the biggest bud, often the one near the top of the panicle, and spent some effort and time forcing its proboscis into the unopen flower. No chance of pollen transfer there...Does it matter? Probably not. The flowers selected appears to me to be ones that opened at 2:00 pm. Likely the flower will have produced more nectar by the time it opens.

2:00 pm - only a few flowers fully open, all female. An equal number are half open half closed, so are either female flowers shutting down, or male phase flowers getting ready to open. No pollinating insects are anywhere about.

4:00 pm -
only a few flowers fully open, all female. Everything else is closed. At this time of day there are no large bodied pollinating insects anywhere on the property, even in the best sites. It is cold and overcast. In contrast, the avocado flowers have 3 absolutely minute diaphanous-winged gnatty-things, a random 'mega vinegar fly', an insect that looks like a vinegar fly (Drosophila) but probably isn't, and an ant-sized black wasp, probably some tiny parasitic wasp. Not bad, given the conditions.

June 10
Today it is sunny, warm and still from late morning onward. The air temperature is only15oC, but it is a lot warmer than that in the sun. A recently planted very small 'Carmen' Hass (known for winter flowering and fruiting) has many panicles for a tiny tree, and has had small numbers of flowers open for several weeks now. There are 10 flowers open at 1330 hours, in the male phase and clearly shedding pollen. A 3 minute observation clocked up one bee, assiduously visiting the flowers, including a partially open/closed flower. Ants are also feeding at he nectaries. I have checked the tree at midday on several occasions recently, and it is always in male phase and always actively shedding pollen at around noon. There are always ants feeding in the nectaries. This is the first time I have noticed a bee visit. Other days may not have been sunny, but air temperatures would have been similar. Radiant heat temperature effects may have been much less.

Bees cannot fly continuously at ambient temperatures below 20oC. They must stop and 'bask' to build up temperatures, especially in the thoracic region 'powering' the wings. It seems to me that bees won't travel far when there is little prospect of longish periods of 'flight-powering' direct sunshine. The corollary is that it might be a good idea to have a beehive close at hand for winter flowering avocados.

June 11

Sunny and warm, temperature at midday is 17oC. Overnights have been mild due to northeasterly weather. At 1200 3 young Carmen Hass trees have flowers half way open/closed in the male phase. They are shedding ample pollen. A young Pinkerton tree around 100 meters away has commenced its 'main early' flowering and is full-on female at the same time. The stigmatic surfaces are white and receptive. No insects were observed on the flowers (apart from the ever present thrips). Numerous bees and bumblebees are visiting the immediately adjacent dentata lavender hedge, but none were observed visiting the Pinkerton avocado tree.  There were several tiny (midge sized) flies. One with wonderfully  purple iridescent eyes and an irridescent black body, offset by improbably cream little halters. One was observed feeding on nectar, but they are as likely to be basking, raising and lowering their legs in a mysterious signal system known only to other little flies, no doubt. Around 3 or 4 Nysius bugs are still present, and their stylets are clearly active around the staminoides, leaving me with the impression they are perhaps accessing nectar, rather than sucking the flowers juices. But they seem rather home-bound animals, and don't seem to move far from a group of 'home' racemes. Not much of a pollinator. Carmen Hass might be a key pollinizer for Pinkertons early flowers - if a pollen transferring insect can be found.

August 3
It has not been as cold a winter (so far) as we might expect. The lowest temperature so far (in july) was 3oC. Equally there were some warm days, and on one day the temperature reached 18oC. The day started cloudy, but is now sunny, breezy, and warm. At 11.00 it was 17oC.

The flower buds on the best Hass trees are now starting the early phase of panicle expansion. Only one of the Pinkerton trees has some panicles fully expanded and starting flowering. All the rest are about with Hass. Several of the Carmen Hass continue to flower. One has more panicles than leaves, it seems to me. This particular plant has been 'splutter' flowering for ages, and I can see it is nearly at the end of its 'panicles to come'. It is in male phase at 1100, and the flowers are attended by ants (as always), but there is no other insect activity.

August 13

It has been exceptionally mild and warm the last few days. Even night temperatures were plus 12oC here. The only avocados flowering are Sharwill and Fuerte (and various Mexican type seedlings). Sharwill has barely started. I would be astonished if any fruit set this early in flowering, in spite of favorable temperatures.

Carmen Hass has more or less finished an early winter flowering, with nothing set. A few panicles splutter on here and there. But there are no insects  - presumably the honey and pollen reward is too meager to attract them. But Roldana petasitis, (velvet groundsel), a plant from Mexico-Central America that is "sparingly naturalising in the Auckland region" (identification Ewen Cameron, botanist, Auckland Museum) is flowering heavily, and its honey scented flowers are covered in bees and (a some flies) in spite of overcast and breezy conditions. The bees are collecting pollen as well as nectar.

September 3

It has been sunny and still in the morning, and sunny and breezy in the afternoon. Temperatures in the shade are 15oC. Bees, bumblebees and Monarch butterflies are active in the early flowering 'Pixie' plum blossom.

Literally 2 or 3 Hass flowers are open on one tree. Carmen Hass continues to have flowers open. Carmen flowers are attended by ants, as usual. A solitary Pinkerton flower is open, and a bumblebee vigorously worked it, then searched some of the in-bud panicles for more. The reward from one flower seems to me disproportionate to the effort it expended. Dusa, which had a few flowers open about a week ago, now has no flowers open at all.

I noted a 'small hoverfly' around the property, but not near Carmen (the only reliable flowering avocado at the moment).

September 7

It has a been a warm (18oC +) day, but very windy. At least it is a northerly. Rain is expected by mid afternoon, and then the weather turns cooler, with a 14oC maximum predicted over the next few days.

1:00 - All the Pinkerton trees are flowering relatively heavily, and all flowers are female. A bumblebee is actively working the flowers, in spite of the wind. There are no pollenizing varieties nearby. All the flowers on the Hass trees are closed.

2:30 - There are only one or two flowers open on the Hass trees, but they are all female. The Pinkerton trees have a mix of open and closing female flowers, and flowers in 'male configuration', but not shedding any pollen. Rain is starting to set in, and no insects are around, beyond a few ants in the avocado flowers.

