Violets and others: Erodium lebelii vs Erodium cicutarium

A summer project to learn about Geraniums and Storksbills ended up with difficulties separating Common Storksbill ( Erodium cicutarium ) from Sticky Storksbill ( Erodium lebelii ). The nearest place to Cambridge where E. lebelii can be seen is Landguard Point , Felixstowe in Suffolk, so off I went to the coast.

Erodium cicutarum is known to be variable in flower colour, size and amount of glandular hairs so the starting point is what an 'normal' plant looks like.

Common Storksbill with bright pink flowers

1) Common Storksbill (Erodium cicutauium) ( inland plant)

Common Storksbill leaf with non-glandular hairs.

The sepals are pale with green strips. Occasionally plants have been found with very short sepal tips.

Short sepal tip 24th June 2019 Science Park, Cambridge

Next example of variation is some plants have white flowers.

Note almost white stigma, 24th June 2019, Science Park.

Hairs can be very useful in the identification of Geranium and Storksbills and are normally consistent, however the Common Storksbill is variably hairy. The flower stem can have flattened hairs , short and long upright non-glandular hairs plus upright glandular hairs. Glandular hairs are quite common on the flower stems and are often the dominant type . Plants with glandular hairs on the stems, often have white slightly curved non-glandular hairs on their leaves.

E. cicutarium leaf showing hairs, 18th June 2019, Devils Ditch.

2) Coastal variations of Erodium cicutarium.

The first impression of the Storksbills at Landguard was there were very few plants that looked like the normal inland deep pink flowered type. Most had smaller flowers of a variable paler colour and some white flowered plants, were also present.
The warning first sentence from the Hybrid Flora came into mind ' Small-flowered densely glandular maritime variants of E. cicutarium are often mis-determined as E.lebelii, but the only reliable distinction lies in the mericarp morphology'. This was backed up by the statement in Harrap's Wild Flowers, ' Identification must be confirmed by examination of the fruits'. Mericarp is the seed case which is attached to a long beak. Five beaks make up the storksbill.

E. cicutarium Mericarp seed case with hairs enclosing the dark brown seed.

The mericarp, a hairy case containing the seed, attached to the beak.

The critical feature is the shape of the pit where the mericarp attaches to the beak and whether the hairs over-reach the pit.

Following the Harrap's Wild Flowers advice I went looking for plants with more densely glandular hairs ( often picking up sand) and more greyish-green leaves. Flower colour was not going to be much use as it noted that colour could be pale pink to white. Then I would confirm ID by looking at the fruit.

Leaf Shape Variation, 26th June 2019 Landguard

On the way I did find an interesting variation in the leaf shape but neither had any glandular hairs so thy must be both be E. cicutarium. Hairs were the normal curved white non-glandular type and the leaflets were cut to about 2/3rds. The version on the right is not typical being more rounded leaflet tips.

Leaf Detail showing dense white hairs

Another variation were some plants having dense wavy white hairs which might make the leaves look grey at a distance. The hairs were mostly non-glandular and the seed showed these plants were E. cicutarium. Note glandular hairs on the leaf stem. These leaves had rounded tips.

Hairy plant with seed in bottom right corner

The leaf shape in this version is more rounded without the more pointed tips. The seed pit shows a distinct rim and the hairs do not overreach the pit making this plant a E. cicutarium.

Easy Plant.

Once the ID had been established by inspection of the mericarp , I noticed that all the deep pink flowered examples were found to be E. cicutarium. It appears that at Landguard the deeper pink flowered plants are always Common Storksbill. This was backed up by the leaves not being covered in glandular hairs.

E. cicutarium , pink flower with pink anthers and 5 pointed dark pink stigma in the middle.

Note the flower stem and sepals are covered with glandular hair.

Paler flowered plant.

Most plants at Landguard were a very pale pink with shades of blue/lilac. Flowers were small being typically only 8mm in diameter.

Photo of flower

E. cicutarium

The paler flowers tend to have paler stigma, in this case a pale yellow and the same as the E. lebelii ( see below). The flowers of both species seemed to be indistinguishable.

