Category Archives: planting

Trees for bees

Well, not just for bees, for wildlife generally.

But bees in particular ūüėČ

I’m a beekeeper and I’m keen to keep bees on Ardnamurchan. This isn’t the place to describe the pleasure I get from keeping bees – though here are some clues:

Local honey

A garden without beehives feels empty to me and it’s something I intend to fix soon.

Honey bees and bumble bees

Honey bees (Apis mellifera) are the only species of bee managed in Europe for honey production. The clue is in the name; in Greek, Apis means bee and mellifera means honey-bearing. They are small, not particularly hairy and they live as a large colony Рa superorganism Рcontaining a single queen, up to 50,000 workers (all half-sisters and daughters of the queen) and a couple of thousand drones (males).

Queen, workers and a drone honey bees

Honey bees overwinter as a colony, using large amounts of stores (honey and pollen) to keep the bees alive through the coldest months of the year, until the days lengthen and the queen starts laying eggs again.

In contrast, bumble bees are (often) large and very hairy, they form small colonies containing perhaps 200 workers. The queen hibernates alone overwinter and starts a fresh colony the following spring.

Bombus terrestris – the buff tailed bumble bee

Ardnamurchan is awash with bumble bees of several different species. I’ll post something more about them when I finally get around to identifying them all 1.

In contrast, where I live there are no honey bees at all ūüôĀ

Throughout 2019 I searched for honey bees. There are lots of wildflowers and heather for forage but no honey bees.¬†The willow trees – more about these in a minute – up the hill are thick with bees on a good day. In late April, on a calm day, you can hear the bees working the willow from 25 metres away … but none of them were honey bees.

Where have all the honey bees gone?

There are honey bees on Ardnamurchan. I know of a few other beekeepers on the peninsula. None are particularly close to me Рeither via the B8007 or as the bee flies.

Honey bees forage up to three miles from the hive. So this suggests that there are no hives within range. But what about wild (or feral) honey bees? These live in hollow trees or cavities in a wall.

This is where the story gets a bit more complicated and I have to make a distinction between Ardnamurchan and ‘just about everywhere else’.

The vast majority of honey bees colonies in the UK (and globally) are infested with an ectoparasitic mite called Varroa. The mite feeds on developing honey bee brood and, while doing so, transfers a smorgasbord of unpleasant viruses that Рquickly or slowly Рkill the bee.

Beekeepers use a variety of control measures to reduce mite levels and so avoid losing colonies. If done conscientiously Varroa control is extremely effective.

Varroa parasitised honey bee

In the UK the average annual colony losses are 25-30%, largely because of the ravages of the Varroa mite and the viruses it transmits 2.

In the absence of Varroa control the colony will die within a maximum of two to three years. Many colonies die within just one season, often when overwintering due to the mite reducing the longevity of worker bees.

Which brings me back to wild or feral colonies. Before the mite was introduced to the UK (in 1992) wild colonies were commonplace. Now they are exceedingly rare; studies show they have high pathogen levels and probably only live for a year or two.

So does that explain why there are no honey bees in my part of Ardnamurchan?

Probably not … parts of Ardnamurchan, and other regions on the far North-West coast of Scotland, are still free of¬†Varroa.

As a beekeeper this is an exciting prospect and something to be very carefully protected 3.

So where are the honey bees?


Although honey bees are distributed globally, northern Scotland is at the far north and west of their native range.

Global distribution of the honey bee (Apis mellifera)

It’s likely that the climate is borderline for the long-term survival of wild or feral honey bee colonies.

Rain (of which we have more than sufficient) reduces the time colonies are able to forage for nectar and pollen. Damp, cool conditions are tolerated well by bumble bees (this is one of the reasons they are both larger and hairy) but can leave honey bee colonies low on stores and close to starvation.

Loch Sunart, July 2018

Beekeepers can compensate for this. Hives can be supplemented with nectar or pollen, particularly very early in the season when natural forage is limiting and the weather is poor. This enables the colony to build up sufficiently to be self-sustaining through the remainder of the season.

