Dear paper committee. Here is a revised draft, which incorporates almost all of your comments. It is this draft that I would like to use as a basis for our discussion at Thursday's meeting. [Agenda] Major issues to discuss at our meeting. --- S.A. suggests that a complete description of CC Coherent Pion need not be included in this draft, and that only a reference to the SciBar paper is necessary. A counter-argument to this is that, while SciBar is the only detector that can *measure* the coherent pion content, but SciFi did also see a discrepancy (and has reported it at conferences and in the 2005 oscillation paper) which is now significantly resolved. The draft reflects this new understanding, but we may consider removing it. --- In this draft, I have clarified that the "unmodified" predictions for pmu and thetamu are from the 2004 published spectrum fit (but with no CC coherent pion, and NQE/QE = 1.06), for the purposes of showing baseline distributions before the MA fit. We now have an unpublished 2005 spectrum fit, which will be (may be) revised once before next spring. We may discuss whether the unpublished 2005 spectrum should be used for these plots, and determine what information must be included in this case. --- C.W. says that we should be very clear about how and why we perform the Delta-theta cut to separate the QE and NQE enhanced samples. This is a vital part of the method of the analysis. This is included more completely now. --- C.W. also suggests that the effect of proton rescattering, especially the difference between the zero final state interactions and what we actually use, is something that has not been reported clearly in the literature. Since it is vital to the delta-theta method, a more clear statement of that is given. I have included a plot of the distribution of the number of tracks with our regular rescattering, and with final state interactions, based on a slide from Walter, NuInt01. We should discuss whether this is the best way of illustrating this point. --- This is the first significant physics-results paper from the SciFi group, and it is intended to be a PRD paper with no official length restrictions. Does it adequately represent all the important details of SciFi that are not already covered in the NIM papers, or oscillation papers? Or does something need to be added, or expanded with a plot, withouth distracting from the main focus of the paper? --- Members of the paper committee are invited to express any other concerns, or questions that should be addressed before this manuscript is circulated to the collaboration. --- S.A. has expressed some ideas about the interpretation of, and how to extract physics meaning or constraints for the nonQE/QE parameter. I have not yet followed up on these ideas, though we may choose to talk about them at our meeting. Other comments about changes in this draft. --- The plots are improved, but not yet good enough. Plots take a long time. --- As I said I would, this draft has a single normalization parameter, common to both data sets. The results and plots are updated. As expected, the results did not change, but there is one more degree of freedom now.. --- The contribution to the total error from the relative flux and normalization was estimated incorrectly. I accidentally included nonQE/QE in that estimate. The corrected value is \pm 0.06. The total error is still 0.12. --- M.Y. asks why the efficiency and purity are different than they were for the 2005 oscillation paper. You will also notice that the muon momentum threshold for K2K-I has been changed. For the MA anlaysis, we do not use the Lead Glass stopping events, which is a large effect on all these numbers. Also, the updated MC, with the correct SciFi mass has an effect. There may be a further effect if the numbers in the oscillation paper were not with the fit spectrum, but in the MA paper, they are certainly with the published fit spectrum reweighting. --- I have removed all instances of stat and syst error separately. In most places only one error is given (the total error 0.12). In the results section I attribute 0.03 to statistical. --- M.Y. says "the spectrum shape in the higher energy region should be correlated with the MRD energy scale." I have not addressed this specifically, though I have done a large amount of work on this issue. I will try to prepare some more details, possibly as an internal note for the K2K collaboration. --- S.A. suggests a longer comment about what could be different between the K2K-I and K2K-IIa data samples. I think her meaning is that the two fit values diverge, probably at the edge of the 1-sigma errors, if only uncorrelated errors are included. I may still add such a description, but it is not there yet.