Commelina caroliniana ?

Commelina caroliniana Walter, Fl. Carol. 68 1788. (syn: Commelina hasskarlii C.B.Clarke) ?;

Himalaya to India (as per WCSP)

Commelina caroliniana
, sometimes known as the Carolina dayflower, is an herbaceous plant in the dayflower family native to India and Bangladesh. Both the scientific name and the common name are misleading as the plant was described based on specimens found in the southeastern United States before it was known that the plant had been introduced from India. It was most likely introduced to South Carolina in the late 17th century along with rice seed from India. Its flowers emerge from summer to fall and rarely into the winter.
Commelina caroliniana is an annual herb with a diffusely spreading growth habit. It will readily root at the nodes when they come into contact with the soil. The stems are decumbent (i.e. lying on the ground with rising tips) to scandent (i.e. climbing). The glabrous (i.e. hairless) leaves have blades that are lanceolate to lanceolate-elliptic or lanceolate-oblong in shape, measuring 2.5 to 10.5 cm in length by 0.7 to 2.4 cm in width. The leaf margins are scabrous (i.e. with rough projections), while the apex is acute to acuminate in outline.[3]
The inflorescences are important for making a proper identification, and especially for distinguishing it from Commelina diffusa. The upper cyme, also called a cincinnus, is typically vestigial and included inside the spathe, though it is can be rarely 1-flowered and exserted. The spathes are solitary and bright green in colour, though becoming paler towards the base. Unlike Commelina communis, it lacks veins with contrasting colour. They are pedunculate (i.e. supported on a stalk), with the peduncles measuring 0.6 to 2.3 cm. They are generally not falcate (i.e. sickle-shaped), though they may be slightly so (cf. Commelina diffusa). They measure 1.2 to 3 cm in length, and rarely up to 3.7 cm long, by 0.5 to 1 cm in width. Their margins are not fused and are usually ciliate (i.e. with a fringe of hairs), while the apex is acuminate in outline. They are glabrous or very lightly pilose (i.e. with fine soft hair).[3]
The flowers are bisexual with all three petals being blue, though the smaller lower petal is white towards the centre. The centre-most stamen has a white connective (i.e. the portion connecting the two halves of the anther). There are 3 cruciform staminodes present, each with yellow antherodes, these often showing a maroon spot at their centres. The fruit is a capsule, each having 3 locules and 2 valves and measuring 6 to 8 mm, though they may be rarely as small as 5 mm. Each capsule contains 5 dark brown seeds that measure 2.4 to 4.3 mm long, and rarely up to 4.6 mm, by 2 to 2.3 mm wide, though rarely they can be as narrow as 1.6 mm. They are smooth to faintly alveolate (i.e. with a honeycombed surface) with a mealy texture. The chromosome number is 2n = ca. 86.[3]

(From Wikipedia on 15.10.15)

Commelina caroliniana is an endemic species to the India Subcontinent. It is commonly found in fields, swamps and irrigated fields. It was introduced in South Carolina in the year 1696 and spread across to the southern east part of United States. It is also  present in Phillipines and Guam. In India it is found in Andaman, Delhi, Gujarat, Goa, Kerala, Maharastra, Rajasthan, Tamil Nadu and Uttar  Pradesh. The species has been located from 32 localities in the India sub continent. The area of occupancy and extent of occurrence are wide.  Data on population trends and conservation has not been reported. It is therefore listed as Least Concern.
Commelina caroliniana is commonly found in fields, swamps, yards, waste places, along roadsides , rarely in the forests. It is also a weed in crops, especially in heavy irrigation field such as rice.
Citation:Rehel, S. 2013. Commelina caroliniana. The IUCN Red List of Threatened Species 2013: e.T177167A7382133. . Downloaded on 15 October 2015

