atmospheric phenomenon? What causes condensation trails to converge? This air plane just caught my eye. Two contrails apparently are flowing backward, slightly off-centered and then ultimately converge, giving the overall shape of a very narrow rhomboid parallelogram, with one diagonal being on the order of about four plane-width's.
Please explain what I am seeing here as detailed as possible. 
Some of you may even know what class of plane this may be and thus infer the scales involved, to give some hints of the scale of the contrail under investigation?
I looked up other pictures, and if I am not mistaken, most contrails are slightly angled outward. Is this by design or due to expansion and convection of the gases involved? Can someone confirm this?

Edit:


*

*I did observe the described contrail-shape, with quite some consistency, over varying angles of ascension and declination.

*I have never heard of conspiracies regarding contrails (e.g. chemtrails) until after writing this post. It shows that good, hard explanations for atmospheric contrail phenomena are in demand.

*Given the right circumstances (large distances) some apparent contrail convergence can be attributed to perspective alone.
Links:


*

*Contrail Science (impressive resource, which I just peeked at so far)

*Contrail gallery

*Contrails - clouds (ppt)

*http://www.es.lancs.ac.uk/hazelrigg/amy/Introduction/The%20Science.htm

*http://www.epa.gov/oms/aviation.htm

*http://www-pm.larc.nasa.gov/newcontrail.html

*http://www.epa.gov/oar/oap.html

*http://www.history.com/shows.do?episodeId=464914&action=detail

*http://mynasadata.larc.nasa.gov/P4.html
 A: I will make it an answer instead of a comment.
My guess is that the convergence is an optical illusion. This plane is flying at a level where the relative humidity is small.This means that the trail evaporates, it will evaporate faster from outside (the trail itself is humid) and finally what is left is merged the dissolution giving the impression of convergence.
in addition, this image 

and this second image 

shows clearly that there is a latteral expansion in the thinner air of the more dense exhaust, which would contribute to the joining of the two tracks in your image.
The narrowing is due to the faster evaporation in the moisture depleted air at the outer boundaries.
A: All aircraft producing lift have a wake, because the wing deflects the air downward, while the air further away is not deflected. This takes the form of a pair of "wingtip vortices".
Since they transfer energy to the surrounding air as they dissipate, they expand in radius, because momentum is conserved. So they capture the condensation trails from the engines, causing them to twist and swirl with the vortices.
Those condensation trails eventually evaporate, at a rate depending on the humidity.
This shows the vortices clearly:

Note that the vortices are more pronounced when the plane is moving at a relatively low speed, as it is during takeoff and landing.
(I think it's on approach to landing, based on wheels, flaps, low nose, and position of camera.)
Why the trails seem to evaporate outside-in, if they actually do, Anna's answer is not a bad guess.
In this video, the vortex is clearly shown. Note that the smoke is showing the vortex from the right wing.
EDIT: This video gives a pretty good idea of wingtip vortices, wake turbulence, and how pilots study about these things.
A: The vapor pattern that we can see on this pictures is due to the instability of trailing edge vortices. Here are some picture showing an instability of this vortices (in French for English see aslo this JFM). Different plane will have different trailing edge vortices resulting in different observable vapor pattern. On Wikipedia the trailing edge vortex are revered as wingtip vortices. The geometry of the plane create a trailing edge vortices and this groupe of vortex are unstable causing the rhomboid parallelogram.
