Grasses are tough

Grasses are tough, not only do they grow just about everywhere, they can be very difficult to identify.  Unlike so many flowering forbs, grasses tend to all look about the same to the casual observer.  This situation is reinforced by our tendency to mow grasses down on a regular basis so that all we see are a collection of short green leaves.  In the US there are 169 genera and 1,398 species of grasses.  In the Northeast there are roughly 600 species growing in the wild.  Over the past couple of years I’ve making an effort to identify grasses growing on and around our property here in central Maryland.   

The first level of grass identification is to determine whether you are looking at a true grass (Poaceae) rather than a sedge (Cyperaceae) or rush (Juncacaeae).  You can make a quick distinction using the rhyme:  Sedges have edges, rushes are round, and grasses have knees (joints) and bend to the ground.  Sedges have triangular stems; when observed end-on the leaves will project out 120 deg apart.  The stems of rushes are round.  The stems of grasses have joints or knobs where each leaf blade originates.  In addition grass stems are hollow while those of sedges and rushes are solid.

While there are a few grasses that can be identified with some reliability on sight, most require close examination of the flowers and leaf surfaces with a hand lens to determine the grass down to the species.  Some easily identifiable species include the very popular little bluestem, Schizachyrium scoparium, and invasive species like Japanese stiltgrass, Microstegium vimineum, and Johnson grass, Sorghum halpense.  But for the most part you need to do a close examination to really determine a grass down to the species level.  There are a number of dichotomous keys for grass identification down to the species level.  Some rely heavily on flower (spikelet) characteristics, while others include other characteristics.  These keys ask a sequence of yes-no questions about the plants to slowly lead you to a result that matches the characteristics of the plant in question. 

Little bluestem is particularly recognizable in fall when the stems turn bronze
and the fuzzy seeds catch the light.  In summer the stems have a bluish green cast.

Johnson grass is a non-native weedy species that is often seen
 on the edges of fields and drainage areas.  It is tall 3-7'
and kind of looks like skinny corn stalks.  

Japanese stiltgrass in an invasive grass species, particularly in the
eastern United States.  It can be distinguished by the silvery midrib
on the elliptical leaves and its lax habit with many joints along the stems.

The Native Plant Trust (formerly the New England Wildflower Society) has developed an on-line key called GoBotany that allows one to enter a variety of plant characteristics simultaneously and provide a best fit.  This is particularly useful if you are trying to do an identification at a time when no blooms are available.  The plants in this database include grasses, forbs and woody species, but are limited to those found in the New England states. 

 To do a grass identification you’ll need an identification key and a hand lens (10x).  A camera is also useful to document your observations.  Since I made a commitment to ID some of my grasses I’ve been gathering a few different resources.  The Manual of Vascular Plants of Northeastern United States and Adjacent Canada, by Gleason and Conquist is a very technical key (in small print) that, while challenging to use, does provide detailed descriptions of each species. Grasses an Identification Guide by Lauren Brown is easier to use and well illustrated.  It also includes some common sedges and rushes in addition to grass species.  Its drawback is that it describes only the more common grasses.  Both of these references contain species found in the northeastern quarter of the US.

One resource I haven’t used much (yet) is Manual of the Grasses of the United States by A. S. Hitchcock.  This two volume set describes all of the grasses found growing in the US.  It does include some illustrations which makes it somewhat friendlier than Gleason and Conquist.  I’ve been intimidated by the size of this collection (especially large considering this includes only grasses, no rushes or sedges).

Here in the Mid-Atlantic region an easier to use key is the Field Guide to Grasses of the Mid-Atlantic by Sarah Chamberlain.  This guide includes botanical sketches of plants, and plant parts, to aid in the identification and includes only true grasses that are found in the Mid-Atlantic States (VA to PA).  So you don’t wade through extra species that are not likely to be present.  It includes many more grass species than Lauren Brown’s book, though it lacks the sedges and rushes. 

An additional book that I have found critically useful in all plant identifications is Plant Identification Terminology by Harris and Harris.  It provides illustrations and definitions of all the plant parts that are described in botanical Latin terms in the keys. 

Going through a dichotomous key can be time consuming, especially early on as you are wrestling with botanical Latin.  A drawback of many keys is that it is difficult to go backwards when you get something wrong, i.e., you get to the end of the key and your sample is nothing like the choices in the key.  Some of the more formal keys are easier to back up on when each step is numbered.  I’ve found with some of the book formats that they do not number the steps (choices) so that going backwards involves leafing through a lot of pages.  Perhaps putting some sort of bookmarks, like post-its, in the book would make it easier to retrace my steps. 

