Here is what i have figured out as a measure for the “real” magnification of panel A in the paper by White et al, 2008, on rat and mutant SP-D molecules.  My original observation was that the SP-D dodecamers were too large for their 100nm bar marker. That I attempted to confirm by creating two circles, one the size of their bar markers, the other a circle that intersected at least 2 of the CRD in the SP-D dodecamer.  The discrepancy was large, being about 132%.  I next could see that the background had a larger grain in A of their figure than B (and now i believe it is off even more in C) and measured that, and A was about 115% larger shadowing grain than B.  Thinking of different shadowing on different grids which must have some variation, I felt that could be improved upon, and surmised that the mutations in their mini SP-D was only in the collagen-like domain, not in the neck or CRD.  That said, I measured height and width of the CRD in panels A and B and compared the mean of those measures.  (there were 31 observations in panel A and 80 observations in panel B.  Some i didn’t measure as they were obscured by other details. The means and sd showed a significant increase in the size of the CRD in panel A over B.

Here is the micrograph, and the pink dotted lines show distances measured (a height and a width per CRD).  This too showed that A was about 115% enlarged over panel B. So now when one (as i will) measures the decrease in the collagen-like portion of the mutant (mini) SP-D, i will enlarge the images in mini SP-D in panel B before comparing them with the images from panel A.  That the neck and CRD in panel C are again smaller i wont work with this since there maybe some change (as observed in the “lack of separation” in some of the CRD in the images, so other things are changing the shapes there.  (please note that the background grain size is even smaller than in B. (so much hassle for such a simple error)
Thank you to two great websites which I totally love, as i can cut and paste from excel and get numbers i want.  SD calculator, and t-test calculator.

Working at this moment on a cover submission for the American Journal of Human Genetics which brought me to a place of pseudocoloring chromosomes in a faux multicolor FISH.  I began to see these as little south of the border hat dancers…. forgive.  Actually they represent chromosome pairs 1-5.