Treating Garlic Mustard in the Offseason

Late last winter I began testing the effectiveness of spraying garlic mustard, Alliaria petiolata, rosettes with glyphosphate.  Spraying in late fall through winter has the advantage that most native species are dormant at that time.  While I did not have early satisfaction of find a lot of dead garlic mustard plants it did seem as though there were many fewer blooming stalks by mid-spring.  You can read about that in my blog post ‘Fighting Garlic Mustard with Fire? Or Something Else…’

Here's a typical patch of garlic mustard as it looks in November.  It is accompanied here
by Japanese honeysuckle, another invasive that can treated at this time.

This fall I am repeating the spraying.  I had to wait until after some sub-freezing temperatures to make sure that the native vegetation was dormant and hence unaffected by the glyphosate spray.  The advantage of spraying in late fall is that there is less risk to the native vegetation, particularly the spring ephemerals, like Dutchman’s breeches and spring beauties.  Another reason to wait until fall is that a large number of garlic mustard seedlings (80-90%) do not survive the first year, as reported by The Nature Conservancy. So by waiting there will be fewer plants to treat and less herbicide used.  The drawback to fall spraying is that the fresher leaf cover on the ground can hide more of the rosettes.  Since I spray the individual rosettes rather than blanket spray I can move the leaves out of the way as I work and can use a lot less herbicide.  You can find a report on the effectiveness of winter spraying in this paper by Frey, et al. 

Another invasive that can be treated in fall is Japanese honeysuckle, Lonicera japonica.  Here on my zone 6-7 property Japanese honeysuckle does not go completely dormant and is still susceptible to glyphosate spray.  I sprayed a plot that was fairly dense with the honeysuckle this past week.  We’ll see in the spring if it made a difference.

There are a number of other invasive ground covers that are still green now, but according to the literature I found foliar glyphosate is not particularly effective at this time.  Among these are vinca, Vinca minor, creeping Charlie, Glechoma hederacea, and mock strawberry, Potentilla indica.  So despite my desire to be rid of these I did not waste any of my spray on these unwelcomed plants.

By target spraying I can avoid the native species that still have living foliage.  These include some plants with rounded leaves similar to garlic mustard like white avens, violets, and golden ragwort.  Other natives to avoid spraying include sedges and cool season grasses, ferns and any other early spring plants coming up early.

 

Winter rosettes of white avens have whitish veins.  You can see how
it compares to the deeply veined leaves of garlic mustard,
marked with white*'s.

In comparison to garlic mustard,
violet leaves are smooth and somewhat glossy.

Golden ragwort leaves are palmately veined
and are regularly toothed on the margins.

A fall trip through the woods also turns up other invasives with distinctive foliage or berries like winged euonymus, barberry, and Oriental bittersweet.  Small specimens of these can be pulled from the moister fall soil. These can also be treated with 20% solutions of glyphosate using the 'cut and paint method.'

I spotted this burning bush/winged euonymus because it still had foliage on it. 
Others with their namesake bright red foliage were even easier to see and pull.

To control garlic mustard one needs to use a combination of tactics appropriate to the situation and season.  Winter spraying with glyphosate will be easier and more effective than fire.  In the spring, pulling or targeted spraying would be most effective.  And for those plants remaining in the late summer, cutting close to the ground or continued pulling will be in order.

Summary of Garlic Mustard Control Measures

Method	Timing	Pros	Cons	Effectiveness
Fire	Late Fall to Early Spring	Non-toxic.	Difficult to achieve ideal conditions; Need to keep under control.	A good moderately hot fire is effective, but difficult to achieve.
Herbicide Spray	Dormant season	High kill rate; can be targeted; no soil disturbance.	Spraying toxic materials; may affect non-target species.	Dormant season spraying reduces non-target species effects.
Cutting	Summer, after flowering	Non-toxic; cutting at ground level nearly 100% effective; minimal soil disturbance.	Labor intensive; disposal of cut stems/flower stalks; use of weed whacker causes collateral damage.	Very effective when done right.
Pulling	Anytime ground is soft	Non-toxic; very effective as long as most of the root is removed.	Labor intensive; disposal of pulled plants required once flowers are present; soil disturbance.	Very effective.

