One way how a tree benefits from bark exfoliation is to prevent epiphytes (e.g. mosses, lichens, etc.) from establishing themselves. It also allows expansion of the truck without causing fissures.

Yes, your second proposition could be also valid for many tree species: trees are growing faster than the bark adn so causing  the bark to rupture. However, especially in the tropics, it seems that the story is much complex.

Well, actually no-one will ever know since the tree can never tell you! It is teleological question. However, it can be assumed that they do it for the above reasons – to exfoliate, remove disease, and get rid of damaged tissue, etc :-)

Thanks for the philosophical input. The exfoliating of Oreomunnea impressed me very much... so i was wondering if there is any review or synthesis on this phenomenon.

Couldn't it be a side effect of fast growing without obvious benefit for the tree ?!?

Yes, in fact this is very often the first answer you can get, especially for the temperate trees, and especially among foresters. The story seems to be slightly complicated in the tropics with some extreme examples - there some other evolutionary/ecological adaptations must be considered.

In subalpine and northern forests a reduction of epiphytes reduces the snow load in winter. Does the trees in warn areas deposit any toxic substances in the bark?  And do they (1) get rid of them that way or (2) make their surroundings toxic against other organisms? At least the leafs and parts of the fruits of some Juglandaceae contain the toxic Juglon and other naphthoquinone derivatives.

This is an interesting addition. I will include it to the list of potential explanations.

In tropical forests, exfoliating bark appears to be effective in reducing the number of climbers (vines, lianas, bush-ropes) on trees.  Examples include Terminalia oblonga (Combretaceae), Calycophyllum spruceanum (Rubiaceae), Macrolobium spp. (Caesalpiniaceae), Eucalyptus deglupta (Myrtaceae).  Bursera simaruba (Burseraceae) is an exception because the papery thin outer bark exfoliates exposing the photosynthetic, greenish inner bark.

Thank you very much for this additional information and examples.

Shedding of bark is not as prominent in the tropics compared to temperate trees. Is it not because the tropics do not experience temperature ranges as wide as that of the temperate regions? During the winter in temperate places, the ourter cells of the bark must be getting damaged much more than in the tropics.

I have no idea. I think that all three theories you have mentioned above could be possible, depending on the species and conditions of occurrence. It could be also interesting to do some review of trees and shrubs which peeling their bark and participation of such a species in the floras of particular biomes.

The process of peeling can also be explained as mechanical reaction of the bark to the radial growth of the tree trunk. In this case the process depends on the character of the bark and a manner of its breaking/cracking, eg. Betula versus Zelkova carpinifolia or Acer pseudoplatanus, Caria laciniosa etc.

In the Mediterranean Arbutus andrachne and in great part also Platanus orientalis (at least some specimens) remove their bark every year.

Thanks for these valuable remarks and ideas. Indeed, this could be interesting to do kind of synthesis on a global scale.

With reference to the answer of Jose some small factual contribution to Gregor's endeavor:

The bark of some deciduous temperate zone trees is damaged ('sun burn') by high temperature/radiation in summer  (e.g. beech/Fagus; maple/Acer) rather than by frost as postulated. 

Commiphora bark peels in (dry) tropical areas. 

Some Eucalyptus sp in tropics too do bark removal but it true mostly it happens in temperate zone species. The exfoliation process is related to temperature and  climatic response of trees to adjust the temporarily change. Physiologically it can improve the regeneration of new bark cells as process of secondary growth in trees.

Thank you very much for these additional indications.

 Anatomically, Shedding of bark is the part of secondary growth in the cortical region of the stem due to formation of cork cambium or phellogen. According to Esau's plant anatomy it depends partly on the characteristics and manner of the growth of the periderm itself and partly on the amount and kind of tissue separated by the periderm from the axis. Exfoliating bark has different structural bases. If thin-walled cells of cork or phellem are present in the periderms of  outer bark (rhytidome), the scales may exfoliate along these lines. Breaks in the rhytidome can also occur through the cells of non peridermal tissue. In Eucalyptus breaks occur through phloem parenchyma. The advantages of exfoliating bark may be the same as you have explained.

Thanks for these further anatomical details and explanations. I see now, that my question is too general - what I wanted to know / better understand was the "extreme/exceeding exfoliating" as observed in only few tree species (e.g. Oreomunnea, Platanus, etc.) but not a "typical" renewal and growth of the outer bark. In any case your hints are very valuable, indicating the importance of anatomy in explaing this phenomenon.

Gregor, While not strictly pertinent to your later more specific query, others may be interested in two additional possible purposes for peeling bark, and these only in fire-resistant trees. Bark that regularly fissures and separates probably facilitates penetration by epicormic shoots, a common characteristic of pyrophytes. And those who have witnessed eucalyptus fires may have have noticed how loosely-hanging bark can conduct ground fire up to the crown from where convection and wind may waft burning pieces several km to spread the conflagration. Never underestimate Mother Nature!