A warm day (18oC at 1400 hrs), sunny, clear, still in the morning, a slight breeze in the afternoon. Bees and bumblebees are in abundance after a week of rain and cloud. The lowest night time temperature in the last few weeks was 6oC. The highest daytime temperature was 22oC. The only avocados flowering (apart from some seedlings) are Fuerte and Carmen Hass. A queen wasp and Polistes chinensis was observed on a Carmen Hass, but no bees. Several steel blue ladybirds were busy eating pollen. The Fuerte was in male phase at 1400 hours, shedding pollen. The only flowers open are on the very lowest portion of the tree. All avocados are heavy with developing buds and panicles. It will be a big flowering season ahead.

September 13
A warm day, but totally overcast,with patches of light rain. The sun came out about 1545 hrs. The air was totally still, the skies clear and blue. Warmth soaked the trees.

An early flowering Hass seedling (l'il Hass) is being worked by several 'large hover flies' (Melangyna novaezelandiae), half a dozen worker German wasps, 2 Bumblebees, 2 Asian wasps (Polistes chinensis), one bee, one drone fly (Eristalis tenax) and 4 or 5 brown blowflies (Calliphora stygia).
As usual, the worker German wasps, Bumblebees, and bee are by far the most active and persistent in their work. The blowflies spend significant amounts of time just basking in the sun.

At 1600 Carmen Hass and a young Pinkerton with no crop load were both in full female phase.

By 1615, a storm front was rolling across the sky, bringing thunder, lightening and splotty warm rain with it. Would any pollen deposited on stigmas be totally washed away? I don't know.

September 21
A warm day (18oC at 3 pm), but breezy and overcast, and in the afternoon there were a few spots of splotty rain from time to time. Northerly. A few days ago we had some cold but clear weather, and night temperatures dropped to about 5oC.

3:00 - The Hass trees have a smattering of flowers open, all in male configuration but not shedding pollen. One Hass tree has a very persistent and thorough bee working the flowers, so clearly they are producing nectar,
as are the smattering of Ruby (seedling) flowers. The Ruby has a large bumblebee and a German wasp on it. In both Hass and Ruby the inner whorl of stamens is upright against the female style. The females stigma is white, and therefore receptive, but the male parts are yet to shed. In contrast, Sharwil, in identical male configuration, is shedding heavily.  Dusa has a very good number of flowers open, and all are female (by 5.30 pm the male flowers were opening, overlapping the female, but not yet shedding. The little Carmen is flowering 'hard out', and is in female phase, but attracting no insects beyond the pervasive ants (ants are in all cultivars open flowers). By 5.30 pm Carmen's male flowers were opening, but not shedding.

September 23

This morning it is sunny, breezy, warm to cool, depending on the wind. It has been cold for the last 2 days, and the overnight temperature the day before yesterday was 7oC, and last night it dropped to 5oC. I'm very pleased to see the sun today, and at 1030 it was 16oC.

There are 4 or 5 bumblebees working the lavender and other spring flowers, along with 1 of each Polistes species, but I can see only 1 bee. The adjacent Pinkerton is closed, but a  'large hover fly' sits basking on the largest closed flower bud. All avocado flowers are closed except for the 2 'B' flowering cultivars Fuerte and Sharwil, and a very 'Mexican' seedling of (probably) Fuerte x Bacon. All have a lot of flowers open, all in the male phase, and all shedding heavily. The only insects observed were numerous ants, and in the Fuerte hybrid, lots of thrips.
Carmen Hass, while mostly closed, has a smattering of flowers half-closed, with pollen evident.

September 24
This morning it is alternately sunny and somewhat overcast, with a moderate breeze with calm intervals, warm. The morning was particularly warm, and the days high was 20oC. The overnight temperature was 12oC, a lot warmer than recent days.

Ar 1615 hours, I was surprised to find one of the Hass trees (the one with a tendency to have 'loco' flowering) had quite a number of flowers open on the east side of the tree. The flowers were fully open, and some were male and some were female. The classic overlap. There are very large numbers of ants in the flowers, so numerous the flower looks yellowish black! Clearly Hass is pumping out the nectar. Ants seem to fit nicely between the stamens as they move in and out, and rarely travel up to the anthers, where the pollen packets are. The sky has clouded over, and there is an intermittent light breeze, so it is still warm. Conditions seem ideal for fruit to set.

Carmen Hass also has both male and female flowers open. Sharwil and Fuerte have numerous flowers open, and the abundant shedding pollen is very obvious. The bearing Pinkertons have a smattering of flowers open, and a young Pinkerton tree has a very good number open. At 1615, Pinkerton is full-on female. But none of these trees have any pollinating insects obvious. In contrast, 2 seedling trees, in halfway male phase, had several bumblebees and other insects on them. The difference, one might guess, could be the quantity and concentration of sugar on offer.

By 1730, the female Pinkerton flowers had closed, and the male stage was opening. These flowers almost all had brown, non-receptive female stigmas. The anthers had not yet started shedding. Self pollination is almost impossible. Many of the male flowers had thrips wandering about.

September 25

Night time temperatures have been around 10o C for the last several weeks, except for a dip to 2o C and 5o C on two nights. Daytime temperatures have broadly been between 15oC and 18oC (out of the direct sun). There have been quite strong equinoctial winds, with mixed cloudy and sunny periods.

The early flowering Hass trees continue to flower, although flower numbers are relatively small. Pinkerton continues to flower well. Pinkerton's female flowers seem to open in the late afternoon at the moment. Hass can shed pollen from mid morning to late afternoon (depending on the day!). Female flowers have been found in the morning on one day, and in the late afternoon on another. Always in small numbers,or just single flowers.

Pollinating insects are often, but not always around the flowers in small numbers at the times of my brief visits - the bumblebee is a stalwart, no matter the weather, there is often a bee or two, almost always thrips, sometimes a fly or two, and sometimes a Polistes wasp. The number of wasps visiting seems to have dropped off somewhat. Some Pinkerton and Hass flowers are heavily visited by ants when the nectar is flowing, but they seem too far below the stamens to be useful.

September 27th
Dead calm at 0830, overcast, humid, warm, already 16.5oC. Breezy by 0930, some odd sun moments between low grey clouds. Overcast. By 1030 there were more sunny breaks, warmer, hitting 23oC in the shade! Humid. Dew on grass.  Breezy/windy, with calm periods, threatening showers. Lots of insects around. At 1245 temperatures are 21oC. Overcast, bit of wind, humid, warm. About 18oC at 1600, totally overcast.

