WMAP: A Radiological Analysis II.

WMAP images have an exceedingly low signal to noise (just in excess of 1). Final images are made from 12 section images, which in turn are processed using separate linear combinations of data. ILC coefficients do not remain constant from year to year. In contrast to standard practices in medicine, difference images are presented at substantially reduced resolution. ILC images are not presented for year 2 and 3. Rather, year-1 data is signal averaged in the 3 year data set. Proper tests of reproducibility require viewing separate yearly ILC images. Fluctuations arise from the inability to properly remove the galactic foreground and in the significant yearly variations in the foreground itself. Variations in the map outside the galactic plane are significant, preventing any cosmological analysis due to yearly changes. This occurs despite the masking more than 300 image locations. Any ``signal'' observed by WMAP is simply the result of foreground effects not only from our galaxy, but indeed yearly variations from every galaxy in the universe. Contrary to published analysis, the argument suggests there are no findings in these images other than those related to image processing, yearly galactic variability and point sources.