Since I tried numerous grafted tomato plants I figured I enter some of my findings. I won't argue that there may be some disease resistance benefit, especially in the Southeast where soil borne diseases are more of an issue. I'm in the northeast where those diseases are not a major concern. First I'll state that grafting is a chalenge. sometimes it appears that the graft was a success when in reality a shoot from the rootstock developed and grew. Since the rootstocks recommended are Beoufort(Sp) or Maxifort (the vatiety I used) and neither produce desirable fruit you can imagine the frustration at the uncertainty of what I was really growing. Then there is the planting depth issue. You must be extremely carefull to only plant to below the graft or your grafted top may take root and defeat the purpose. Lastly, you need to "Prune". Many are against the concept of "Suckering" lateral branches. If you graft be prepared to NEED to vigorously sucker the unwanted new vegetative growth from the rootstock. Otherwise you will have loads of strange little yellow wierd flavored fruits that are of no use on stems that you never planned for.

My bet is 90% as bad sucker growth if planted perfectly. It just seems to be the nature of the beast. What you should do? I dunno. Packing dirt around them would be a temptation. Many trees live like this a long time in nature near streams and lakes. Replant? Probably not. I have a Bradford type from the previous owners. If several of my larger trees had not required removal it would be gone now. Guess I am shell/chainsaw shocked lol. I just can't kill another! Now that said, my "right" to complain when my neighbor plants bamboo or zoysia near my fescue has been limited by my own decision to keep this invasive stinky mess until the snow or wind removes its blight from my zip code. damn Post Tree Removal Disorder lol.

But getting back to basics, we have another case that appears similar. According to Darlington (Evolution of Genetic Systems, 1958) I am sure that we could take each example, of the many that were published many years ago, and follow the trail of more recently published research that provided useful information for determining if the original conclusions were reasonable. We would probably learn a great deal about many interesting subjects. But it would be time-consuming and from my viewpoint I am handicapped by having poor access to libraries with older research journals. I expect that time would be more efficiently used by examining the original research papers and determining whether accepted scientific practices were used and thus their conclusions warranted. I am going to continue with Frankel's research because although I do not think he was a fraud or a fool I do consider that his conclusions were unwarranted. It is clear from his published comments that he knew exactly where his experimental design failed to eliminate alternative explanations. His 1956 paper was considered by van Marrewijk as having shown an effect of the grafting so I will use it as the basis for my comments. From a quick examination of recent research published by the Chinese and Japanese groups on graft-induced genetic changes I expect that, in general, similar comments could be made in relation to many of the papers published on similar subjects. Frankel's paper gives no indication that he randomized the growing of any of his plant material and thus any systematic environmental differences present were not eliminated from affecting his observations. It is normal practice to have a randomized design to account for such potential effects. Nor did he provide any description of how he classified plants as sterile versus fertile. If there was any subjectivity involved in that classification then 'blind' methods would be needed to eliminate any researcher biases (that is the person classifying the plants would not know which plant was from the experimental group and which was from the control group). And most seriously he had no control for the effect of grafting. The description in the paper indicated that the effect of grafting was severe both on the scions and on the seeds produced from them. He needed a group that was fertile scion grafted to fertile stock to be able to compare and eliminate the effects of maternal environment on the offspring. These are known in plant species and can be short-term or long-term. If he had a control group he would then have had to analyze the differences in fertility between the two groups to show that the experimental treatment had a significant effect. His research material was open-pollinated, genetically variable and thus permitted alternative explanations involving selection, etc. There was no description of how he chose the plants to provide the scions, was one plant used for all scions? were a number of plants chosen at random used? and so on. He clearly understood that his design allowed alternative explanations as he states: "For several reasons these data do not permit extensive speculation about the nature of this cytoplasmic sterility." and he went on to mention the high heterozygosity, the lack of large numbers of offspring, the possibility of distorted segregation ratios due to high seedling mortality. In my opinion his conclusions are unwarranted and although he thought, "There is a remote possibility, for instance, that nutritional deficiency, which might have been induced by grafting, might cause a disturbance in cytoplasmic enzyme activity in such a way as to lead to an increase or decrease of sterility-determining entities of the cytoplasm." I consider that is the most likely explanation (omitting the specification of enzyme activity). With hindsight, from what is known now about cytoplasmic male sterility and mitochondria, the most likely explanation is that his fertile stock had individuals with substoichiometric mitochondrial variants present and that the stress induced by grafting caused a change in proportions. That type of explanation would also help explain why the results of replicating the experiment in Petunia could fail - that is, they would fail if the fertile stock chosen did not harbour the mitochondrial variant and they would fail if the grafting was easily successful and did not provide high levels of 'stress'. If 'Rosy Morn' has spontaneously produced cytoplasmic male sterile individuals then that would be evidence that it contains individuals with substoichiometric (one variant DNA copy per every, eg 100-200 cells) quantities of the variant mitochondria. It is intersting to note that a report (Bentolila, et al)quoted "In Petunia, a single dominant nuclear gene termed Rf confers fertility to lines carrying the only known CMS cytoplasm in this genus (16)." Reference 16 is a paper by Izhar, et al and Bentolila, et al thank Izhar for providing Petunia stocks; Izhar worked with Frankel. Apparently none of the cytoplasmic male sterile lines from grafted individuals are available from Frankel's work or the other research groups. It doesn't really matter at this point whether transposons, mitochondria, viruses, plasmagenes or some as-yet-unidentified phenomena are involved. Although for Touchstone, as the graft has already been made and a variant found this does apply, in general I cannot agree that it does not matter. For example, if one is interested in genetic variation caused by a transposon then it would be easier and more efficient to begin with a stock that has an active transposon than to graft a stabilized strain in the hopes of reactivating the transposon and producing interesting genetic variants. The question returns to "does grafting cause any changes that are different from those that can be caused by alternative methods?" and "can those alternative methods be more practical/effective/efficient than grafting?" If the cytoplasmic male sterility history (Frankel) is a general example I would conclude that the answer to those two questions, in order, is no and yes.

There is a technique used to graft a healthy nurse piece between the top and lower trunk and roots of an apple tree. Though not often done, the purpose is to bypass a heavily damaged trunk area and sustain life and growth of the tree top works. The technique is used on smaller trees to repair damage done by deer rubbing their antlers on the trunk and scouring away the bark. I think it would be a natural for the purposes you describe. In your case you would simply wind up into a loop all the stem you desired to eventually eliminate, and I believe this is pretty much the way it is in a tomato green house. Then you would cross the stem at the point where you would make one graft by sciving away about 1/3 to 1/2 of each stem area where crossed, bind and tape these together until they bond and grow together, then cut out the coil after bonding had occured. In other words you jumper the stem around the coil, allowing it to heal before cutting out the coil. Certainly some experimental work would be necessary but I believe it will work well. When I operated an orchard I did make a couple of the grafts described on deer damaged small trees and both did survive. (However, they looked mighty strange for a number of years until the nurse piece was big enough to support the load and the old damaged section could be cut out.) I think one tree is still going after 13 years.