Warm temperatures, calm weather, and humidity mean conditions are very good for successful pollination. Hass started opening a few flowers around 0900, and Pinkerton had a few flowers open at 0830. All are female. By 1100 there were several Hass flowers open, and both a Melanostoma and an Asian wasp was seen evaluating some of the fattest flower buds on one tree. A single bee visitation was observed on another. By 1245 Pinkerton had a good smattering of flowers open, all female. There were ants in the flowers, but that is about all. The small Pinkerton tree planted hard up against the lavender hedge had open flowers within 40 centimeters or so of numerous bees working the lavender. None detoured to even investigate the avocado flowers on offer. In another situation  isolated from competing nectar sources there are 4 Pinkerton trees of about head height. By 2 pm they were covered in open female flowers, and were being attended by
a bumblebee, at least 5 bees, and around 4 Asian wasps.

Pinkerton is said to set fruit best when there are 'B' type pollenizer trees close nearby. The only 'B' pollenizer in the vicinity is a young 'Ettinger' tree, but its racemes are still developing, and it won't flower for at least some days yet. In contrast, a Sharwil tree in a tub and a branch of Sharwil in-grafted into the small Pinkerton by the lavender hedge have both had male flowers open in very good numbers since 0830 in the morning. However, they didn't start to shed pollen until some time after midday. By 2 pm every flower was shedding pollen. And almost every flower was attended by ants.

By far the most promising pollenizer for this early time in the main flowering season seems to be Dusa. It had quite good numbers of flowers open at 0830, all male, and all shedding pollen. The numbers of flowers both open and shedding increased quickly, and pollen was shed all day until 2 pm when female flowers stated to open. The male flowers were well closing down by 4 pm, but even then the anthers in the semi closed flowers were still shedding pollen. At 1100 a bee was persistently working the Dusa flowers, as was a Melanostoma. The bee drifted down to the in-grafted Hass portion of the tree and actively worked its flowers. Like Sharwil, the Dusa racemes have a noticeable odour.

The only other possible pollenizer variety here is Edranol, but it had very few flowers open, and none shed any pollen until the early afternoon - and even then, only one or two flowers had pollen available.

September 27th

An overcast, humid morning yielded to a slightly windy, but warm, sunny, humid day.

At 1030 it was 19oC in the shade. Pinkerton had quite a good number of female flowers open, and Gwen had a few flowers open. Fuerte and Sharwil had high numbers of male flowers open and shedding pollen, Edranol only a few. Hass and Maluma remain closed. Insects of all types abound - German wasp queens, Polistes, flies, bees, small hover flies, drone flies, ladybirds, thrips, miscellaneous bugs large and small.

By 1330 both Hass and Maluma were open in female phase, and Fuerte, Sharwil and Edranol were shedding pollen.

At 1900 hours one Hass tree still has numerous female flowers open. Pinkerton and Gwen still have the odd female flower open (Pinkerton more than Gwen). Both Hass and Pinkerton also have good numbers of male configured flowers, but are not shedding. Fuerte is still male flowering like crazy, pumping out the pollen, as is Sharwil. The only insect working the flowers is a lone Queen German wasp.

September 28th

The weather is similar to yesterday, but sunnier earlier in the day. The temperature in the shade at 0930 is 18.5oC. Last nights overnight low was 10oC.

The Hass trees and Carmen Hass have only a rare flower open at 0930 hours - except for the 'Hass selection' (a heavier fruiting type), which has numerous male flowers open and shedding pollen. Gwen also has a good number of flowers open, all male and shedding. A Pinkerton tree in a sunny location has a large number of male flowers open and shedding, where those more shaded have almost no flowers open.

Of the 'B' pollenizer types, Edranol has a few male flowers open, Fuerte has large numbers of male flowers open, but only some are shedding, Sharwill is the same, but also has a few female flowers open. Hashimoto is also male, with more flowers shedding than the other 'B' types.

It is clear that at this time of year, and these temperatures, the 'B' types Fuerte, Sharwil, and Hashimoto flower most heavily, have male flowers both open and providing pollen all day long from morning to night. Edranol is inferior in both these aspects. In addition, Fuerte is absolutely outstanding in its ability to attract pollinating insects - although Sharwil is not far behind.

1300 hours - shade temperature is 21o C
Edranol has a few flowers open and shedding. Fuerte has heaps of male flowers open -hyperfloriforous, in fact - and all shedding. It also has a few female flowers open. The tree is well attended by bees and Polistes wasps. Sharwil is also male flowering and shedding pollen, and it, too, has a few female flowers open. Once again, bees and Polistes are well present. Hashimoto has a good number of male flowers open and shedding pollen, but there are no female flowers present.

Pinkerton has good numbers of female flowers open, attended by Polistes (mainly). Maluma also has some flowers open, but not as many as Pinkerton (a few bees and the odd Polistes wasp attend). Hass and Gwen flowers are still closed.

1600 hours - shade temperature is 17o C
Edranol and Hashimoto are male phase and shedding pollen, Fuerte is still shedding vast amounts of pollen (its few female flowers are no starting to close), and Sharwil is the same. Gwen is closed, and Hass, Carmen Hass and Maluma have only a rare female flower open, but are otherwise closed. In contrast, Pinkerton has good numbers of female flowers open. Either Gwen and Hass opened as female and closed again rather quickly, or it is too cool for them to open.

September 29th

The lowest night temperature over the last 2 weeks was 5
oC. Fine and sunny at 0830, 15oC in the shade, clouded over by 0900. Perfectly still.

Fuerte has been flowering for a while now, and Sharwill has also started in earnest, and, as usual Lil Hass.
0830 hours  
Apart from a chance seedling, nothing else has flowers open this morning. Only L'il Hass has insects on it, a few bees and German wasps. This tree has massive numbers of open flowers, and may be a good nectar producer.

1430 hours - Sunny, slight breeze, some passing cloud, shade temperature 20
'Select' Hass has some flowers open, in male phase and shedding pollen. No flowers are open on the other 2 Hass trees adjacent. Fuerte remains in male phase, shedding pollen.

1700 hours - Sun low on the horizon, and the temperature has fallen to 12oC. There are no insects on the avocado trees - except for a couple of immobile steel blue ladybirds. A few female flowers are open on Hass and Carmen Hass - and the odd male flower as well, shedding pollen. There are a lot of female flowers open on Pinkerton, and, like Hass, the odd male flower, shedding pollen. I  sighted the first flower of the season open on Hashimoto and on Maluma. Sharwill still has heaps of male flowers open, shedding pollen.