This isn’t a luxury that wild or feral colonies have. If they starve to death there are too few other colonies on the peninsula to re-populate the environment 4.

Natural supplements

Although bees might be able to fly three miles when foraging, they’d prefer to fly 30 metres and would use up a lot less energy doing so 5. It therefore makes sense to try and enhance the pollen and nectar sources in the immediate vicinity.

Most importantly, sources available early in the season are likely to provide the greatest benefit.

Many of the best early season pollen and nectar sources are trees, not flowers.

Therefore, before I get bees I’m busy planting trees.

Willow and hazel are both excellent sources of early season pollen. Willow also produces reasonable amounts of nectar. The plan is to plant significant amounts of both.

Goat willow, male catkins

Both hazel and willow have additional advantages. They are native trees and are a focus for lots of wildlife other than honey bees.

Hazel provides almost year-round interest. Pollen-bearing catkins are produced in February to April, followed by the fresh green leaves which remain on the trees late into autumn.

Willow grows well in damp areas. We have some very damp areas which are of little use for much else. Willow is rich in salicylic acid which can make the leaves unpalatable to insects and browsing animals. Anything that is unattractive to deer has to have a better chance of surviving as there are far too many of them in the area 6.

It is important to try and provide overlapping sources of early season pollen and nectar, rather than just lots of a single type with short availability.

I will therefore be planting quite a bit of alder which, like willow, is good in damp ground and provides pollen. In addition there will be native fruit trees – hawthorn, blackthorn and cherry – all of which are great for wildlife and provide flowering interest and nectar early in the year.

Making space

I’ve been clearing the triffid-like rhododendron from large areas of the site over the last year. This has been backbreaking work, but I’m starting to make progress … and have generated at least a full years’ supply of firewood.

One day all this will be under rhododendron

All this now looks very different …

Rhododendron creates such oppressive ground cover that there were almost no native trees or plants left in the cleared ground. However, within the year, a few things are starting to sprout. I’ve scattered some native wild flower seeds in the hope that they will get established before the land is invaded with bracken or reinvaded by rhododendron.

There’s still (lots) more to do, but for the first time I have some space to plant trees.

Native transplants and willow cuttings

Most trees will be planted over the winter(s) as bare-rooted ‘whips’. These establish quickly and are relatively inexpensive. However, they are only available when trees are dormant and they can be lifted from the seedbed.

I’ve previously discussed the fate of the majority of self-seeded native trees. They germinate, grow a few inches or a couple of feet … and are then browsed to the ground by deer.

I’ve therefore spent the year ‘rescuing’ self-seeded trees from the area – mainly hazel, rowan and oak – and moving them to parts of the plot I’d prefer them to grow, protected by a tree guard 7 where appropriate.

Willow and rhododendron

I’ve also taken willow cuttings from some local trees up on the hill. Willow are dioecious – with separate male and female catkins on different trees – and there are (conveniently) one of each adjacent emerging from a sea of rhododendron within a short walk.

If you cut foot long willow twigs somewhere between a pencil and a forefinger thick they readily produce roots if left standing in water.

Willow cuttings

I took cuttings in mid-April and they were ready to plant by mid-June or early-July. Within 4-6 weeks there was new leafy growth and by the end of the growing season some had put on 12-18″ of growth.

Rooted and shooting willow cutting, early August

Others, less well protected, had been eaten by the deer ūüôĀ

I’ve been reading the excellent¬†Handbook of Scotland’s Trees by Reforesting Scotland¬†which has a chapter on how to avoid the attention of deer. One suggestion is to plant huge 3m willow ‘cuttings’¬†i.e. entire branches, so that the growing tips are out of reach of the deer. I’ll be giving this a try in 2020.

In the meantime the first ~100 bareroot ‘whips’ have arrived and are heeled into pots waiting to be be planted. This is planned for Christmas and the New Year, each being supplemented with a sprinkling of mycorrhizal fungi and organic blood, fish and bone to help it get started.