When I do go through a key and end up without a match I will go back to a point where I was not absolutely certain about my choice and then run down that new path.  Sometimes you just don’t have a good example of the feature that they are asking about; like size or shape of a seed, the number of veins on a lemma, or a surface is slightly or very hairy. 

Here you can see the individual florets of switchgrass.  This species has
only one floret per spikelet. Highlighted here are the glumes,
bracts that enclose the floret(s).  One of the florets is in full bloom
 with the anthers and stigma projecting out of the floret.

After getting frustrated about not getting satisfactory results I found it helpful to practice on some known species.  That way you can get some confidence with identifying botanical structures.  I had examples of switchgrass, Panicum virgatum, and purple top, Tridens flavus, that I could practice on.  After that I worked on an unknown grass that was a little past bloom, the flowers were a little tattered but mostly recognizable.  I found that one of my problems is counting florets in dried flowers.  Sometimes they just look like a bunch of chaff.  With a little practice I’m getting better at recognizing the flower parts.  I was able to ID the unknown as the relatively common, though non-native, smooth brome, Bromus intermis.

Here's the flower stalk for purple top.  The spikelets, which contain
4-9 florets, are deep purple in color.  Hence the common name.

One thing that surprised me in keying out grasses is that plants that are close together in a key may not look that much alike when viewed at a distance.  That is because many keys focus in on differences that are not easily seen rather than on obvious features like height or size of leaf blades.  While not part of some keys, knowing things like bloom season or habitat can be very useful in confirming an identification. 

In addition to running through keys, working with people who are familiar with grasses can be of great help.  For many, once you know and recognize a plant you don’t need to run through a key to identify it, they just know.  In this time of COVID-19 it is even more difficult to find and work closely with a grass expert.  I did recently find a Facebook group focused on identifying grasses called Grasses,Sedges and Rushes of the Northeastern USA.  They have been helpful in pointing me in the right direction when I have been adrift. 

The American mannagrass grows vigorously in early spring
 and is in full bloom by the end of June.

This was the case when I thought I had ID’ed a grass as drooping woodreed, Cinna latifolia.  This was thrown in doubt when I saw that the grass was going dormant in mid-summer while Cinna normally blooms in mid-summer.  Based on a lead I got from the Facebook group, I redid the ID using a different key and came up with American mannagrass, Glyceria grandis.  On reviewing the keys it appeared that I had totally messed up characterizing the flower (spikelet) parts by overlooking some of the less developed florets.  On a side note, it was interesting to see that areas that were thick with mannagrass had very little Japanese stiltgrass filling in in late summer.

 

July 2020 What to do about Japanese Stiltgrass NOW

In late July through early August Japanese stiltgrass, Mircostegium vimineum, puts on a major growth spurt prior to blooming.  This presents an opportunity for removing a large amount of this invasive annual grass before it blooms and begins to set seed for the following years.  There are several methods that can be used now, which is best depends on the particular situation. 

 

Pulling  This is my method of choice in areas of mixed species. In late summer stiltgrass can grow to 3-4 feet tall.  At that height it is above many desirable plants and you don’t need to bend over too far to pull it out.  By using a loose grip and grabbing higher up on the plant mass I can selectively pull out the stiltgass and leave most perennial and more deeply rooted plants in place. By wiggling the grass side to side as I’m pulling the relatively weak roots are broken and the stems remain intact.  

Here stiltgrass and the native grass nimblewill, Muhlenbergia schreberi, are growing together.

Pulling the stiltgrass with a light touch leaves the well rooted nimblewill in place.

I usually leave the piles of stiltgrass to dry out in the sun for a couple of days before disposing of it.  If the grass is not setting seed I’ll put it in a brush pile or in a segregated location where it can break down but will not accidentally be spread elsewhere.  If the grass has bloomed it will need to be bagged or put into an isolated location where the seed cannot escape.  This is the number one reason for pulling before bloom!

This is a woodland area where I have been pulling stiltgrass of 2-3 years. 
Though it looks dense there are not that many individual plants.

 

After 20 minutes of pulling I was able to clear this area of stiltgrass. 
What remains is a ground cover of mostly rosy sedge, Carex rosea.

Low Mowing  Mowing as close to the ground as possible in late summer (just before bloom) is a common recommendation for combating stiltgrass.  This prevents blooming and seed set in the upper stems.  Unfortunately stiltgass will also set seed in the stem at the base of the plant, 1-2 inches above ground level.  These cleistigamous flowers are difficult to remove without cutting really close to the ground.   If you wait too long to mow, after seeds begin to develop, mowing may only serve to spread seed unless you have a well fitting grass catcher on your mower.  ( In that case you should dispose to the clippings in a way that the seed will not escape into the environment.)  These extra flowers make stiltgrass very difficult to remove from turf areas.