September 30th

Very warm, and sunny at 1415 hrs. In fact, it was 18oC in the shade; the days high was 21oC; the overnite low last night 10oC.

There are lots of insects about. The bees are feeding on forget-me-not flowers.

Fuerte has heaps of male flowers open and shedding, as does Sharwil. Fuerte has a good smattering of female flowers are open, rarely seen! A big german wasp Queen is taking some nectar, but seems easily put off by ants. A Polistes chinensis is also feeding on it, and is busier and more organised. The 'A' type flowers - Hass, Pinkerton, Maluma, and Gwen are all closed. A bee and a Polistes chinensis scramble back and forth over the closed Gwen buds, looking for a way in. Gwen must be pumping nectar in the bud, in preparation for opening later today or tomorrow.

October 01

A warm, but windy day. Nice in the sun but out of the wind. The days high was 23oC, the overnight low was 10oC. At 1400 hrs it was 17oC.

unknown insect in avocado flowerThere were quite a number of insects in the most floriferous and stinky avocado trees, mainly German wasps, Asian wasps, and flies. There were a number of miscellaneous insects, including the quite hairy nectar drinker on the left (a ladybird species, possibly Rhyzobius lophanthae). It is probably so small it moves below the level of the stigma, and is unlikely to contact it in its wanderings.

At 1300 hrs Sharwil, Dusa, and Fuerte were flowering heavily, and all had copious amounts of pollen clearly visible. However, while Hass, Maluma and Pinkerton had only a few flowers open, and the vast bulk of their flowers were closed. Gwen remained totally closed.

October 2

A warm day, sunny with some cloud. Temperature at 1430 is 25oC, and last night's low was 10oC. These are good temperatures for fruit set, but a bit late in the day.

There are lots of bees and wasps active on the flowers, but apart from Fuerte Sharwill, Pinkerton and a few seedlings, flowering has not really started in earnest yet.

Pinkerton is in female phase at 1520, so if pollinated, pollen tube germination and growth is starting in a time of declining temperatures.

October 3

A warm, clam day, sunny with some cloud. Temperature at 1330 is 21oC, and last night's low was 8oC.

There are lots of bees and wasps active on the flowers, but apart from Fuerte Sharwill, Pinkerton and a few seedlings, flowering has not really started in earnest yet.

1130 - Gwen is closed, but Polistes humilis and some bees are trying to bust into the buds, so the flowers nectaries must be very attractive at this stage.

October 4

Last night was cool, 8oC. Warm and overcast, 21oC this morning, falling to 18oC by 1400 hours as cloud cover deepened. It is very humid, and by 1430 a steady drizzle commenced. By 1730 hours the drizzle has gone and the clouds remain.

                  species on Sharwill avocado flowersThere are a fair number of bees, wasps and flies around, but not as many as usual. There were several Zorion sp. on the open Sharwil flowers. It looks a little different to the common species with the white spots - but I may be mistaken.

At 1430 Pinkerton is female phase. Sharwil is male and shedding pollen. Late in the day, at 1730 hours, Pinkerton is still in female phase, but I think it is beginning to close and I think male flowers may just be starting to
open. There are still a useful number of bees on the Pinkertons, in spite of the late hour and cool temperatures. The only other insect evident is the German wasp.

9 October

Sunny and a warm northerly this morning. At 0830 it was 23oC in the shade! Past 4 days overnite low was 9oC. By midday it was totally overcast, breezy, cooler, 18oC in the shade.

Note: A quantum shift in the number of insects! There are far more bees, they are staying longer, the diversity of insects is there - flies, flower fly, bumblebee, German wasp, Asian wasp - but there are more present at the same time. Is it the early morning heat, or do avocado flowers produce more nectar when it is warmer? Or a bit of both? Or are insects speeding up their life cycle?

At 8.30 in the morning there are lots of flowers open, all male. The trees have attracted numerous insects - 3 or 4 bees, a bumblebee, a metallic blue hoverfly, several flies, a German wasp. A small Pinkerton tree 20 metres away is in full female flower, with lots of flowers open. Despite being immediately next to a Lavendula dentata hedge that has numerous bees on it, the only insects observed on the flowers are ants. So there is some nectar there, but obviously not enough to draw the bees off the hedge. And yet the Hass tree, 10 metres from the hedge, is very successfully attracting bees...

By 1.30 pm there are heaps of Hass flowers open. They are all female. The males are either non-existent, or are folding down as they close. There are 4 to 5 bees present per tree, and the odd other visitor, such as the 'small hover fly', and a large bodied hover fly, possibly the three-lined hover fly.

By 4.30 pm the youngest Hass tree are still female phase, but closing down. The air temperature is 17oC. There are still several bees in the younger Hass trees, even altho it has been cloudy and fairly cool in the late afternoon. The oldest Hass tree is quite some distance away, and is shaded by some young eucalypts (not that it matters on an overcast afternoon), and there don't seem to be any insects on its flowers. In contrast, 12 metres away there is a Hass seedling tree that is a mass of flowers. I have never seen a female phase on this tree, it seems perpetually male, but whatever, it is particularly attractive to insects. The pollen parent is almost certainly Fuerte, and Fuerte also seems particularly attractive to insects. Anyway, 12 metres away, this tree has at least 7 bumblebees, 5 or 6 bees, 2 or 3 German wasps, and at least 3 blowflies, all avidly collecting nectar. And this is not a very big tree. Admittedly the Hass tree has only a very small number of flowers in comparison, but even so...

Pinkerton is flowering very heavily. Pinkerton was in the female phase at 1.30 pm, and at 4.30 pm it still has a good number of female flowers open, but is closing down. There are many 'half open/closed flowers', either females closing or males opening.
The Pinkerton trees are being worked by a bee, a bumblebee and a German wasp.

10 October

Warm, a mix of sun and cloud. The nights have been quite cold - down to 8oC overnight. These temperatures are forecast to continue for the next 4 nights.

Mid afternoon there were very good numbers of bees on the Pinkerton trees. The temperatures have been such that phasing has been reversed, and the male pollen shedding phase has been in the morning to early afternoon. The female phase is just starting, and there is a good overlap between the two phases, plus lots of bees. But cold night temperatures may prevent effective self pollination. We will see.

12 October

Rain and strong winds, mostly westerly, overnight. The overnight low was 6.5oC. Cold at first in the morning, with a little scittering of hail.
The temperature at 1145 was 16oC. The sun burst through at midday, but strong westerlies continue.