Bareroot willow, blackthorn and gean (wild cherry) whips.

With a bit of patience and some tender loving care they should develop well over the next few years, providing pollen and nectar for the bees, and generally improving the habitat for other wildlife.

Happy Christmas and New Year


Trail cameras

There is a lot of wildlife on Ardnamurchan. It is one of the reasons I returned here time and again … and came back more permanently in 2018.

Some of the wildlife is obvious.

Siskin (male)

Siskin (male)

The never-ending parties of tits, siskins and finches on the bird feeders.

A white tailed eagle effortlessly gliding along the scarp slope of Beinn Bhuidhe.

The stag ghosting away in your peripheral vision as you drive back late along the B8007. Did you really see it at all?

An otter rolling amongst the bladder wrack exposed at low tide as it searches for crabs.

Otter, Loch Sunart, Glenborrodale, October 2018

Otter, Loch Sunart, Glenborrodale, October 2018

But most of the wildlife you never see because it is about when you’re not.

Either the wildlife makes itself scarce when we’re blundering about … or because it’s the middle of the night.

Smile, you’re on candid camera

But with a bit of technology, some patience, a bit of trial and error and a little luck you can see what you’re missing.

Trail cameras (also know as trailcams or camera traps) are movement-activated cameras that record still images or short videos any time of the day or night. During daylight hours they record colour images. At night they use infra-red LED lighting and record in black and white.

I’m using camera traps to detect the wildlife I don’t see because I’m chainsawing out the rhododendron and making a mighty ruckus.

And I’m using them in the vain hope that they’ll help identify the gaps the undesirable wildlife get through to eat the native flowers and trees I’m planting in the newly rhododendron-free areas ūüôĀ

Red deer hind scouting out a rhododendron-cleared area for newly planted trees

There are dozens of makes to choose from at prices ranging from ¬£25 to at least 20 times that, depending upon the features required. The more expensive models have 4G or wifi built in (and I’ve no experience of these).

Optically – at least from the technical specifications provided by the manufacturers – trail cameras above ¬£150-175 don’t improve very much. Less expensive models tend to offer lower resolution and longer trigger times (I’ll discuss the significance of both of these shortly).

It’s worth noting that none of these ‘hobbyist’ camera traps are going to produce stills or video to compete with the sort of stuff you see on Seven planets, One world¬†1.

However, they can produce perfectly acceptable results.

Roe deer buck

I’ve used the Browning Recon Force Advantage and Browning Spec Ops Advantage¬†cameras.

Yes … the names are totally ridiculous ūüėČ

I felt like I should spend the weekends disguised as a bush, or at least be wearing a camouflage hat and jacket, when I ordered them over the phone 2.


Both these trail camera models deliver still images of 4 – 20 MP (megapixels).

This is nothing like as good as it sounds I’m afraid. This is because the larger images are¬†interpolated when they’re scaled up – effectively adding pixels as the image is expanded.

I’ve only tested the still camera function during daylight and can see no significant quality differences between 4 MP and much larger images.

Do not expect images comparable to your digital SLR … they’re not even vaguely close.

Video is better being up to 60fps 1920 x 1080p HD 3. This is perfectly acceptable and I usually use them at 30fps to save SD card space.

Less expensive models usually offer lower video resolution and/or a lower frame rate. Both reduce the quality of the resulting video, though perhaps not so much it actually matters.

Lights, camera, action

Well, lights, battery life and action.

To work at night the camera needs to illuminate whatever it has picked up using the motion-sensitive sensor. It does this using infra-red LEDs. The two cameras (embarrassingly) named above use different types of LEDs.

The Recon Force Advantage has ‘low glow’ IR LEDs. These are visible to the human eye and they are certainly visible to animals (as I regularly capture video of them reacting adversely to the light).

Badger – note the surprise when this¬†poorly-sighted fellow sees the ‘low-glow’ LEDs

The Spec Ops Advantage has ‘no glow’ IR LEDs. These are essentially invisible to humans 4, but are effective over a shorter range (~70% the distance of the low glow LEDs).