sk2015sept08/08 - Commelina sp. on the bank of the pond : 12 posts by 4 authors. Attachments (10)
These photographs were recorded on 02-Sept-2015 and today. This species is thriving on the bank of ponds, village drainage, even on sides of railway tracks, possibly in damp soil.
Branching and rooting at nodes.
Flowers are the size of C. benghalensis as can be seen in the last pic (flowers in right is C. benghalensis) where two species growing side by side. Leaves are upto 7.5 cm long and 2 cm wide. Only a very few leaves are narrower than this. This species doesn't flower much, only a very few shoots bear flower.
Please help.
Commelina species so far in efi     
Isn't it Commelina forsskalii ?
Thank you very much ... for the suggestion. But, I simply do not know how to identify a Commelina. The problem is BSI restricts the distribution of C. forsskalaei Vahl. in the Peninsular and NW India.
Many many thanks ... If BSI checklist didn't prevent me I would have very much identified it with the same taxon you have suggested.
There is very much confusion between C. diffusa Burm f. and C. caroliniana Walter (in US). C. caroliniana is conspecific with C. hasskarlii of India -
This species is C. hasskarlii C. B. Clarke.
Attached here relevant docs (downloaded from the net) for your perusal and further opinion.
Attachments (2)- 116_IJAR-5827.pdf & 2907-11801-1-PB.pdf
This species is very close to Commelina diffusa Burm f. 
I can rule out the possibility of Comelina caroliniana Walter (or Commelina hasskarli C.B.Clarke) because of two reasons
1. It is very clear that, this species has an upper cincinnus
2. Spathe is glabrous as opposed to serrulate to pubescent spathe.
Also, Comelina caroliniana Walter (Fl. Carol. 68. 1788) is the correct name since the name has priority over Commelina hasskarli C.B.Clarke (Commelyn. Cyrtandr. Bengal 13, t.3. 1874). A careful reading of the abstract of the paper would have helped 😊
NB: There is one more species of Commelina in this post.
Image number P1160722 has Commelina benghalensis (right side) in addition to the species in quiry.
Thanks to ...
BSI checklist for this region can be viewed at efi thread.
As per per the checklist above C. diffusa Burm. f. not to be found in this region
Even if we assume that the BSI checklist is outdated, and C. diffusa later finds its way to south Bengal also, I doubt if such an introduced (in this region) species can be so pervasive. In fact this is the most common Commelina thriving everywhere in SINGUR, HARIPAL, DANKUNI, CHANDITALA blocks of Hooghly districts. Presently the species is not in flowering state. Only one or two can be seen in one or two population. All my previous uploads of this species in this group and in facebook group (4 years ago) had been identified as diffusa!

Now, something about ... points (ruling out caroliniana Walter) -
  1. no where it was/is mentioned that caroliniana cannot have an upper cincinnus. Instead what author noted in his paper is "upper cyme usually vestigial (rarely well developed and 1-flowered)." The same has been copied in FoNA. Against this FBI, Bengal Plants, of-course outdated, noted "upper cymes 2-4 flowered.". The author's view in the paper (page 45) I referred can also be taken account in this regard.
  2. no where I could find mention of spathe in caroliniana is serrulate!
  3. spathe of caroliniana "margins distinct, usually cilliate, apex acuminate, glabrous or very sparsely pilose."
If, based on the above 3 points, we want to negate the probability of this species being hasskarlii (caroliniana) yet accepting the probability of diffusa as Manudev Ji has indicated we need to see/read the author's view in the paper I referred - "... Although a detailed comparision between C. caroliniana and C. diffusa and the variation exhibited by these species in the United States will be the subject of another paper, specimens of the two species can be separated by the following key :
  • Spathes not all to slightly falcate; upper cyme usually vestigial (rarely.....).... capsules....   .... C. caroliniana
  • Spathes usually distinctly falcate; upper cyme in larger spathes usually well developed and 1-several-flowered; capsules... C. diffusa
Interesting ..... until we examine a number of spathes and cymes at different locations of several populations!!!
(as for naming convention I followed BSI checklist which doesn't mention caroliniana)
(as for benghalensis I mentioned it in the initiating post itself !!!)
Since we are discussing upper cyme, three more pics of this specimen of same population. Attachments (3)
I fail to find any reason to rule out possibility of C. caroliniana Walter (C. hasskarlii C. B. Clarke).
Few more days required to conclude its identity. For time being keep it in Commlelina diffusa.