The only insects on the avocados are German wasps, and the odd fly (plus a few different species of pollen eaters and nectar sippers - a Zorion beetle is the only one I recognise).

Sharwil, Fuerte and Edranol are in male phase, but only Fuerte and Sharwil are clearly shedding pollen. At 1145 hours Fuerte has quite a useful number of female flowers open, which I find suprising, given the temperatures. Pinkerton has heaps of flowers open at this time, all female phase.

October 22
Last night's minimum was 8oC. The morning broke bright and sunny, but there was fog down in the valley. It clouded over fairly soon after, and then cleared to a sunny, warm, day, with light cloud and light breezes. The previous day was warm and sunny, as was the day before. Finally we make up for the cold avocado-discombobulating weather just prior. Perhaps the flowering behave will start to return to supposed 'normal'.

There are very good numbers of bees around, black ones, yellow ones. Queen German wasps have largely disappeared, probably to get on with breeding. Polistes are around, but not many. I noticed the first native bee today, but not on the avocados. There are, as yet, almost no blowflies present, which is odd given the number of possums I have buried under the avocados over the last weeks. The other odd thing is the absence of ants in the flowers.

At 1300 hours it was 22oC, fine and sunny. Hass and Gwen are in female stage, Fuerte is in male stage and shedding copious pollen. There are large numbers of bees on the trees. This is a group of small trees growing close together, and the Fuerte is only half a meter away. Bees regularly shuffle back and forth between the immediately adjacent Gwen and the Fuerte.

By 1500 hours it was 23oC, and while a good number of female flowers are still open on Hass, it is clear they are beginning to shut down. Gwen is also closing, and seems further along in the change. There are still good numbers of bees on them.

At 2130 hours the temperature is 15oC and Hass (only) is in male phase and shedding pollen, albeit there are not large numbers open. Of course, the only insect around is the odd moth attracted to the torch light.

October 23
The day dawned cloudy and cool. At 0830 there is a light is a faint cool breeze wafting, the temperature is 18oC. The overnight low was 9oC.

In spite of the cool, there are bees and one only blowfly on Fuerte, Gwen, and Hass. All are in male phase but with few flowers properly open or shedding.

November 02

Insect activity is in high gear today. Small hover flies (Melanostoma fasciatum) and large hover flies (Melangyna novaezelandiae) are in evidence everywhere. Bees are working forget-me-nots, drone flies (Eristalis tenax) and wasps are present in good numbers, and blowflies are about. It has been sunny, showery, hot, humid, and breezy; the soil is moist. This would be the primo time for cross pollination of avocado varieties.

But the only male pollen providers at this time are edranol and fuerte, and probably Sharwil (to a lesser extent).

Hass, Reed, and Pinkerton are all in female phase at 1230. Gwen and Esther are in female phase by 3.30 pm.

There are already some tiny little fruitlets apparently set on Reed, and on one Hass. The jury is still out on whether anything has set on Pinkerton - but conditions are ideal today for the late flowers to set. We will see. (April 2012 - 2 fruit set on one tree, in an area where a flowering Sharwil (B type) is adjacent.)

November 3

Sunny, with a cool southerly breeze. The high yesterday was 19o C. The overnight low was 10oC. Currently it is 15o C.

At 10.00 am several bees are actively 'robbing' nectar from fat Pinkerton terminal flower buds. These buds will open in a few hours. There are some flowers half open/half closed, but these are not as attractive as the buds. Even a large flower fly is forlornly flying from bud to bud, before settling for 'second best', the half-mast flowers.

By 4:30 pm the trees are in female phase. There are are around 3 bees per tree, a good number considering the small size of the trees. The weather is sunny, and although there is a cold southerly blowing, the trees a fairly well sheltered, so it feels warm.

Not many flowers remain on the Pinkerton panicles. (Note from april 2013 - initially 1 or 2 fruit set per tree, but after fruit drop 1 tree has 2 fruit, 2 trees have 1 fruit.)

The young 'early setting' Hass has very few flowers left, but has (apparently, at least) set a good number of terminal fruitlets over the last 4 to 8 days. A later setting Hass 10 metres away has far more flowers to come, but appears also to have set at least some fruit. By 1630 it is in female flowering phase, but no insects are around in this area. Another Hass tree, which started to flower later than these two, has a lot of flower in the upper parts of the tree. In contrast, the panicles in the lower part of the tree are all but over.

(Note from march 2013  - the lower panicles set only one or 2 fruit. The only fruit set are in the upper part of the tree. Whangarei avocado growers reported that early november colder than usual, with overnight temperature ranging between 2.5oC and 6oC. The coldest orchards have no fruit set. Other orchards have fruit set in the warmer parts of the orchard, but not in the colder parts. Some early set fruit dropped when temperatures dipped in november. A commercial orchard on our road at a lower elevation only has fruit set in the top of the trees, the same as us. Whangarei reported that later flowers set, but earlier flowers didn't. The pattern seems the same for all of us.)

Sharwil has for all intents and purposes finished flowering.

Edranol is now in full swing, and its flowers overlap the period of time when fruit are most likely to set on Hass - whereas Sharwil is no longer effective. Bees happily move from Edranol to an adjacent Pinkerton in female flower phase - in spite of ample other nectar source immediately available (and being exploited by other bees). Edranol is still shedding pollen in the afternoon, and a German wasp is on the flowers.

Fuerte is also (still) flowering, albeit it is coming to an end. Like Sharwil, it is almost always male and shedding pollen, no matter the time of day.

The youngest of the two Reed trees here has commenced flowering, mainly on the lower panicles. At 4:30 it, too, is in female flowering phase. A bee, a worker bumblebee and a Large hover fly are in attendance.There are already a good number of 'apparently set' fruitlets on the lower part of the tree. (Note from april 2013 - only 10 fruit set and held on the tree.)

November 04
At 1400 hours the 'A' types are in various versions of female phase - most are half open, with almost no female flowers fully open. Only Pinkerton - so reliable - has a smattering of female flowers fully expanded amongst the open and half closed males. In the case of Pinkerton there is a clear overlap between male flowers still open from overnight and the first female flowers to open in mid -late afternoon. However bees - any large bodied flying insect in fact - are totally absent.

November 05

It was cold last night, with a southerly blowing. The morning is clear and sunny, warm out of the wind, but a bit cool in it. There are ample pollinating insect numbers in the avocados - german wasps, Polistes, flower flys, blow flies, and, of course, both bees and the odd worker bumblebee.

