How tree branches are attached to trunks

[sloth]

If you could that'd be most appreciated, I think I cleared it out when I moved...

[Gary Prentice]

Its at home and I'm at work, I'll bring it in to scan in the AM

[sloth]

Still at work eh, long day! No rush, and thanks again

[KTSmith]

So is there a question on the table? The most recently posted images to this thread simply show branch stubs or what an American lumberman would call "pin knots". Normal branches in every way. They died (due to shading, probably) and were "shed" in the sense of the breaking of symplast connections, yet remained attached. Successive years of vascular cambium growth of the stem encased them. The encasing wood is appressed to the branch stub but not confluent with it. The branches are, as always, only attached at the base.

Or have I missed the point entirely or is there no point of discussion on the table.

If so, a fresh thread might be in order.

This thread had been asking for Shigo's 1985 attachment paper. It's readily available from any good library or interlibrary loan, but I'd be happy to send it along. Perhaps I already have! This thread, like me, is a bit long in the tooth.

[sloth]

Hi KTSmith, I'm certainly not posing any question in particular. I just feel Slater's paper makes some big statements. It got me thinking that I always just took Shigo's model of attachment as fact without question. When someone comes along 20 odd years later and casts doubt on it publicly for the first time, I feel I ought to have a closer look at the whole situation...

[Gary Prentice]

Hi KTSmith, I'm certainly not posing any question in particular. I just feel Slater's paper makes some big statements. It got me thinking that I always just took Shigo's model of attachment as fact without question. When someone comes along 20 odd years later and casts doubt on it publicly for the first time, I feel I ought to have a closer look at the whole situation...

 

I believe that the only significant difference in Matthecks model is that the tails of the branch vessels don't unite below the collar, as in Alex's model. There's a lot of work by others which complement Shigo but nothing that backs up Duncans clever clip.

 

After reading his work I spent a few days pulling unions apart on saplings and twigs. In most of the species I tried I couldn't reproduce the clip. Mattheck explains it in mechanical terms quite well, in his talks.

 

Kevin, could you make much sense of the confusion between co-dom forks and branches?

 

Edit; I forgot that article this morning, an early start and it went out of my head.

Edited January 14, 2015 by Gary Prentice

[KTSmith]

Duncan Slater and I had an extended, collegial, and useful discussion on his contention that the principal difference between co-dominant forks and branches was one of relative size of the members, what is sometimes referred to as aspect ratio.

 

I respectfully submit that is nonsense. My point of view was published some time back in Arboricultural Journal and is also linked here: Letter to the Editor [response to "Towards a new model of branch attachment" by D. Slater and C. Harbinson]. I think that was presented and discussed in Arbtalk at that time.

 

Does that argument still need to be made? If so, I'd pull up my notes from the earlier exchange, but a fresh thread might be in order. Or perhaps the British point of view has this all sorted out and there is no need for discussion. This is all pretty "old news" to me, and perhaps to the rest of the community as well!

 

The Shigo 1985 branch attachment paper is attached.

Shigo Branches 1985.pdf

[daltontrees]

Ahhh the penny has just dropped. You are in fact that KT Smith! If you look back at around posting 15 in this thread you can see I posted your paper. I havppen to agree with you etirely about the fallacy of codominant stem unions just being very steep branch unions. That alone was enough for me largely to dismiss the Slater work.

 

I also comemnted recently on someone's request for a definition of meristem. It reminded me that ny model or theory for branch attachment has to recogniuse some very simple basic truths oabout cell division and differentiation. I believe that once elongation of primary meristem has taken place (elongation and lateral deviation, the source of all branches), the residual meristem activity then creates specialisms in the function of the cells to create vascular paths adn thereafter secondary meristem is repsonsible for putward growth and all future annual inctements; seeing it in these stages underlines the fundamental difference between steep branch unions and codominant forks. They can never be the same in terms of attachment or connections (in the Shigo sense of the word).

 

But it also creates a problem for the Shigo model that just doesn't go away. The Shigo model is an unresolveable topology problem. The upward vascular connection from the stem to the branches cannot put on tissue next year. Once created it is overlain entirely by stenm vascular tissue. The annual increment next year cannot arise from the outer face of the branch-in-stem tissue because, even if there were meristematic cambium cells present they would not be able to expand and they would be crushed against the inner face of last year's stem wood.

 

And so the only resolution to the topological problem is for the branch-in-stem wood to have externally connected cambium. Running parallel to the stem cambium until it flows under, around and over the bark collar. Ths could accommodacte all the apparent attributes of the Shigo model including the overlapping appearance, dye testing, branch collar regeneration, as long as the usual transverse flow properties of wood are accepted as being present in unions just as they are in any other adjacent wood cells.

[KTSmith]

Ha, Ha, I thought I replied to above an hour ago, but I must have mashed it up.

 

I don't see where the problem is. Consider a healthy stem and branch system, no cankers, burls, scars, etc. Topologically, the vascular cambium is a single, continuous layer of cells covering branch, stem, (and roots...).

Regulated by hormones in a manner not completely understood, the VC of the branch begins division earlier in the growing season than the VC of the stem. The orientation and differentiation of the daughter xylem cells is a function both of the orientation of the VC and of the cascades/pulses of those hormones. The annual ring of the branch completes formation followed by the annual ring of the stem. But throughout this process, the VC remains a single, continuous surface always to the outside of the derivative daughter cells.

OK, perhaps this will help: Visualize the lower surface of the angle formed by the stem and branch. Some portion of the VC surface will produce branch xylem early in the growing season and stem xylem later in the growing season. There is no "branch VC" getting covered by stem xylem. It's one, continuous surface. So does that make sense?

[Gary Prentice]

Does to me.