E. cicutarium

The colour of these photos is not quite correct missing out a slight lilac blue tinge which was shared with the E.lebelii.

Photo of leaf

E. cicutarium leaf with 99% hairs being glandular

This particular leaf is a typical shape for Common Storksbill with quite pointed tips to the leaflets.
The cuts being 2/3rds deep, a standard that both Common and Sticky Storksbill conform to. The glandular hairs are erect and typically 0.2mm long.
Some leaves on this plant had no glandular hairs, possibly older leaves can lose the glandular tips through evaporation, if this were true one would think some trace of the glandular sac would be left, so it would appear that some plants have different amounts of glandular hairs on the leaves.
The main point is that Common Storksbill can have very hairy glandular leaves (which will pick up sand) although this is unusual. Sticky Storksbills have very glandular hairs which leave your fingers feeling damp after touching the leaves, whereas even the glandular hairy Common Storksbill does not quite have that effect. More investigation required as it could be that the amount of stickiness does seem to be a factor. It has also been seen that the leaf shape can be more rounded so it seems likely that leave shape may not be a reliable way of separating these two very similar species.

E. cicuturium highly glandular leaf.

Photo of mericarp (seed case)

Mericarp of E. cicutarium

This photo shows a pit with a definite rim and beyond the rim there is a clear grove. In practice the groove is a variable feature of E. cicutarium and some literature suggests the groove is not present on the coastal form. I found it was present in most of the coastal E.cicutarium at Landguard.
Hairs over-reaching the pit, a key feature which distinguishes Common from Sticky.
The hairs do not overreach the pit in this photo above . Unfortunately not all plants show the key feature quite as clearly, with some plants thought to be E.cicutarium based on non-glandular leaf hairs having mericarp hairs that over-reached the pit by up to 25% of the pit diameter.

If Mericarp features are not clearly typical of either species, it is best to look at another mericarp from a different stalk on that plant.

Another photo of mericarps .

E. cicutarium

Not quite so easy to see the rim but just possible to make out the groove. More mag needed.

Same picture magnified up to show rim and grove but note that the hairs do over-reach the pit slightly est. 20% of diameter . Problem is that this level of detail is not easy in the field and you really need a X20 lens, more power than most people carry.

E. cicutarium

3) Sticky Storksbill ( E. lebelii )

Photo of flower showing pale off white (slight yellow tinge) stigma and anthers with pale pink-orange colour. Stigma colour is the same as the coastal E. cicutarium whereas the deeper pink flowers can have a darker pink stigma. The anther here is the anther case which splits to reveal the yellow pollen. The whole anther structure then falls off and I assume the flower continues on for several days waiting for an insect to bring pollen from another flower.

E lebelii, Landguard

Photo of leaf

E lebelii leaf showing glandular hairs.

The leaves on the E. lebelii tended to be small and rounded however the coastal form of E. cicutarium can have rounded leaves and also can have glandular hairs so I can't say there is a difference except that E. lebelii tends to have more stickiness.

E. lebelii

Clapham.Tutin.Warburg mentions the the leaves of E lebelii are almost bipinnate, so I though I would look at leaf shape. Bipinnate implies the leaflets are on a separate stalk to join the leaf. This is case for the lower leaflets but not the ones nearest the tip of the leaf. This arrangement is the same in E. cicutarium. Vein pattern is limited to a single vein in most leaflets although a vein split can be seen in the top right corner and this is probably a function of leaf size. See below for a leaflet of 100mm long for Common Storksbill , three times larger than the leaflet for E. Lebelii. shown above.

E. cicutarium leaflet showing vein split off each lobe.

Photos of Mericarp (seeds)

Small pit, no groove and overreaching hairs 50%, E. lebelii

Hard to make out rim detail, hairs overreaching pit just over 50%. E. lebelii

Hairs overreaching pit by 80%.

E. lebelii Features like hair overreach visible at green stage.

E.lebelii Mericarp hair over-reach just over 50% across pit.