The other significant feature to look for in a trail camera is the trigger speed i.e. the time it takes from detecting motion to capturing video (or a still  image).

0.4 seconds might not sound very long, but a small animal – like a stoat – moving fast across the field of view will feature for a further half second on the video before you get 10 – 20 seconds of just herbage waving in the breeze.

Mouse … one of several hundred I have videos of ūüôĀ

Both cameras use eight AA batteries. Unfortunately, not just any old AA batteries. You’re strongly recommended to use lithium AA batteries which are appreciably more expensive 5.

Why lithium? You need batteries capable of delivering high current – at least 2500mAh – and they often need to work at low temperatures. Standard rechargeables or Duracell simply cannot do this.

I get 9-12 months use out of a set of lithium batteries in a well-placed camera (i.e. not too many false triggering events) set to take short videos, about 50% of which are recorded at night. So, although the batteries are not inexpensive, they also last well.

I’m also dabbling with building a solar-powered rechargeable battery pack, though this will only be suitable for areas which receive full sunlight at least some of the time. It will make it less portable, but might be worthwhile for a really good location (or a very remote one).

Irrespective of the technical details of the camera trap, the biggest influence I’ve seen on actually capturing useful and usable images is due to the positioning of the camera.

Location, location, location

Firstly, things to avoid …

  • Glare from the sun. Try and position the camera facing north. Or at least somewhere in the arc between north-east and north-west. Not an issue at night of course, but why compromise daytime images beings spoilt by glare and dazzle? 6
  • Herbage moving close to the lens. This will repeatedly trigger the camera during windy weather. There’s only one thing less interesting than screening 89 short videos containing the same small branch being blown intermittently in and out of the field of view.
  • Mounting the camera on something that moves. This almost always ensures lots of false triggering. Even 8-10″ tree trunks sway perceptibly in strong wind. A fencepost is good, as is a tree-stump 7.
  • Mice. This is what’s less interesting. That infuriating branch I just mentioned is insignificant when compared with a couple of hundred ten second videos of mice scurrying around in the foreground ūüôĀ

Red deer stag and glare from poorly positioned camera

Obvious places to try …

  • Game trails. Large animals usually leave a track or trail, particularly if the ground is soft. I usually point the camera along the trail to get a head-on (or tail-on!) video, rather than a fleeting glimpse as they cross the field of view.
  • Paths. Many animals will follow man-made tracks rather than making their own way through the undergrowth. This is where I’ve had most success in filming a wide range of wildlife – from badgers to woodcock.

Roe deer hind on manmade ‘path’ hacked through the bracken

  • Bait. I’ve not tried this but you’ll see many camera traps set up pointing at bait of some sort. Some people use an egg, others use cat food or – for pine martens – a digestive biscuit with jam.
  • A waterhole. Don’t expect hippos and gazelle, but you might well get an otter or animals coming to drink. I don’t have anywhere really suitable so can’t comment on how effective this is.

The camera should almost always be mounted 30-60cm above the ground. Most trail cameras have a small screen you can use to check the positioning. If the camera is only 30cm from the ground this can be tricky to view 8.

Try to position the camera vertical and avoid too much sky in the field of view. The former makes the resulting videos/photos look a bit better quality and the latter avoids lots of ‘blown out’ highlights in the images.

Trail camera firmly strapped to the lower trunk of a tree

The great thing about these cameras is they’re pretty much ‘set and forget’. You can leave them for a week or a month. You certainly don’t need to check them on a daily basis. They just sit there, quietly grabbing images when triggered.

SD cards

Camera on game trail

All of the images are recorded onto full-size SD cards. I only use 16 or 32 Gb cards and have only ever run out of space when I setup the camera incorrectly and recorded very long video clips.

With 16 or 32 Gb SD cards now costing only ¬£6-8 it’s easy to carry spares and swap them with the SD card in the camera. I’ve also built a portable, battery powered, Raspberry Pi Zero backup device which copies the card in situ. However, this gets press-ganged into other roles and, frankly, it’s faster to just swap cards over.

