Viola pilosa or Viola canescens ABApril_2020/01

[Ashwini Bhatia]

This is the second species from Dharamshala which is still confusing. The flowers can be from pure white to deep mauve and vary in size from 1.5 to 2.2cm. Most are wider than long. 

The petals are of different shapes and sizes. The two uppermost are the broadest at ca. 7mm, the laterals are roughly 5mm wide (both about 13mm long). The lowest petal (or the uppermost if you consider resupination, but for clarity I am using the lowest) is the narrowest at 4mm and excluding the spur about 10mm long. The spur is cylindrical, white, and can be hooked or not and is 3mm long. 

The leaves are slightly paler underneath, hairy on both surfaces with scattered white hairs. The petiole is hairy and the stipules are laciniate (with long hair-like projections on both long edges). 

The pedicel projects beyond the leaves mostly and is covered on white hairs too. Sepals are lanceolate, equal and have hairy margins. 

The anthers are pale yellow, two with nectaries projecting into the spur. The nectaries are yellow-green, hairy and about 2mm long. 

The ovary is hairy at the top. The stigma appears truncated at lower magnification but up-close reveals a crater like perforation at the top of the club-shaped style. 

The seeds are yellow with dark purple warts on their surface. Elaiosomes are prominent.

Depending on how we interpret this data, we could either believe this to be Viola pilosa or Viola canescens. All comments are welcome.

Thanks.

Ashwini

All photos taken between 1750 and 2200m in Dharamshala, Himachal Pradesh. the flowering season is usually late February to June. A few plants can be found well into autumn too.

[J.M. Garg]

I will go with Viola pilosa.

--
With regards,
J. M. Garg

[J.M. Garg]

Sorry, I wanted to say Viola canescens

--
With regards,
J. M. Garg

[Tabish]

This type of Viola, with all sepals equal and with ciliate margins, I would think it is Viola pilosa.

  Tabish

-------------------------------------------

www.flowersofindia.net
The waterhole of flower lovers

[J.M. Garg]

Pl. see Viola canescens. I think style is the best clue. Here style is club shaped. Leaves are also not acuminate.

As per keys in Flora of Pakistan:

13 (12)	Stigma beaked. Ovary glabrous		(14)
+	Stigma club shaped. Ovary hairy		2 Viola canescens
14 (13)	Leaves acuminate. Sepals lanceolate, acute, ciliate-dentate		3 Viola pilosa
+	Leaves obtuse. Sepals ovate, obtuse, entire		1 Viola odorata

--

With regards,
J.M.Garg

'Creating awareness of Indian Flora & Fauna'

Winner of Wipro-NFS Sparrow Awards 2014 for efloraofindia. 

For identification, learning, discussion & documentation of Indian Flora, please visit/ join our Efloraofindia Google e-group (largest in the world- more than 3,000 members & 3,00,000 messages on 23.8.18) or Efloraofindia website (with a species database of more than 13,000 species & 3,00,000 images of which more than 2,50,000 images are directly displayed on 31.1.20).

The whole world uses my Image Resource of more than a thousand species & eight thousand images of Birds, Butterflies, Plants etc. (arranged alphabetically & place-wise). You can also use them for free as per Creative Commons license attached with each image.

Also author of 'A Photoguide to the Birds of Kolkata & Common Birds of India'.

[J.M. Garg]

As both species are quite confusing, these key features are not visible in images of Viola canescens at FOI.

I have doubts about correctness of these images in FOI.

[Tabish]

This is from FOC description of Viola pilosa:

"styles clavate, base slightly geniculate, gradually thickened upward; stigmas ± flat, not margined, very inconspicuously short beaked in front, with smaller stigma hole at tip of beak."

  best wishes

  Tabish

[Ashwini Bhatia]

Thank you Mr. Garg and Tabish ji for your comments. We can all appreciate the difficulties involved. 

The issue boils down to which source(s) we trust the most in our identifications. Especially when the sources we usually refer to do not agree with each other.

When I began looking at the two viola species in Dharamshala in 2015, the first characteristic I looked at was the stipules. I was told that the fringed ones meant canescens and toothed ones pilosa. Now I have learnt that these can be variable. FOC describes them for V. Pilosa as "stipules mostly free, brown or green, lanceolate, margin long or shortly fimbriate-dentate, apex long acuminate”. 

Flora of China does not include V. canescens in its list of viola species in China but makes a brief comment under V. pilosa description "In FRPS (51: 90. 1991), the name Viola canescens Wallich was misapplied to this species.” FRPS is Latin for Flora of China. This is not elaborated further so we will not know if the author(s) does not approve of V. canescens as a valid species or is merely suggesting that these features were earlier confused with V. canescens. 