E. lebelii mericarp 4.7mm long and with hair overlap 50% 26Aug2019

4) Other features from the literature.
a) Mericarp length 5mm and less for Sticky, 5-6.5mm for Common. A limited sample measured were

Sticky Common (coastal form)
4.5mm 5.1mm
4.8 5.1
4.9 4.8 ( with 20mm beak)
4.8 5
4.0 5 (with a 32mm beak)
4.5 5

Almost follows the rules so worth checking this feature but see apppendix below for more data which follows a visit on 28th August 2019.

b) The Wild Flower Key ( Rose) suggest flowers in groups of 3-7 for Common, 2-4 for Sticky but in practice at Landguard the Common had either 2 or 3 flowers coming from a stem ( typically about a 50/50 split ) and Sticky had either 1 or 2. One Sticky plant had only single flowers but typically plants had an even split between 1 and 2. Whether this pattern is limited to this single site is not known but it dos not follow the literature.

c) Beak length 15-40 mm for Common, beak up to 22mm for Sticky.
Beak length seems highly variable. The Sticky plants I saw had beaks between 10 and 20mm but overlap with the coastal Common 15-35mm makes this feature of limited use.

Conclusion.

The identification of E. lebelii from the coastal form of E. cicutarium is very difficult. The coastal form is so very different from the normal inland version of the Common Storksbill that you can incorrectly assume it is E. lebelii. My first visit to Landguard resulted in total confusion as I was surprised how similar the Sticky and coastal form of Common were. The confusion was increased by the fact that the hair overreach of the Mericarp pit, as a distinguishing feature, is not as clear cut as you might think.
A second visit was made and care was taken to really check out the mericarp detail before proceeding to take photos of other features. This was helped by finding an area which only had plants of E. lebelii , before moving onto areas with both types present.

With practice the mericarp detail does provide a good way of separating these very similar species just as the books suggest. The hair over-reach in E. lebelii was always at least half way across the pit whereas E. cicutarium was never more than 25% across and often the hairs did not over-reach at all.
There are however exceptions to this attempt at a rule, see appendix below.
The pit rim can be very hard to see but you just have to look at more mericarps on each plant to make a determination. Many storksbill seeds are not viable having empty mericarps, so you have to find good non-sterile ones to check details like mericarp length.

It is very dangerous to make any generalisations based on a single site, so I would welcome any feedback from experts familiar with other populations which may be different from those at Landguard. The situation at Landguard makes identification very difficult and I wonder if this is true of other sites?

The name of Sticky Storksbill has recently been changed from E. lebelii to E. aethiopicum previously also being know as E. glutinosum in older books. There are records of hybrids from Wales and South Lancashire but no plants were found at Landguard that fitted the description of the hybrid, which is possibly surprising since both grow together.
The coastal form of E. cicutarium is know as subs. dunense. but intermediates occur so it not really given any special status.

Basic method for finding a Sticky Storksbill.
1) Find a plant with small leaves that are covered in glandular hairs. Those without glandular hairs on leaf will be E. cicutarium. Note that some stressed plants can lack glandular hairs and this applies to both species. Identification is so difficult I would bypass stressed plants.

2) Find brown storksbill with good mericarps still attached. Using X15 mag lens examine the pit at the end of the mericarp attached to the beak. It will help to pull off the sepals to get a better view.
Check hair over-reach of the pit. Greater than 50% is good for E.Lebelii. Some E cicutarium have hairs that over-reach up to 25% of diameter of pit. but many will have no over-reach at all. Check presence of a rim to the pit. this can be very hard to judge. E lebelii should not have any sign of a raised rim. E. cicutarium often has a clear rim and a groove below it.

3) If the mericarp features are not conclusive find another brown mericarp on that plant and check that out, as it may have easier to see features. I find that the Olympus TG5 camera using photo stacking micro mode is almost as good as having a microscope in the field and that gives more mag that my x15 lens. Mericarp features vary across a single plant so taking a sample of 2-3 is good, as on average a determination can be made.

Additional note.

The colour of the anthers and stigma are variable but cannot be used to distinguish between the two species as far as I can tell. In general the stigma on E. cicutarium is normally pink, sometimes quite a dark pink, but white flowered and coastal whitish variants have pale almost white stigma.
The stigma in E. lebelii are pale white with a hint of green or yellow, the same as in white flowered Common Storksbill.