Late summer mowing had been used in this area for 2-3 years.  Here it is in 2017. 
There is a lower density of stiltgrass, but still a lot.

I’ve been trying to remove stiltgrass solely by timed mowing in one area but have made only minor progress over 3-4 years.  What I have found to be most effective in lawns is to use a pre-emergent herbicide, marketed for crabgrass, in early spring.

Here is the area in 2020.  In addition to late mowing I treated this area with the
preemergent herbicide Dimension.  Above the line has been treated for
two consecutive years, below the line only once in spring 2020.

 

Weed whacking  Using a weed whacker can be effective where you can selectively cut down to ground level to remove those stem flowers. It can get tedious if you are trying to preserve desirable plants intermixed with stiltgrass.  You can leave the cut stems in place if the grass has not begun to bloom, but if it has you should consider raking out the debris.  (If plants are beginning to set seed weed whacking will likely spread more seed than doing nothing.)

I had good success in a woodland edge area using this method.  Starting out, the area was nearly a monoculture of stiltgrass so I did not have to be too careful.  In the second and third years I also raked out the cut stems.  I have also seeded the area with Virginia wild rye, Elymus virginicus, a shade tolerant native grass.  Now six years later I can manage that area with just a little pulling. 

 

Fire  Because stiltgrass is a weakly rooted annual burning at ground level can be an effective means for killing it.  I have been experimenting with burning stiltgrass at various times during the year.  Burning in early and mid spring reduces the amount of stiltgrass seedlings but does not eliminate its presence.  This is in part due to the extended time over which stiltgrass can germinate in the spring.  Burning in late spring seems to thin out the amount of stiltgrass but can also reduce the amount of desirable vegetation.  In one spot where I did a mid-spring burn the area was overgrown with stiltgrass by the end of July.  On removing that stiltgrass there were very few other plants growing there. 

This is a small area of lawn dense with stiltgrass that I burned with a garden torch
two days ago.  It's not necessary to consume the entire green plant with fire,
just burn the roots at the soil surface.  Perennial grasses should
bounce back in a week or two.

In this area I pulled most of the stiltgrass then used the torch to burn the surface
to kill any remaining vegetation.  I was then able to immediately plant several
plugs of switch grass, Panicum virgatum.

Where burning has proved very effective is in hard to reach areas where clear ground is desired, such as pathways and fencelines.  Burning stiltgrass is particularly effective in late summer. In lawns with cool season grasses you can burn areas infested with stiltgrass with a garden torch.  The cool season grasses are dormant because of the hot weather and dry conditions.  As the weather cools in September the clumps of cool season perennial grasses will resprout, without the presence of stiltgrass.

Here is the garden torch in action, burning the roots of some 3 foot tall stiltgrass. 
I can stick the torch through the openings in the fence the get to hard to reach areas.

It is interesting to note that heat is conducted better in slightly moist soil than in dry. While burning stiltgrass leaves no chemical residues you must use great caution to keep flames under control and not allow any fire to spread out of control.  There may also be local rules to follow.  

Herbicide   While I have had very good results from using preemergent herbicides in early spring to get stiltgrass out of the lawn, I have not had personal experience using postemergence herbicides to control Japanese stiltgrass.  Based on information provided by a number of university agricultural extensions there are several herbicides that can be used in summer time (up to the onset of flowering)  to control stiltgrass.  Low levels of the broad spectrum herbicide glyphosate, about half the normal concentration, are reported to be effective against Japanese stiltgrass.  Glufossinate is another broad spectrum herbicide that is effective against stiltgrass. 

I pulled this from the Rutgers Cooperative Extension (NJ) site:

Glyphosate and Glufossinate (various trade names) can be used to spot treat Japanese stiltgrass in gardens and planting beds. Both are broad spectrum herbicides that should be applied only to the unwanted plants. If applied to the foliage, stems, or woody portions of desirable plants, it could damage them as well.

Sethoxydim (tradename Bonide Grass Beater Over-the-Top Grass Killer®) and Fluazifop-P-Butyl (Ortho Grass B Gon Garden Grass Killer®) are selective herbicides that can be applied to growing stiltgrass in landscape beds. When used according to the label, these herbicides will not damage most non-grass ornamental plants. Be sure to follow the label closely and heed all precautions.

According to the labels on the Bionide and Ortho products these products are not effective on sedges.  That is a good feature for use in woodland edge habitats since native sedges are a major component of the ground covers growing there.  I not sure if that means that those, mostly desirable species would be unharmed, or, just not killed at a high rate.  I noticed that the label on the Bionide product indicated that it could be used safely on lawn with red and Chewings fescue.

Here is another example of the effectiveness of a preemergent herbicide. 
To the left of the line I applied dithiopyr (Dimension) in mid-March,
to the right is untreated.