 

Leaves of Three

Poison ivy has three leaflets, the center one has a longer stem.  Leaves have a central vein
with secondary veins branching off from it.  Young leaves are glossy, but older ones are variable
 and leaf margins are all over the place.

After a mild winter in the Mid-Atlantic, it's been a cool, damp spring.  This seems to have brought out a lot of growth in the understory.  Included in this lushness is plenty of poison ivy, Toxicodendron radicans, formerly, Rhus radicans.  As I have been doing some weeding I have been vigilant for ‘leaves of three’, the rhyme for identifying poison ivy.  But that is not the only plant out there that sports three leaflets. 

Following are some common plants that bare some resemblance to the dreaded poison ivy.  Probably the most common look alike in the Mid-Atlantic region is box elder, Acer negundo.  When I first encountered this tree I thought OMG it’s a  poison ivy tree!  This tree reseeds prolifically generating a myriad of seedlings with bright green leaves divided into three leaflets.  While superficially similar to poison ivy, on closer examination you can see that box elder has an opposite arrangement of leaves and branches, while those of poison ivy are alternate.

Box elder has three leaflets that appear very similar to
poison ivy, but branching is always opposite.  Leaf margins are variable

Here they are side-by-side.  Box elder seedling at top has opposite branching,
poison ivy in mid-frame is opposite.  The yellow flowers are green and gold,
  Chrysogonum virginianum.

A close second in my experience is Virginia creeper, Parthenocissus quinquefolia. This rambling and climbing vine has the same habit as poison ivy, but it usually have five leaflets rather than three.  The confusion comes because the younger shoots often sport leaves with three or sometimes four leaflets.  When I spot these I carefully trace the vine back a little ways to see if it also has leaves of five.  Some people have sensitivity to Virginia creeper, but the reaction is not as severe as the rash most people get from poison ivy. 

Virginia creeper is a vine with a similar habit as poison ivy. 
Most, but not all, leaves have 5 leaflets.

Another understory tree that has three leaflets is common hoptree or wafer ash, Ptelea trifoliata. This is not likely to be confused with poison ivy, although seedlings or growth from the base of the tree could cause some concern.  On the hoptree each leaflet does not have a distinct stem (petiolule), rather the leaf tapers sharply to the point of attachment.  While the leaflets of poison ivy are distinct with the central one considerably longer than the two lateral ones.  

Common hoptree grows well in shady locations not unlike poison ivy. 
Note how the leaflets lack distinct stems.

Aromatic sumac, Rhus aromatica, is less commonly encountered.  In the wild it is an upright shrub.  But in the landscape trade there is a shorter, spreading cultivar called ‘Grow Low’ that is becoming very popular.  When I’ve bumped into mine in the woodland edges I froze for a second until I noticed that the leaflets have different lobes and the leaflet stems (petiolules) are all the same length.  On poison ivy the middle leaflet has a longer stem than the other two.

The leaflets of aromatic sumac tend to have rounded lobes concentrated at the tips.

White avens, Geum canadense, has a number of leaf forms.  Some of its basal leaves can have three leaflets but they are not particularly glossy and are generally rough in texture.  Also this species grows in clumps, it is not viney.  

The younger leaves of white avens tend to have three leaflets.  To the lower left you
can see some of the more complex leaf forms.

Barren strawberry, Geum or Waldsteinia fragarioides and the non-native W. ternata, have leaves of three, but the leaf margins are more deeply toothed and the leaflet stems are all very short.

The glossy leaflets of barren strawberry appear to merge together looking
 more like lobes than separate leaflets.

  Mock strawberry, Potentilla  (or Duchesnia) indica, is another prolific ground cover with three leaflets.  It’s leaf margins are regularly toothed and it’s habit is different, spreading by stolons, not vining.  

The leaf surface of mock strawberry is much rougher in appearance than poison ivy.

