1) The results go in opposite directions, so the relation is negative linear or inverse.
2) It's pretty clear that if you compare the change in E / Change in d, for the 1st to 2nd point and the 2nd to 3rd point that you do not have a linear relationship
3) Since it is inverse in some way: E = a/dn where n is not known, you should calculate 1/d and see how it compares to E. (You could also compare di/dj to Ej/Ei for pts i and j to see if the ratio is = which would imply a simple inverse relation)
4) The inverse relation fits pretty well as the first to the second point halves 1/d and the E halves as well. Checking 3rd and 4th point, we find the 1/d ratio is .55 and the E ratio is .59.
If you plot lnd vs lnE you will get a slope of -1 which tells you that E = ad-1