Absence of evidence is not evidence of absence

The two most obvious signs that badgers are in an area are their extensive latrines and their corpses at the side of the road ūüôĀ The badger (above) was the first I’d seen on Ardnamurchan and I had no reason to suspect they were there.

The camera never lies.

Pine marten, Ardnamurchan, May 2018

In contrast, it’s not uncommon to see pine martens lolloping across the road and their scats (faeces) are often left in prominent places 9.

Pine marten scat

Of course, the ‘biggy’ in terms of mammals on Ardnamurchan, is the Scottish wildcat. Although this area was designated as a wildcat ‘haven’ about 5 years ago, things have gone a little quiet since then.

Scottish Wildcat Action are now concentrating on Morvern which, although not far away, is across Loch Sunart and likely forms an isolated population.

Nevertheless, over the next year or so I’ll continue to leave at least one or two trail cameras well away from areas of human disturbance in the hope of capturing one on video.


Deer and reforestation

The moist and mild climate of Ardnamurchan provides almost ideal conditions for the germination of plants and trees. Wherever you walk, if you look carefully, you can find miniature birch, diminutive rowans or little oak trees.

Birch seedling

Parts of the peninsula have extensive broadleaved woodland and the area is famous for its oakwoods around Salen and Ariundle. We have hazel copses, a large amount of willow and alder carr, some grand old oaks and lots of birch.

Or lots of rhododendron, depending where you look, though it is gradually being cut back.

Mixed established deciduous woodland has excellent biodiversity and oak trees are famed for having more associated species of wildlife – including bacteria and fungi, lichens, free algae, mosses, vascular plants, invertebrate animals, birds and mammals – than any other tree in the UK.

Oak, late November

However, unfortunately, it’s true to say that on Ardnamurchan – and in lots of other places in the Highlands –¬†mighty oaks from little¬†acorns¬†do not¬†grow¬†1.

The missing generation

It’s easy to find seedlings a few inches high and the mature trees are obvious. What’s missing, or at least significantly under-represented, are saplings and small bushes.

For example, we have two or three rowans (mountain ash) close by the house. All are mature trees (~15+ feet tall) laden with berries from late summer onwards. All spring and summer I’ve been finding 3-5″ tall (short?) rowan seedlings, perfect in everything but stature.

But there are no rowans on the plot that are waist or head height.

It’s all, or almost nothing.

Well, not quite. You can find these missing generation(s) of spindly saplings growing from cracks in steep rock faces or other inaccessible spots.

You can also find a few shin-high, multiply-branched, twiggy, de-leaved, stunted specimens. These often have fat little trunks and extensive root systems, out of proportion to the tree height, like a poor quality bonsai.

Browsed rowan

The grim reaper

The inaccessible saplings and the thick-limbed but dwarfed individuals give an obvious clue as to the fate of the missing generation.

They are being, or have been, browsed to extinction.

There’s no regeneration of native woodland because as soon as a tree is tall enough to be noticed it’s a midnight snack for deer.

On very steep slopes the deer cannot graze so the saplings escape. However, there’s often little soil present, or the ground is unstable, or it dries out too much, so the tree never reaches maturity.

Accessible saplings get grazed and grazed again. The new juicy growing tips are nibbled away, leaving ever-thickening twiggy branches. Year after year this results in a stunted, distorted little bush, with lots of woody growth but few leaves.

Too much of a good thing

There are six species of deer in the UK – red, roe, fallow, sika, muntjac and Chinese water deer. Of these, the first two are native to the UK with the rest being imported. However, fallow deer were here before the last ice age and were re-introduced after the Norman Conquest (11th Century).

Red and (particularly) roe deer are widespread and, in places, common.

Deer … they’re everywhere

Too common.

In Scotland, red deer densities (PDF) can be greater than 30 per km2 in places, with very high levels in the Grampians and north and west of the Great Glen.

