Pruning trees and shrubs in late summer is NOT recommended because such pruning causes new soft growth that can’t harden off before winter. You will find this advice everywhere, but is it correct?
Pruning Trees and Shrubs
The woody stems of trees and shrubs go through an annual process of maturation. In spring new growth starts and continues into early summer. The growth at this point in time is called ‘softwood’ because it is soft and pliable, and can be bent to a large degree without breaking. If you bend it too much it will snap and break but in some species you can actually tie the stem into a knot without breaking them. This softwood is not yet mature, and it can’t withstand freezing.
As the summer progresses, the plant slowly converts the softwood into hardwood. The bark thickens and and the color changes from green to brown. The hardwood becomes much stiffer and does not bend very well. Along with the physical changes there are internal chemical changes that make it more resistant to low temperatures. It is getting ready for a freezing winter.
Pruning trees and shrubs stimulates new growth. If pruning is done in spring, the new growth has time to mature and become hardwood before winter starts and it can then survive the cold of winter. However, if pruning is done too late in the season, there is not enough time for the new growth to harden off–it runs out of time to become hardwood. When winter comes, it will freeze and die off.
To combat the problem of winter die off, it is commonly suggested that you do not prune trees and shrubs in late summer or early fall. If you don’t prune at this time, you will not cause new growth, and winter will not damage your woody plants. This all makes a lot of sense.
But ….how early is too early to prune late in the season?
This question was discussed on social media a few months ago, and it was suggested that pruning in mid August might not cause new growth. The following experiment was carried out to determine if it is OK to prune in late summer.
Experiment Design
On August 15, 2014 I went around my garden and took cuttings from a number of shrubs. I tried to include a variety of species to get a good cross section of shrubs. In most cases, two different types of cuttings were made:
- This years growth (starting to harden off)
- 2-3 year old growth (fully hardened)
I wanted to see if there was a difference in response to the two types of cuts. In theory, the tip of this years growth should contain higher levels of auxins (plant hormones) and be more responsive to a cut. Older wood should require a longer time to mobilize chemicals to the right spot and wake up dormant buds–at least that seemed like a rational explanation.
If pruning in late summer causes new growth then most, if not all, of the cuts would generate new growth. In that case a second round of cuts was planned for September 15 to see if pruning done at the start of fall would initiate new growth.
The following is a list of the shrubs used in the test. All were well established and have been in the current location for 3 to 7 years. Unless noted otherwise (see notes below), the shrubs had stopped their spring/summer growth period. The third column indicates whether new growth took place in the 30 days following the cuts.
| Plant Name | Type of Cutting | Did New Growth Occur? |
| Buddleja davidii | Current year | Yes |
| Spirea X vanhoueti | Current year | No |
| Spirea X vanhoueti | 2-3 year old | Minimal new growth |
| Cephalanthus occidentalis | Current year | No |
| Cephalanthus occidentalis | 2-3 year old | No |
| Berberis ‘Rose Glow’ | Current year | No |
| Berberis ‘Rose Glow’ | 2-3 year old | No |
| Sambucus Canadensis | 2-3 year old | No |
| Salix purpurea ‘Nana’ | Current year | No |
| Salix purpurea ‘Nana’ | 2-3 year old | Minimal new growth |
| Philadelphus coronaries ‘aurea’ | Current year | No |
| Philadelphus coronaries ‘aurea’ | 2-3 year old | No |
| Spirea nipponica ‘Snowmound’ | Current year | No |
| Spirea nipponica ‘Snowmound’ | 2-3 year old | No |
| Hydrangea paniculata ‘Pink Diamond’ | Current year | No |
| Hydrangea paniculata ‘Pink Diamond’ | 2-3 year old | No |
| Syringa ‘Atheline Wilbur’ | Current year | No |
| Syringa ‘Atheline Wilbur’ | 2-3 year old | No |
| Abelia mosanensis | Current year | No |
| Abelia mosanensis | 2-3 year old | No |
| Caryopteris divaricata ‘Snow Fairy’ | Current year | Yes |
| Physocarpus opul ‘Coppertina’ | Current year | No |
| Physocarpus opul ‘Coppertina’ | 2-3 year old | No |
| Cotinus coggygria ‘Grace’ | Current year | No |
| Rose, no name, small pink | Current year | Yes |
| Hibiscus syriacus ‘Red Heart’ | Current year | No |
| Hibiscus syriacus ‘Red Heart’ | 2-3 year old | No |
| Viburnum opulus ‘Roseum’ | Current year | No |
| Viburnum opulus ‘Roseum’ | 2-3 year old | No |
| Syringa pekinensis | Current year | No |
| Syringa pekinensis | 2-3 year old | No |
Notes:
1) Caryopteris divaricata ‘Snow Fairy’ and Buddleja davidii were growing at the time the cuttings were made and these shrubs tend to keep growing until frost. The amount of new growth was the same on cut stems and non-cut stems. Both shrubs are root hardy in zone 5, but are not stem hardy. They are treated more like perennials in this climate and are cut back each spring.
2) Plants with only a ‘current year’ cut, are plants that are normally cut to the ground in spring, so there was no 2-3 year old growth to cut.
3) The Rose is a vigorous fall grower in zone 5, and normally keeps growing and flowering until frost. There was more growth on the cut stems than the non-cut stems. The rose stems usually die back to ground during the winter.