At 1000 hours all the 'B' type avos have flowers with 'spent' and shrivelled anthers, the flowers half closed - except for Sharwil, whose flowers are fully closed. The 'A' types, from Reed to Hass and everything in between, all have male flowers, many shedding pollen. There are no female flowers at all on the various B types, not on Fuerte, not in various 'B' type seedlings around the place. Edranol, also a 'B' type, is now flowering well.

By 1300 hours all the 'A' types had flowers firmly closed (except for Hass, which had a few 'late' males open). Pinkerton seems to have a fast turnaround - while all the flowers were firmly closed, there were already just a few female flowers opening. In the 'B' types, Hashimoto had some fresh males open and shedding, but not a lot. Some stigmas in Hashimoto were brown, some white. Sharwil was the only other 'B' with a few fresh male flowers open, albeit not all were shedding.

By 1800, with light beginning to fade and the cold night air descending, all the 'A' type avos were still tightly closed, with the exception of Maluma, which had just  few female flowers open, and Pinkerton, which had a lot of female flowers open. On the 'B' side, Edranol was shedding pollen, but not many flowers were open. Hashimoto, however, had a very good number of shedding flowers, and so did Sharwil. However, no insects were about in the cool wind and fading light.

At 0815 the weather is windy, clouds scudding about, sunny, rather hot already, at 19oC in the shade. The cool southwesterly winds of the previous days have now turned northwest and warm. Night temperatures have been pretty good over the last few days - the lowest temperature was 12oC, not bad.

Except for Reed, which has just started, most avocado flowering is now all but over. There are good signs of set on some trees, probably from several weeks ago, in the vital late October period.

Reed flowering seems a little late this year.

At 0800 hours the 'A' types are mostly closed from yesterday and last night's male flowering. A few have not closed properly, and the pollen shedding stamens are still visible - and sometimes
slightly protruding - through the slightly parted tepals. Interestingly, the Gwen tree has male flowers open and shedding on the 'cold' side of the tree away from the morning sun, and closed male flowers on the sunny side. Hass has come pretty close to the end of flowering. There are only a few buds left to open, and then that's it for the year. There are no insects on the 'A' trees.

With regard to the 'B' types, edranol is towards the end of it's flowering, and in a sort of unconvincing semi-open male phase. Sharwil has all but finished flowering, Fuerte is mostly closed, but, interestingly, has a smattering of female flowers open, and the ever-reliable Hashimoto has lots of male flowers open and clearly shedding pollen. Edranol had a blowfly (and a zorion beetle) on it, but nothing else. At least the blowfly was actively working the flowers.

November 06

This morning is hot - 21oC at 1000 hours, sunny, cloudless, with a light breeze.

Maluma is still mostly closed, but those that are open are male and shedding pollen. Pinkerton, which had a good number of female flowers open in the early evening yesterday is now mostly closed, barring a smattering of knocked about flowers shedding pollen. Most of the Pinkerton flower panicles have the majority of flowers with signs of having set. So most the 'A' types missed female flowering yesterday (apart from Pinkerton), but the male flowering was in the warmest part of the day. If most flowers set at the male flower phase, then the lack of the female stage is no problem.

November 07

It has been mixed sunny, warm, humid, and overcast; the soil is moist after several days of intermittent warm showers (at last - it has been rather dry for weeks). Bees have been working Reed heavily over the last week or so. Prior to the showery days, Reed was a wall of flowers, plastered with bees (and the odd worker bumblebee). Conditions were ideal for fruit set - hot, humid, night temperatures never below 10oC.

Pinkerton racemes have almost finished flowering, and are looking skeletal. There are a few flower buds to open yet, but not a lot. The odd bee continues to attend.

The only male pollen providers at this time are edranol and fuerte, and probably Sharwil (to a lesser extent).

At 0900 Reed is semi-open and being heavily visited by bees. There is a substantial smattering of female flowers open on Pinkerton.
At 1200 Hass is pretty much closed, just the odd female phase flower open. Pinkerton has a good number of female flowers open. Maluma is just opening a few flowers, female phase. The only male offering pollen is Edranol, which has an abundant supply of males, shedding well. Reed is in female phase, starting at 1230.

At 1.30 there are still a few Hass female flowers open, but sod all. There are no insects (apart from ants) on the racemes, which is hardly surprising given the low amount of nectar available on the tree.
Even by 1.30 there are still relatively few flowers open on the Reed, but the sparsely distributed open flowers are well visited by bees, albeit some bees 'bud bust' yet-to-open flowers. Clearly, Reed is a very rewarding nectar producer, as far as bees are concerned, anyway. In contrast, while Pinkerton still has a good smattering of female flowers open, insects are largely absent (apart from ants).
Edranol is still chucking pollen everywhere at 1.30, but again, apart from the ubiquitous ant, there is no obvious interest from the insect world - in spite the good number of flowers open. Interestingly, the modest flowering on one of the new Hashimoto trees is also in male phase at 1.30, also with good pollen production.

There are lots of tiny little fruitlets apparently set on Reed, and on the two young Hass trees. I'm not sure if anything much has set on the row of Pinkertons - 'spent' flowers are hanging on in numbers, but no sign of the little green developing fruitlet yet. I am not confident, I have seen heaps of 'apparent sets' on Pinkerton before, only to see them all fall off a day or so later. Conditions are ideal today for the few remaining late flowers to set, so there is still hope.

In contrast, a younger Pinkerton tree far distant from the row has a good number of little green fruitlets just visible under the shrivelling flowers tepals. There is a Sharwil in-graft close by, now finished flowering. The only difference between this tree and the distant Pinkerton row is that there is an Edranol tree a few metres away, flowering heavily, whereas there is no Edranol in the vicinity of the row of Pinkertons. However, the Edranol has been flowering ever since the Pinkerton was put in, and it has never set fruit previously, so no conclusions can be drawn. (August 2014 - the Pinkerton row went on to produce a very good crop, whereas the younger Pinkerton tree adjacent the Edranol set nothing.)

November 17

It is oppressively humid, hot (21oC at 1000), totally overcast and still. It has been warm and dry for days and days. We have not had any rainfall other than a brief shower. The soil has once more dried out to a deep level. Night temperatures have been a little cooler, with one night dropping to 8oC.

Avocado flowering is essentially over except for Reed and Hellen.