It is estimated that the current red deer population in Scotland is in excess of 450,000 head. This number has increased at least three-fold since 1959 when The Deer (Scotland) Act was introduced in response to a growing awareness of the damage to agricultural crops and native woodland.

In 1959 the recommendation was that an optimum number of red deer in Scotland was ~60,000. ‘Optimum’ in terms of reducing crop damage and allowing natural reforestation without ugly and expensive fencing.

Presumably not ‘optimum’ in terms of animals available for stalking. Which is a significant part of the problem …

Simon Pepper (ex-WWF Director for Scotland and Deer Commission) has written a brief account of “the deer problem” in Scotland and Reforesting Scotland nicely describe the impact and management of deer. I won’t rehash the case they make here but I think it’s compelling.

The camera never lies

Our plot is not fenced and deer regularly come down off the hill or out of the surrounding woodland. I occasionally spook them when I’m wandering about clearing rhododendron or birdwatching. At night their visits become more frequent … we might not see or hear them but the dog detects them with her ultra-sensitive radar and growls quietly.

They trample regular ‘game trails’ across parts of the plot, moving from one block of woodland to another during the night, or going down to the shore.

I’ve positioned a trail cam in likely looking spots and regularly ‘see’ red deer and, about 10-fold less frequently, roe deer. Whether this reflects the population density of the two species or is a consequence of the locations used for the camera is unclear.

It’s very common to ‘see’ a red deer hind with a calf in tow. Stags are a lot less frequent.

At times they descend from the hill mob-handed, late in the evening, intent on the destruction of almost everything except rhododendron (which unfortunately they don’t eat due to the grayanotoxins in the leaves).

Reforestation and stalking

It’s ironic that deer are woodland animals, yet their presence in large numbers prevents the regeneration of the forests in which they should flourish. Red deer from open moorland are smaller, less well conditioned and have higher winter mortality than woodland deer.

Inevitably, large scale culls have been unpopular with the general public and the landowners who provide stalking.

Importantly though they have been effective. They have reduced the numbers of deer on the land and have allowed the natural regeneration of forest without the need for fencing (which brings its own problems).

Richard Baynes has written about the consequences of the deer cull in Glen Feshie, both in terms of the uproar in the shooting community and in the resulting improvement in reforestation. The income from stalking is the same now as before the cull; clients are “prepared to pay more to hunt a scarcer, wilder animal”.

For a more extensive discussion of reforestation and biodiversity, accompanied by outstanding photographs, I recommend you have a look at Alan Watson Featherstone’s blog.


I love deer.

Particularly braised slowly with a good quality, full-bodied red and winter vegetables ūüėČ

However, I don’t like venison so much I could make a serious dent in the local population of deer.

Therefore, other measures are needed to deter deer and to encourage the regeneration of trees. In turn this will decrease the triffid-like invasion by rhododendron and increase the native wildflowers and understory vegetation.

Deer fencing remains an option but is unsightly, expensive and Рbecause of access issues  Рwould require gates and/or cattle grids. Properly maintained it is the only certain way to exclude deer.

Lion dung has proved tricky to source, but may not be effective anyway.

I’ve planted a few tree seedlings ‘rescued’ from inaccessible areas, together with a dozen or so willow ‘sticks’ to provide early season pollen for bees. Several of these have already been decimated by the deer.

Transplanted rowan seedling

I’ve therefore protected some with mesh tree guards to help them get well-established.

I’m intending to plant another hundred or so bareroot native trees this autumn. Rather than littering the plot with dozens of unsightly corrugated tubes I’m hoping to exploit the topography to barricade access points, only putting tree guards around the saplings on the periphery. Additional temporary fencing will be added where needed if the barricades alone prove to be ineffective.

Willow cutting with tree guard

Or when … ūüôĀ


Laga Farm on Ardnamurchan has a large fenced area of hillside overlooking Laga Bay in which native trees have been planted. The sparse woodland is only six years old but is already developing really well and is wildlife-rich … and will undoubtedly get better. In addition to the planted trees there is also a large amount of natural tree regeneration now that the deer are excluded.

Laga Farm Native Woodland