After about 26 emails and about 9 months, in October 2004, Crowley told me that he had "mis-placed" his original data, but did send me a smoothed and transformed version. The version in the email ties to Figure 2 of Crowley and Lowery , but differs from the spliced versions in the spaghetti diagrams of Mann et al. [EOS 2003] and Jones and Mann . The version in the spaghetti diagram of Briffa et al.  contains an extension from 1983 to 1987, not shown in the original Figure, in which only 2 and then 1 proxy are used.
Although Crowley and Lowery  presented the composite of all 15 proxies, later versions used a 13-site subset, which altered the relation of the MWP and 20th Centuries in the proxy record.
In my next post, I’ll show the effect of individual proxies in the Crowley reconstruction.