105 years of Average Daily Temperature Differences

I have been examining the average of each day's temperature difference for two closely spaced towns for the entire time that temperatures have been taken at the two nearby stations. Again I will look at the lunacy found in the record of Montevideo and Milan, Minnesota. When I take the average temperature difference for each day in February from 1894 to 2008 I find the average temperature difference is as seen in the following chart.




The question I have is why is the 105 year average temperature difference for Feb 6 1.7 deg F different than the average temperature difference for Feb 8th? It seems to me that there shouldn't be that large of a temperature difference. Once again, if one believes the thermometer record, one is believing in a lunatic data set.

I also would contend that the fact that for 105 years Montevideo has been hotter for 26 out of 28 days on average for February, clearly says that the data is attrocious. Such biases should not exist for towns only 16 miles apart. Whatever we are measuring, it isn't global temperature.