Golden Alexanders, Zizia aurea, has leaflets that are twice divided, 3 sets of trifoliate leaflets (biternate).  The leaflets are ovate to lanceolate with finely toothed edges.  The long petioles may give the impression of young poison ivy stems.  

Leaflets of golden Alexanders can be seen in the circle at bottom right. 
Also present in this image is Virginia creeper and mock strawberry.

The native clematis, Virgin’s bower, Clematis virginiana, is a weak-stemmed vine with three part leaflets.  While you would want to routinely test the vine strength to distinguish it from poison ivy, an examination of the veins on the leaves would show the difference.  Virgin’s bower has a palmate pattern, with the major veins radiating from a single point; whereas poison ivy’s veins branch out along a central middle vein.  

The ribbed stems of virgin's bower are too weak to support itself
 and it needs something to climb on.

The list of trifoliate plants seems to go on and on.  After I thought I had this pretty well wrapped up I took another look and noticed the leaves of Jack-in-the-pulpit, Arisaema atrorubens, and that reminded me of the trilliums.  These can be recognized by their relatively large leaf size and regular arrangement of leaves.  Arisaema leaves are in a ‘T’ arrangement and those of trilliums are arranged in a regular triangle pattern (120 deg apart).  Can you think of any more poison ivy look-alikes to be on the watch for?

Fighting Garlic Mustard with Fire? Or Something Else...

A couple of years ago I added a garden torch to my arsenal of tools to maintain the landscape.  Fire can be a natural and effective tool in controlling invasive species and is one of the better ways to maintain a prairie ecosystem.  But, fire can also get out of control and its effectiveness is diminished if it is not timed properly or its intensity is not sufficient to destroy the target species.  On the garden scale a garden torch is a good way to kill weeds growing in cracks in a drive or walkway.  It is also effective at killing young seedlings, particularly annuals.  I found it particularly useful for  for clearling plants away from the partially buried wire fence around the vegetable garden.  Perennials and established plants require much more heat to kill them.  For established weeds repeated burning 2-3 weeks apart may be necessary.  Though it seems counter-intuitive, flaming moistened soil is more effective at killing young plants and seeds than dry soil because the moisture helps conduct  the heat through the soil.

Here's my Benzomatic garden torch.  It is light weight and a good size to fit into tight spaces. 
The one pound tank lasts an hour or so, depending on how big a flame you use.  

I originally got my torch to use on Japanese stiltgrass.  While spring burning does kill the sprouts effectively, other plants are actively growing at this time and the fire will set those back as well.  Also stiltgrass germination occurs over an extended time period, so multiple burns would be necessary.  For stiltgrass in a cool season lawn, burning in late summer, while the perennial grasses are in summer dormancy, is very effective at preventing the stiltgrass from setting seed.

This success with stiltgrass got me to thinking about using the torch on garlic mustard, Alliaria petiolata.  Garlic mustard is a monocarpic biennial. It spends its first year as a rosette of deeply reticulated, reniform (kidney-shaped) leaves that builds up energy reserves in its fleshy white roots.  The following year it sends up a 2-4 foot flowering stalk in mid- to late-spring. It seems that winter time might be a good time to take out the garlic mustard.

Winter rosette of garlic mustard.

In the past when I have tried to burn garlic mustard I found I had to hold the flame on the plant for a long time before I could see much damage.  Before burning through a lot of propane, I decided to do a little research on using fire against garlic mustard.  The U.S. Forest Service has reports that document the effects of fire on a number of important plant species, including invasives.  Follow this link to the Forest Service report on garlic mustard.  A study authored by Victora Nuzzo for the Illinois Department of Conservation in 1990 compared the effectiveness of fire, herbicide and cutting for the control of garlic mustard.  They found that for fire to be effective it needed to be moderately intense and carried out in early spring.  Ideal burning conditions were often not available.  Herbicidal spray treatments were most effective in early fall and spring, timed for when native vegetation was dormant and thus less susceptible to the herbicidal effects.  One study out of Ohio showed that winter-time application of glyphosate when temperatures were between 25-45 F was very effective at killing garlic mustard rosettes.

