a tvo, That is not true in this case, 207 The records from the roof pots are much more variable than the records from the gummed paper, not only that,, but they are gener~ ally somewhat higher, So, I have done the simple atunt of reducing them to the same variability. There are two ways of doing that, In one way we can divide each roof pot value by the standard deviation of the roof pot values and*we can divide each gum- med paper value by the standard deviation of the gummed paper values and we will come out with a variance of one in each case, and so the variances will be equal and we will get thie situation, or, we can let these values stay as we record them and multiply the gunmed paper values by ea factor which will equalize the variability between the two, We have been doing this piecemeal at various times and we have collected this data and this factor which we multiply this by, in order to equalize this (indicating) 4s around 1.6, Summing up the 88 values that we have in this group (we actually have 95 weeks in this

Select target paragraph3