We are now at the stage where the pin-head sized little set fruit are clearly visible. It has been a 'bumper year' for Hass fruit set, and clearly the same heavy initial set is happening on Reed. The earliest Reed panicles are covered in little fruitlets, and the main flowering panicles are flowering heavily and well attended by numerous bees and the odd worker bumblebee.

Both Hass and Pinkerton are over except for the odd straggler flower (both cultivars are female phase at 1000). Hass panicles are thick with pinhead fruit, Pinkerton panicles have set very few fruit, and those that have set are clearly from a few weeks ago.

Interestingly, a Sharwil in-graft in the Pinkerton row has a very respectable number of recently set pinhead fruit. In addition, I found a further 3 or 4 pea sized fruit, clearly set a few weeks ago (2014 - all these fruitlets ultimately dropped off) . A small Pinkerton tree on another part of the property is also within a metre of a flowering Sharwil in-graft, but, in contrast, no fruit have set on that Sharwil in-graft. The difference between the two situations is possibly that there are several flowering seedling avocado trees opposite the Pinkerton row, but none near the isolated Pinkerton tree. The seedling tree 1.5 metres from the Sharwil in-graft has flowered a little before Reed, and has only just finished. It might possibly be responsible for cross pollination with Sharwil.

Sharwil has enormous numbers of male phase flowers, but I have rarely seen female phase flowers in Sharwil (Edranol is the same). Sharwil may need higher temperatures to switch to female flowering. Maybe only female phase flowers will set. This implies the female stigma is non-receptive in the male phase, or that it has some self sterility mechanism with its own pollen (Sharwil produces mega amounts of pollen, and the stigma will surely have numerous grains deposited on it in the male phase). This implies it requires pollen from another source.

If extra heat is required to initiate the female phase, then this explanation fails on the simple fact that it is no warmer in the Pinkerton row in-graft than Sharwil in-grafts elsewhere on the property, so I doubt temperature differentials explain it. If the Sharwil in-graft in the Pinkerton row sets fruit again next year, I will have to consider whether or not an adjacent flowering seedling is acting as a pollenizer...but which one? Reed can't be excluded as pollenizer, either. Mysteries abound in the world of avocado pollination and fruit set...

The only possible male pollenizer around is Edranol, but it has only the barest number of flowers left. A European potter wasp attends them.

Hellen is also in female phase at 1000 this morning, and the small tree has a respectable number of flowers open. The small Reed tree 14 metres away has 5 bees and a bumblebee on it, the Hellen has no pollinating insects on it at all. (2014 - Hellen went on to set a heavy crop of fruit.)

Reed has ejected several fruit overnight, and leaf drop has moved into hyperdrive (along with flowering). The new vegetative flush at the tip of the flower panicles is now really starting to move. Hass is by far the most advanced in flushing, and speed of growth. The call on the tree's resources must be large. I will cut most of the flush off, at least on the Hass trees being size-trained.

If it rains, growth on the avocados will be explosive. I hope it doesn't draw resources away from the set fruit.

As always, we will see...

November 23

Hot, humid, sunny. At 0930 it is 24oC already. At 1030 it is 26oC. The days high hit 27oC. Conditions were the same yesterday. It has been hot for many days. There was an isolated heavy rain shower for about an hour several days ago, but the sun came out again after it, and it has been dry ever since.
The Reed trees flowers are mostly closed at 0930, in spite of 24oC temperatures already.  In spite of this, the avocado flowers smell seminal, as they did yesterday morning. (Chestnut flowers have the same odor, as do an orchid species in the genus Coelogyne, the pear species
Pyrus calleryana and some other plants. Apparently semen-like floral odor is not uncommon in plants; it is said to be due to "basic amines such as putrescine, spermine, spermidine and cadaverine". Cadaverine and putrescine are evolved from decaying flesh, which may explain why blowflies are initially attracted to avocado flowers.)

In spite of the heat and the smell, the number of pollinating insects is low relative to days past. There are 3 -4  flower flies (both species) present, 4 bees, 1 Polistes chinensis, and 1 bumblebee.

Interestingly, the tree was in a heavy female phase at 1:00 pm yesterday, but had an area on one side of the tree which was in male phase. This morning (0930) those male phase panicles are now open in the female phase. The bulk of the tree has no flowers open at 0930.

Temperatures have been consistently high for days. Yesterdays overlap of male and female flower opening seems without obvious cause. By 2:30 pm the whole tree was in flower - in the male phase. Even although it was hot, sunny, calm, there were almost no pollinating insects on the tree - just 2 small flower flies, and they were not really doing much. The male flowers were not giving out any odor, in contrast to the female stage.

Hashimoto still has some flowers open, male phase to Reeds female. It would be the ideal pollenizer for Reed - if Reed needed a pollenizer. But it doesn't.

Hot, humid, sunny. At 1300 hrs it was 26oC in the shade, but had hit 30oC earlier. Apart from Reed and Hashimoto, flowering is all but over. There are always a few straggler flowers, but most of the setting has been done. This has been an exceptional year for avocado fruit set. There have been periods in late October -early November that have been ideal for fruit set - warm nights, warm days, not too much wind. We don't have beehives here, but bees have arrived in big numbers, their numbers have been remarkably high. In contrast, bumblebees have been almost entirely absent, which is unusual. In spite of burying 4 or 5 possums under the trees early in the season, blowflies have been almost absent until very late, and then around only in very low numbers. The bees have done their job - there is a massive set on the trees. Most, of course, will fall off, but it still promises a good crop.
Remarkably, our big Fuerte, which sets very few fruit (but lots of cukes), appears to have a good set this year. Another temperature dependant tree which normally only has 8 - 12 fruit is showing a massive initial set, the best I've seen in 30 years. Our tiny Sharwil, which had 3 fruit last year, looks like it might do better this season.

November 24
Hot, humid, sunny. At 1130 it is 24oC. Overnight low was 14oC. At 1130 Reed is in female phase for the second day in a row. There is no smell. 2 flower flies are present, but just flying around the panicles, not working the flowers. in contrast, there are 4 bees on the panicles, working well. Once again, Hashimoto is in opposite phase (male)
to Reed. Esther is also flowering well, and, like Reed, is in female phase.

Today is rainy and windy with total cloud cover and a low cloud base. It is expected to be like this for some days. Temperatures
res are on the low side of mild. The spring flush, still at an early stage in many plants, will now burst away rapidly.

November 27
Avocado flowering is essentially over, with the exception of Fujikawa, which is barely mid flower, and Reed, which is flowering well.