Cutting plants after seed formation had begun in mid- to late-summer was very effective at reducing the number of plants in subsequent years.  Cutting at ground level resulted in nearly complete mortality, while cutting a couple of inches above the soil surface was only about 70% effective at killing the plant.  Since flowers and seeds continue to develop on cut stems, it is important to bag and dispose of all the garlic mustard cuttings.

Pulling mature garlic mustard is also very effective, and can be done anytime.  As with cuttings, if there are any flowers present the pulled plants should be bagged and disposed of.  While completely non-toxic, pulling garlic mustard is labor intensive (pulling and clean-up) and disturbs the soil.  This allows for the germination of additional seeds buried in the soil.

Here are some of the interesting garlic mustard facts that I came across while researching this post:

•        Seeds lie dormant for a year before germinating.  This means you may not see the effect of control measures on new seedlings until the second growing season.
•         Flowers are self pollinating; many are already pollinated before the flower opens.
•         Flowers can photosynthesize.  They can continue to mature to form seed even after they have been cut or pulled.
•        Plants are monocarpic.  They will continue to live until they set seed, even multiple years.
•        Seedling mortality is high, 80-90%, during their first year.  More effort should be expended on removing maturing plants particularly in late fall to early spring.
•       Of the common herbicides, glyphosate and triclopyr are very effective at controlling garlic mustard, 2,4-D is not.

So, Burn or Not?

So to answer the original question, should I use fire to control my garlic mustard?  The answer, for me, is no.  I will need to use a combination of tactics appropriate to the situation and season.  Winter spraying with glyphosate will be easier and more effective than fire at my location.  In the spring, pulling or carefully targeted spraying would be most effective.  And for those plants remaining in the late summer, cutting close to the ground or continued pulling will be in order.

To test this I will target spray with 3% glyphosate in a section of the woods in February if (when) we get a stretch of days in the 40’s with no rain and see how that compares to untreated areas.  (February seems a little early, but we are in the 60’s in early January, so plants may be getting woken up a few months before normal.)  Rather than blanket spraying I will try to target the spray onto existing rosettes.  One of the reasons to use target application is to avoid those few natives that are still in leaf through the winter. 

Some native plants that may look similar to garlic mustard rosettes include violets, Heuchera, Tiarella, and white avens (Geum canadense).  Ground ivy (Glechoma hederacea) and common mallow (Malva neglecta) are some weeds that look similar, but are not currently on my hit list.  (I found no studies that indicated that ground ivy is susceptible to herbicides in winter.  Based on that I wouldn't waste time and materials treating it in the winter.)

Compare the palmate lobed leaves with the deeply veined (reticulated), bright green leaves
of the garlic mustard.  Here, pulling is the only option for removing the garlic mustard.

The leaves of foamflower are distinctly lobed.  The vein pattern
is different from that of garlic mustard.

Golden ragwort has regularly toothed leaf margins and a more linear vein pattern.

White avens over winters as a loose rosette of leaves.  Though most winter leaves
 have 3 lobes, some appear roundish and vaguely similar to garlic mustard.

The leaf shape of common mallow is similar to that of garlic mustard,
but the vein pattern is different.

In areas where I am more actively managing the garlic mustard I will continue pulling plants through late spring (before flowering).  In late summer I will see if there is a practical way to cut the plants close to the ground without leveling the surrounding vegetation.  Otherwise, keep pulling!

This table summarizes the garlic mustard control methods mentioned in this post.

Summary of Garlic Mustard control measures

Method	Timing	Pros	Cons	Effectiveness
Fire	Late Fall to Early Spring	Non-toxic	Difficult to achieve ideal conditions; Need to keep under control	A good moderately hot fire is effective, but difficult to achieve.
Herbicide Spray	Dormant season	High kill rate; can be targeted; no soil disturbance	Spraying toxic materials; may affect non-target species	Dormant season spraying reduces damage to non-target species.
Cutting	Summer, after flowering	Non-toxic; cutting at ground level nearly 100% effective; minimal soil disturbance	Labor intensive; must dispose of stems/flower stalks; use of weed whacker causes collateral damage.	Very effective when done right.
Pulling	Anytime ground is soft and easily worked	Non-toxic; very effective as long as most of the root is removed	Labor intensive; disposal of pulled plants required once flowers are present; soil disturbance	Very effective

It looked like a mild winter until...