4) Two shrubs, Spirea X vanhouetii and salix purpurea ‘Nana’, showed ‘minimal new growth’ on 2-3 year old stems. Minimal new growth is 3-4 very small leaves. Unexpectedly, the pruning did not initiate grow on the current years growth.
5) 2014 was an unusual year weather wise. The middle of the summer was colder than usual which resulted in some plants flowering earlier, and others flowering much later.
Best Time to Prune, Conclusion
At the test site located in Southern Ontario in 2014, pruning on August 15 did not result in new growth for the majority of shrubs. Except for the rose, new growth was minimal or non-existent.
It is possible that a more major pruning (ie removal of a large number of stems) may result in new growth. It is also possible that the results of this test are unusual given the cold summer weather.
Since pruning mid August caused very limited new growth, the mid September pruning was not carried out.
The data from this test suggests that late summer or fall pruning will not result in new growth, and there should not be any concern about winter damage. The commonly accepted recommendation, not to prune in late summer or early fall does not seem to harm the shrubs. Pruning in late fall or winter certainly works, but late summer or early fall should also work in zone 5.
References:
1) Photo Source: (Cloud-pruning in Hallyeo Haesang National Park, Geoje city, South Korea)
Does plant really need pruning? Shouldn’t left it grow naturally? Do you have article about this?
no
I need some advice, please. I live in Upstate New York in zone 5B. Moved to a new home last fall and cut down what I now believe are Endless Summer or possibly Mountain hydrangea. They have come back and now, in early October, I have a few fading blossoms and some new white buds starting.
Do I cut them back to the ground in late fall? Late winter… when they’ll most likely be partially covered in snow? Do I wait till early spring? I believe the blossoms are considered to be growing on new wood since I obviously cut off everything last fall!
Appreciate any advice!
Endless summer blooms mostly on old wood, and then again on new wood. In zone 5 it is not a reliable bloomer at either stage and many people here no longer grow the plant for this reason.
It should not be cut back until after the spring bloom.
This might help: http://www.gardenfundamentals.com/hydrangea-identification/
I think we are getting a bit complicated here. As a horny handed gardener with 40 years experience I can confidently say that some plants definitely shouldn’t be pruned to early (unless you intend to prune again in November…). I am thinking particularly of Yew hedges, which will regrow after an August pruning and the new growth will freeze, looking unsightly. I also used a very rough guide: ‘plant (or prune) Spring flowering species in Autumn, summer and Autumn flowering species in Spring’. Sometimes difficult to decide if something is late Spring or early Summer flowering, but that doesn’t seem to make a difference.
Your rule of thumb will work in most cases, but pruning spring flowering shrubs in autumn will remove the flowers for next year. Spring flowering shrubs develop their buds during summer for the following year and an autumn pruning removes them.
Even if we follow your rule, one has to define what ‘autumn’ is. Late fall after leaf drop is certainly a safe time to prune. What this blog post tries to determine is how early is too early.
What about Yews? It is certainly possible that an August pruning, leads to new growth–they seem to grow all the time. Does this new growth get damaged in winter? That was not tested in the above list of plants–a project for next year.
OK, you are quite right of course, I was maybe confusing ‘autumn planting’ with autumn pruning. Substitute ‘summer pruning’ for autumn pruning, but you have to prune sometime, do you not?
I had a huge yew hedge to look after. It would get a going over in July/ August, followed by a final trimming after the first frost. Not pruning in November resulted in the top growth going yellow due to freeze back, so you learnt about yews that way. There are so many different climate zones to consider that it is difficult to establish firm and fast rules about this kind of thing.
when talking about things like pruning it is always important to know about things like climatic zones, that is why I mentioned that this testing was done in zone 5. If you are in another zone you need to adjust the dates.
Unfortunately people like simple rules and will base gardening tasks on dates–which is not the best way to go. Learn to look at the plants. What are they doing? Base gardening tasks on what the plants tell you–not on specific dates.
I prune early flowering shrubs right after blooming. Some evergreens such as pines as the candles elongate. Everything else gets pruned in winter when I have nothing else to do in the garden.
Pruning late in the season could stimulate metabolic activity at a time when plants should be “shutting down.” Metabolically active cells presumably would be more susceptible to winter injury. More useful that looking for new growth would be to monitor cold injury at the end of winter, which would likely also be dependent on the severity of the winter cold.
Can you stimulate metabolic activity without also starting the growth process? I guess it might be possible to change the internal chemical processes so cells stop getting ready for winter. But I think (just a guess) that the processes to get a cell ready for winter are driven more strongly by light durations and temperatures, than by pruning.
The reason for looking for new growth is that most warnings about pruning too early indicate that it results in new growth that does not have time to harden off. This test seem to disprove this. You are quite correct that it does not disprove the idea that early pruning may still result in winter injury of a different type.
Measuring winter injury would be much more difficult. It is possible that you see more winter damage on stems that had been pruned than none pruned stems? But it is would probably be best to test whole shrubs–some pruned and some not pruned. And it should be a multiyear process to take into account yearly variability. The pruned areas on the shrubs in this test will still be marked in spring so I can see if there is any differences between pruned stems and none pruned stems.
Briefly . . . You could have metabolic activity being stimulated without a gross, visible sign of growth stimulation. Measuring winter cold damage is not at all difficult; just look for dead stems or parts of stems. As with any horticultural “experiment”, you need a sufficient number of test subjects to pin down variability on the treatment you are testing rather than weather, parts of plant treated, etc.