[SVXX]

Nice info cupid! Hyperconjugation makes sense too. Yes, it would've been para directing in that case, I think the same.
As resonance is cancelled, hyperconjugation would be of no use in benzyne, then!

[cupid.callin]

But are you completely sure resonance wont effect benzyne?

I didn't find any good proof of that on internet and my books are not with.
I am going by your words.

[SVXX]

Yes I am dead sure. Think of a simple vinyl carbanion....would you be able to resonate it? Nope. The case is similar here. It's a proven fact too, so you needn't worry on that part!
Now all that is left is orgopete's opinion on the major product.

[orgopete]

If you perform the search embedded here and click on page 614, this should give you Anslyn and Dougherty's explanation.

First of all, I did not know what the products were of different substitution patterns. The point I was trying to make is that is that explanations are rationalizations to explain why the products formed. It would not be unusual for someone to suggest another rational. For example, carbonyl addition models, such as, Cram, Felkin, and others, see http://evans.harvard.edu/pdf/smnr_2000-2001_Siska_Sarah.pdf.

One also must exercise caution in analyzing the results. A meta-bromotrifluomethylbenzene probably only produces a single benzyne while meta-bromotoluene gives a mixture. A different mixture of products can result, but it may not indicate a weakness for a para carbanion. That result from the para bromo however does give the same product and thus proves the para carbanion is less favored.

As I recall, a meta-bromoanisole reacts faster than an ortho. I have seen it argued that this is a coordination effect or an inductive effect. A coordination effect could be supported by simple directed metalation experiments.

I would not discount drawing resonance structures for the benzynes. It won't break any rules and won't hurt anything. You may find it to be useful for understanding. Let me give an example of why you might consider it. If you draw the resonance structure for dimethylaminopyridine, the resonance structure increases the electron density of the nitrogen, yet you should recognize that the hybridization of the nitrogen probably does not change and that it is indeed the non-bonded electrons of the nitrogen that are used in acylation reactions. I might argue that a benzyne is not a triple bond in a benzene ring. Triple bonds are linear. A benzyne might also be thought of a benzene ring with a pair or electron orthogonal to the pi-electron of the benzene ring, and thus similar to pyridine. If you can draw a resonance structure for dimethylaminopyridine, then why not a resonance structure for the benzyne of anisole. If you did that, you may well argue that it controls the addition of a nucleophile and the absence of this effect is why a methyl group results in a mixture.

[SVXX]

I don't think the carbanion formed after nucleophilic addition to benzyne can resonate. We could draw resonance structures for benzyne; not that carbanion. In any case benzyne is highly strained and tries to form biphenylene, so it doesn't last long unless trapped.

[orgopete]

Consider the following:
     

I have drawn a series of resonance structures. I have drawn the pyridine examples to suggest the pyridine nitrogen might have an increased reactivity. I have drawn a similar example with the anisole benzyne. The pi-bond is presumed to be orthogonal to the benzene electrons. This resonance structure would suggest that an increase in electron density may be found at the ortho and para positions. I have drawn a similar set of resonance structures for the carbanion intermediate. The resonance structures of the carbanion do not suggest an increased stability on that basis.

I don't KNOW what the answer is. After seeing the ratios of products that form, I became skeptical of an inductive effect of an N,N-dimethylamino group and an O-methyl ether to favor meta products from the benzyne. I accept a chlorine having that effect. However, a methyl led to approximately equal o/p products for a weak inductive donor. That resulted in my searching for another possible explanation. Since resonance effects are generally stronger than inductive effects, I thought the resonance effect was possible. Also, I don't see any significance to the resonance structures of the adducts as I don't think there is convincing evidence these reactions are reversible. The resonance structures of the benzyne simply indicate that a higher electron density my repel attack at an ortho position. It is just an hypothesis. 

[SVXX]

Well....that was quite a bit of brainstorming on your part, orgopete. My sincere thanks to you for putting in time and effort, and I finally see what you meant by the resonance structures. Neat diagrams!
So the final answer is still inconclusive....can anyone solve this conundrum?

[orgopete]

Before we can solve the mechanism(s) and directing effects, we need more data (and perhaps your text is wrong).

p-Bromoanisole gives equal amounts of m/p products, http://pubs.acs.org/doi/abs/10.1021/ja01584a025

Some reviews: http://euch6f.chem.emory.edu/arynes.html
http://stoltz.caltech.edu/seminars/2006_Tadross.pdf (see p. 12, 13, and conclusions p. 43)

I think one must be careful in deriving mechanisms for reactions solely by looking at the substituents. One cannot look at the electronegativity value of oxygen to figure out why t-butanol has a different acidity than trifluoromethanesulfonic acid. The caltech review had an oxazolylbenzyne adding with different regiochemistry depending on the reaction. Clearly coordination was having one regiochemistry and the lack of it had the opposite.

I foresee a problem in that the generalizations to explain regiochemistry are being contradicted by data. (This is also an objective of people in synthesis, reverse an unfavorable selectivity and success complicates generalizations.) I'm done.

[cupid.callin]

@orgopete
I agree with your concept of triple bond being linear and hecne under great strain in this case ... but in your resonating structures, you drew C and N with 10 and even 12 electrons ... this is merely impossible for them under any drastic conditions ... though i havnt completely read your links as they are too long 

@SVXX
I was saying that R effects can effect benzyne but after looking at them ( I didnt drew them myself .. just drew them in my mind  ) i have to say that i now agree with you now that benzyne is not effected by R effects ... I knew that while triple bond is formed the elec. of the bond orthogonal to the ring wont participate in resonance but now i guess that there wont be any resonance as IMPOSSIBLE STRUCTURES are made!!!!
maybe we have to stick to your earlier statement that R effect donot effect benzyne as now i think  there is NO resonance.

In my view, resonance may be present till bond is not shifted to any1 atom, but as bond is shifted, it would block resonance and structure will be totally dependent on I effects. So bond will be shifted so that structure(not resonating) will be max stable ...
SO i can say Benzyne mechanizm  donot depend on R effects"

though i will confirm it again and get back on this!!!

[cupid.callin]

@ orgopete
In your PdF, it just tells how EWG and EDG effects the mechanism. but it didnt tells EW or ED in what sense? I or R? as in case of OCH3 it is EW by I-effect and ED by R effect. which will dominate is the main quest of the question.

[orgopete]

I realize the purpose of the question. However, as SVXX pointed out, either his textbook was wrong or his understanding was wrong, therefore I became more cautious and skeptical. Maybe his textbook was wrong or maybe it really isn't as clear as it might be. So, you read an interpretation in which someone suggests product distribution followed electron withdrawing or donating. I looked up some data and for the 3-bromoanisole example that SVXX related, the product is meta. If 4-bromoanisole were used to generate the benzyne SVXX illustrated, the actual result is an equal mixture of products. Well, what does that do for all of the donating or withdrawing? It looks as though a 3-substituted benzyne will give a meta substituted product whether you interpret it as electron donating or withdrawing, except methyl has little effect and an equal mixture. A 4-substituted benzyne will give mixture of meta and para products. Interestingly, para-fluoro did give a higher ratio of para.

If you asked me to summarize the effect, I would not. I am not sure there is one that I can identify. If there were more reactions that I could find, then maybe I could find an effect that I would be more confident about.

10 or 12 electrons? I guess you have not seen this interpretation,
http://chemwiki.ucdavis.edu/Wikitexts/UCD_Chem_118B/Chem_118B_Topics/Activating_and_Deactivating_Benzene_Rings

Google doc