We've been having a pretty mild winter this year.  The average temperature in the U.S. in February 2017 was 7.3 F above the 20th century average.  Comparing photos I'd taken in 2014 and 2015 with this year indicates that this year we are 2-4 weeks ahead, based on the blooming of the crocus and forsythia.  On March 14th we finally got a good dose of snow in our neck of the woods (west-central Maryland).

This late snow is out of place with the forsythia that has been blooming since late-February this year.

About a week before the snow storm I took a walk around the woods to see what was starting to come up.  The first thing I checked on was the spicebush.  It usually begins blooming shortly after the forsythia.  This year, while the forsythia had been in bloom for a couple of weeks, the spicebush was just getting started.

March 10th and the flower buds on the spicebush were just opening.

The next plant I checked was the pussy willow, Salix discolor.  This native tree/shrub is one of the earliest blooming native plants and is an important source of pollen to early season native bees.  Since this species is dioecious, only the male plants are sources of pollen; however both male and female flowers have nectar.

The buds of this pussy willow are just opening.  When fully in
bloom the flower buds of this male plant will be covered
with yellow pollen-bearing anthers.

Looking down on the ground in the leaf litter I found a number of Virginia bluebells, Mertensia virginica, that had just come up.  Sometimes the new leaves have a purple tinge to them, but that color quickly fades to green.  The spikes of flower buds follow quickly after this first flush of leaves.

These Virginia bluebells have just come up.

Also showing up on the ground was white avens, Geum canadense, which is pretty common in this area.  While its not particularly beautiful in bloom, it does fill in gaps in the shady understory and its wispy white flowers break up the sea of green leaves.  In the early spring it is by the light colored veins on the deeply divided leaves.

The leaf markings on this white avens rival those of some Heucheras;
however, as it matures the dominant leaves will be smaller and the veination less noticeable.

The plant to the left in this photo is purple deadnettle, Lamium purpureum.  This introduced species is scattered throughout the shady areas.  While weedy, it does not appear to be causing too much trouble with the other plants.

These violets look a little like garlic mustard, ...

Also noted among the fallen leaves were fresh leaves of some native violets, probably woolly blue violet, V. sororia.  These nearly round leaves have finely serrated margins (crenate) and fairly smooth leaves.

They can be distinguished from the over-wintering garlic mustard rosettes that have longer, slender petiole and leaves that are deeply veined and more deeply toothed serrate leaf margins.

Garlic mustard has deeply veined leaves that
look tired, having been out all winter.

Garlic mustard is not the only invasive species that is evident right now.  In fact late winter is a good time to spot some invasive species since they tend to come to life a little before most of the native species.  Multiflora rose and barberry are both leafing out now making them stand out in the woods.  Since the soil was soft I was able to pull a number of these bad guys out of the ground.  This is also a good time to spot Japanese honeysuckle growing up in the trees.  Now is a good time to  cut these climbing vines and deny the roots an early burst of energy.

Most native honeysuckles have the two terminal leaves fused together
like this just below the flower bud.

While not growing with the same vigor as the Japanese honelysuckle at this time, the native coral or trumpet honeysuckle is also leafing out. Flower buds are beginning to form, though the normal bloom season is closer to mid-spring.

Not all plans work out

Three years ago we cleared out a section of garden that was over run by English ivy.  The plan was to put in some new native species as well as watching to see what would sprout up from the soil that had been covered with ivy for a number of years.  To the newly cleared space I put in some goatsbeard (Aruncus dioicus), and a strawberry bush (Euonymus americanus).  As a ground cover I got a bunch of Virginia jumpseed (Persicaria virginiana), which looked like the 'Lance Corporal' cultivar, with reddish chevrons on each leaf. These were followed up the following spring with American alumroot (Heuchera americana), Indian pink (Spigelia marilandica) and Solomon's seal (Polygonatum biflorum).