The anther colour for E. cicutarium is normally a dark purple/blue before it splits to reveal its pollen.
In pale flowered versions it can also be a pale pink and colour it shares with E. lebelii.

White flowered variant of E. cicutarium with pink anthers.

Habitat.

Both species present on sandy shingle areas near Landguard Fort

Landguard Fort, July 2019

Dry grassland with mostly E. cicutarium

Appendix. Update from visit on 28th Aug.

More measurements of mericarp length and hair overlap.

Since the uncertainty in hair overlap is not as clear cut as I had hoped a second definite feature would be very useful, however since neither flowers or leaf have provided this, the length of the mericarp was looking like a good possibility.
More measurements of mericaps were taken on the 28th August 2019 and the results follow.

Common ( E. cicutarium ) 6 Mericarps measured from three plants.

Mericarp length Rim Groove Hair overlap
4.7 Yes Yes 10%

4.8 Slight No 20%

From single storksbill plant to show variation.
4.8 Yes Yes 50%

50% overlap left with rim and groove .

5.0 Slight Slight 40%

5.0 Yes Yes 30%

4.8 Yes Yes 30%

The concern here is that in extreme cases the hair overlap can be as high as 50% of the diameter of the pit making this feature not totally reliable. In addition the length has quite a proportion of overlap between species. The presence of a rim noted as 'slight' was seen using a X40 microscope and side lighting.

Sticky Storksbill ( E. lebelii ) 7 mericarps measured from 4 plants

Length Rim Groove Hairs
4.3 No No 80%

4.4 No No 80%

3.5 No No 80%

4.2 No No 60%

Note The beak ridge goes part way round the pit but this is not a rim.

4.3 No No 20% approx as hairs did not point across pit.

4.1 No No 70%

4.1 Slight No 100%

Quite a small sample, so not surprising that the lengths are on average less than previous sample where 4-4.9mm was the range . The good news is that despite the overlap, on average there is a difference between these two species. The E. lebelii had less than 4.4mm mericarp length and all had long hairs over the pit although since they don't always point in the same direction some interpretation on the percentage across the pit is needed. This includes the view point approximately at right angles to the mericarp axis and the limit of the pit. If all features are considered it still remains possible to identify most plants based on small sticky glandular leaves and mericarp features and it now looks as if the presence of the grove and rim is more reliable than the hair overlap .

Summary of both days results.
Mericarp length
Sticky 3.5mm - 4.9mm
Common 4.7mm -5.1mm ( coastal form at Landguard )

Version 1.3 Updated 29th Aug 2019
Peter Leonard
Rampton
Cambridgeshire
1st August 2019
cambridgewildflowers.blogspot.com

POSTCRIPT.
Following putting up a display poster at the BSBI annual meeting in November 2019, Tim Rich kindly sent me the article by P.M. Benoit from the 'Proceedings of the BSBI' May 1967. This article gives descriptions of both Common , the hybrid and Sticky but these descriptions do not tie up well with what is seen at Landguard. The separation of these two species was based on work done by Andreas ( 1947) who showed that cicutarium was tetraploid 2n=40 whereas glutinous ( leblii ) is diploid 2n=20. The triploid hybrid 2n=(30) is not uncommon where both species co-exist and the article suggests the hybrid is full or largely sterile and has large flowers being a F1.

Of more concern is that the variability of both species is high and that Sticky Storksbill does not always have dense glandular hairs or small flowers. Larsen ( 1958) described a third , hexaploid cytotype as E. danicum which has not been found the the British Isles.

Unfortunately counting chromosomes is beyond my capabilities so giving a definitive identification of plants found at Landguard is not possible.
One other thing, these species are regarded as annual but I have seen at least one plant live for three years. The seeds germinate almost as soon as they reach the ground and the tiny plants grow slowly through the winter months to develop the following spring. Most die off in the summer drought or in the early autumn but if the plant survives this it can make it through the following winter. These long lived plants can occasionally flower right through the winter. These plants always seem to be Common Storksbills with pink flowers and never the pale type (coastal/sticky).