Next, I learnt that to know violas, one has to look at the style/stigma. So I did my best to look at the style closely. I found out that the style is club-shaped, gradually thickening upwards culminating in a perforated stigma placed on the side. 

This matches the description given by FOC for V. pilosa. But Flora of Pakistan gives the stigma for V. canescens as club-shaped but does not elaborate more. 

FOP does not give the shape of the stigma for V. pilosa but tells us that it is beaked. The perforation-like stigma, can be seen like a beak from an angle but I am not sure if I want to base my identification on a feature that is open to interpretation. 

Flora Simlensis says that V. canescens has a truncated stigma and not beaked, and describes stigma for V. serpens (synonym of V. pilosa) as being three-lobed and beaked. The stigma on our plant may look truncated to the naked eye but under a modest magnification, the shape becomes obvious.

Flowers of the Himalaya lists V. pilosa as having a 3-lobed stigma with a beak too but does not comment on the stigma of V. canescens.

So neither the stipules nor the style/stigma shapes are agreed upon in the literature. FOC pdf was compiled in 2007 and could be the most recent study we have on the genus in China/Asia. I am not sure how often the Flora of Pakistan is updated but the books I have consulted are all much older than 2007. But since the study does not include V. canescens at all, we are still left wondering.

Our species matches the descriptions broadly for V. pilosa in FOC and V. canescens in FOP. How do we move forward?

Thank you and regards,

Ashwini

[J.M. Garg]

Attaching a plate provided by Santosh Agarwal ji in Viola confusa?/ABMAR21

Style is more closer to  Viola canescens rather than Viola pilosa

I feel we have to rely more on  Flora of Pakistan as both the species are listed here (and found in the area) and this issue must have been deliberated for long times and keys framed accordingly due to confusing nature between the two.

Leaves are clearly acuminate in Viola pilosa and not so in Viola canescens

There is a clear difference as to how style is visible in the live flowers in both species at Viola canescens and Viola pilosa

One more point is the nature of the lower petal being mostly acute or acuminate in Viola pilosa and obtuse in Viola canescens 

I think the above three point combined together should clinch the id in most of the cases.

[Ashwini Bhatia]

Dear Mr. Garg,

I agree with you that if we are certain about the flowers on our website here being Viola pilosa, then as certainly the flowers from Dharamshala are not V. pilosa. As the images below show, they have a three-lobed stigma and the pointed lip.

https://08511630493324166816.googlegroups.com/attach/18709915a1a38/_DSC0732%20-%20Copy.JPG?part=0.3&view=1&vt=ANaJVrHtj7ZHsf69djLbP1FCOPxM2mjOS-k4-1BGLdC92xV0dz2p5eiPZpbhRnyxjj5OGCfw1gPFLC5YUga35lktvq0Cs9QL9BeT_Y4y4YwJqF-QdnCDEBw

https://08511630493324166816.googlegroups.com/attach/2f0eca70ef548d1e/ID.jpg?part=0.1&view=1&vt=ANaJVrHOJ5wqfYQZIWIRiwBTCrUsRrn21uwlgV_vAwPX0ZUnlZC05Ue92Ojidg7HhNiOnPsPuZ8hx9EuP4n30fgYYnDOuATzm_TJgcd8alVmOV54DAtrrCI

These should be the defining characters then. But that will also prove that the illustration provided by Santosh Agarwal ji is not reliable and neither is the description in FOP or FOC for that matter. The older literature such as Flora Simlensis and Flowers of the Himalaya both win here.

Tabish ji has already included a 3-lobed stigma in his description on FOI, which stand correct.

So if FOC, FOP and the illustration of styles cannot be relied upon, how will we file flowers from Dharamshala under V. canescens?

Thank you.

Ashwini

[Ashwini Bhatia]

Sending here a combined image for reference and comparison.

Thanks.

Ashwini

[J.M. Garg]

Hi, Ashwini ji.

I think style matches with the illustration. Ovary is also hairy.

If we may examine hundreds of specimens, we may find little variation here and there in most of the characters.

All hundreds of specimens can not be put in an image, only a representative images can be made. Same is for the description.

As far as style of Viola pilosa is concerned, we have not seen the close up as you have shown for the other species.

[Ashwini Bhatia]

Thank you Mr. Garg. We can conclude for the time being that the flowers here are V. canescens.

Warm regards,

Ashwini

[J.M. Garg]

Thanks a lot, Ashwini ji.

We are always open and change our ids and pages when we ourselves find new evidence.

[J.M. Garg]

In fact all this new learning has been because of your persistence and super follow up presentations.

[Ashwini Bhatia]

Thank you Mr. Garg. You are too kind.

Regards,

Ashwini