A jumbled mass.  Blue lobelia is recognizable on the left with spent flower stems of the alumroot.  The glossy leaves of Solomon's seal are evident on the right.  In the center you can make out the red flowers of the jumpseed, but the leaves are barely discernible.

Over the next 3 seasons I let the plantings develop without much interference, except for pulling out Japanese honeysuckle and garlic mustard.  Additional species that cropped up included natives like white avens (Geum canadense) and great blue lobelia (Lobelia siphilitica) as well as introduced species like mock strawberry (Duchesnea indica) and a cultivated species of columbine that had been reseeding itself for a number of years through the ivy.

In September as the flowers of the jumpseed are opening up,
something goes after the leaves, eventually skeletonizing them.

While it was interesting to see 'new' plants coming in, the overall planting was too chaotic to look good as a whole.  One problem with the new scheme was that the goatsbeard gets tattered by mid-summer if it gets too dry.  This can be fixed with a little more care and watering through the season. Also the long flower stems of the alumroot add to the confusion. The biggest disappointment was the jumpseed.  It looks nice early in the season with its striped leaves, but by late summer the leaves are decimated by some insects.  What remains is a jumble of wiry stems going every which-way that detracts from the spikes of red flowers.

After letting this area go for several years and not seeing any improvement I decided it was time to make it a little more formal by adding more defined layers to the garden.  While a random planting works in the distance, close up the human eye is looking for more defined structure and order.  In general, lower plants will go in the foreground, near the stepping stones.  Medium sized perennials go to the middle and the bushy goatsbeard stays in the back.  This new planting will be easier to maintain because it will be easier to identify the 'weedy' plants over a background of low sedges than it was before.

First I removed as much of the jumpseed as I could see.  Then I pulled out any encroaching English ivy and honeysuckle.  I also pulled out some native trees that seeded in, like butternut and box elder.  The lower-growing alumroot was already near the path.  For additional low-growing plants, I transplanted several species of sedges from elsewhere on the property.  Their lower stature, defined form, and contrasting texture will help show off the taller perennials.

Here's the garden after removing the jumpseed.  The back right is dominated by two goatsbeard, and the fencing is getting covered with virgin's bower (Clematis virginiana) that was planted in 2013.

In the foreground is one of the sedges I moved into the garden.
It could be Carex laxiculmis but I won't be sure until it blooms.
Also shown here are Alumroot and white avens,
at top with its deep red autumn color.

Since the Indian pinks don't like to be moved I left them in place and instead transplanted all the columbine to the middle ground.

The Solomon's seal was doing well, but they were getting hidden by the jumpseed.  I put some sedges around the Solomon's seal to give them some vertical space.

One of the sedges I chose has 1/4-1/2" wide blades and  grows in clumps about a foot wide.  These went in along the path and are scattered with the perennials to provide some height variation. Also it's evergreen so it will be around all year.  The white avens that is indigenous to this site has deeply lobed leaves that blend well with the alumroot.  The avens can get lanky and will benefit from some cutting back late in the season.

The other sedge is also clumping but has very thin, light green blades.  I'm pretty sure this is rosey sedge (Carex rosea).  It grows very well on partly sunny to shady sites.

The strawberry bush is known for these brightly colored seed pods in early fall.
Where the branches are touching the ground I am seeing new roots forming.
I'll separate these from the parent plant next season.  This is an easy way to propagate this shrub.

This unknown columbine was planted by a previous owner and
has been reseeding itself for a number of years.  Definitely not a native, but it looks good here.

So here's the garden after clean-up and planting.  It looks a little sad now since leaves are falling off and the plants are settling in for winter.

Weeded and replanted, we'll see what happens next year.

If all the transplants come back and the weeds don't return in force the new garden may look something like this overlay drawing.

As drawn here the goatsbeard, Indian pinks, and columbine are shown in bloom.

The sedges and alumroot line the stepping stones, from lower right to upper left.