Reed is setting well, in contrast to Pinkerton, which has set very few fruit. Our tiny Gwen tree, which totally defoliated this spring, has apparently set fruit (I expect them to fall off, either now or later).

It is very obvious that Hass and Reed are essentially self fertile, even if fruit set is increased by cross pollination.

Pinkerton has a very long female flowering season, and Pinkerton has flowers right through the Hass season, and maybe a little past. The weather is warm, and there are insects around, so it is clear that Pinkerton needs a pollinator.

First indications from our hip-high Gwen tree planted at the end of the Pinkertons is that it may be self fertile. We have a very unhappy Esther tree in a half barrel, and it too has apparently set some fruit, in spite of having dropped all its leaves, so it is worth watching. (April 2012 - nothing set)

Bees have barely been present on the flowers. In fact, it was rare to observe a bee on the avocados this year. Even on warm, relatively calm days, there was little bee activity on the flowers - even when they were active gathering nectar or pollen from flowering weeds at the base of the trees.

In contrast, it was common to find one or more German wasps working the flowers, no matter the weather or time of day., and the same is true for the bumblebee. The odd Asian wasp was also common, especially earlier in the season. Flower flies and hover flies were a feature from mid season onward. There were myriads of tiny 'gnatty' things at certain points, always thrips, often ants, and a wide variety of other small insect miscellanea feeding on the nectar (and in the case of the steel blue ladybird, feeding on the pollen).

I doubt poor fruit set on some of our avocado cultivars is due to lack of pollinating insects. It seems some avocado varieties are self fertile, and some more or less aren't.

30 November

Avocado flowering is more or less over. There was a very heavy set of fruitlets on Hass about 3 weeks ago. Almost all of these have now fallen over the last 4 or 5 days, and it is now clear that fruit set is rather patchy and muted on the 2 youngest trees, but OK on the 2 oldest. The young Hass trees are heavily pruned to control size, the older trees are either unpruned or one major branch removed regularly.

The fruit that remain are largely those that set early, probably around mid october. These are now pea-sized or larger. There are a number of very recently set tiny pinhead fruit on some panicles, and possibly some may hold as very late-set fruit.

Reed has also had a huge number of flowers set to the pinhead fruitlet stage. The youngest tree has started pinhead fruit drop, but it is too early to tell what the retention rate will be.

Fuerte has a very large umber of fruitlets set, but pinhead fruitlet drop has yet to start. There are still a fair number of Fuerte flowers open.

Hellen has now finished flowering, but is yet to shed fruitlets. Hashimoto has also finished. There are odd rare individual 'rogue' Hass, Pinkerton, and Maluma flowers. The first native bees have emerged, a bit late for most avocados.

Like Hass, the Pinkertons have now shed almost all their pinhead fruitlets. And like Hass, the only fruit to set are those that formed about 3 weeks ago. Now the fruitlets on Pinkerton are pea-sized or greater it is easier to see the set for the year. Each tree has at least 12 fruit, a greater number than my first rough estimate. This is the largest (apparent) set ever on these trees.

Some references
L. Davenport.
1999. 'A view from Florida on avocado pollination' In: M. L. Arpaia and R. Hofshi (eds.), Proceedings of Avocado Brainstorming. Session 5. Pollination.
Pages 101-104. October 27-28, 1999. Riverside, CA.
Hofshi Foundation. http://www.avocadosource.com.

T.L. Davenport. 2003. 'Evidence for wind-mediated, self and cross pollination of 'Hass' avocado trees growing in Mediterranean environments.'

Proceedings V World Avocado Congress (Actas V Congreso Mundial del Aguacate) 2003. pp. 221-226.

M. Librada Alcaraz and J. Ignacio Hormaza. 2009. 'Avocado Pollination and Fruit Set A Perspective from Spain'
California Avocado Society 2009 Yearbook, 92: pp 113-135

Paul D. Cooper, William M. Schaffer, Stephen L. Buchmann, 1985. 'Temperature regulation of Honey Bees (Apis mellifera) foraging in the Sonoran Desert.'
Journal of experimental Biology. 114, 1-15 (1985) 1

Gardiazabal, F.J., and S. Gandolfo. 1995. A study of self-pollination and cross-pollination in avocado (Persea americana Mill.) cv. Hass of different varieties.
Proceedings of The World Avocado Congress III, 1995 52-56

Harrison, R. A. 1990. Bibionidae (Insecta: Diptera).
Fauna of New Zealand, 20, 25 pp.

Fauna of New Zealand.
Irvin, N. A., et al 1999. The phenology and pollen feeding of three hover fly (Diptera: Syrphidae) species in Canterbury, New Zealand.
New Zealand Journal of Zoology 26: 105-115.

Ish-Am, G., Eisikowitch, D. 1993. The behaviour of honey bees (Apis mellifera) visiting avocado (Persea americana) flowers and their contribution to its pollination
Journal of Apicultural Research 32(3/4): 175-186 (1993)

Janice M. Hickman , Gbor L. Lvei & Stephen D. Wratten. 1995. 'Pollen feeding by adults of the hoverfly Melanostoma fasciatum (Diptera: Syrphidae)'
New Zealand Journal of Zoology, 22:4, pp 387-392

Munro, Martin. 2013. 'Passive pollination'.
The New Zealand Avocado Growers Journal (Avoscene), December 2013 p12

Papademetriou, M. K,. 1976. Some Aspects of the Flower Behavior, Pollination and Fruit Set of Avocado (Persea Americana Mill.) in Trinidad.
California Avocado Society 1976 Yearbook 60: 106-152

Pattemore, D. et al 2018 Low overnight temperatures associated with a delay in 'Hass' avocado (Persea americana) female flower opening, leading to nocturnal flowering.
Journal of Pollination Ecology, 23(14), 2018, pp 127-135

Rader, Romina, Howlett, Bradley G., Cunningham, Saul A., Westcott, David A., Edwards, Will. 2012. 'Spatial and temporal variation in pollinator effectiveness: do unmanaged insects provide consistent pollination services to mass flowering crops?'
Journal of Applied Ecology, Vol 49, 1. pp 125 - 134

Read, S.F.J. 2017. Insect visitors to avocado flowers in the Bay of Plenty, New Zealand
New Zealand Plant Protection 70: 38-44

If you have corrections, observations, or comments, feel free to drop a line to:

 avocado [and next insert the symbol "@"] lauriemeadows "dot" info. Apologies in advance, I don't check email very often, so acknowledgement may be a wee while coming.