Explanation by Liora Engel (with editing by Allen) of the slope-point
method, by which you can get an approximate equation of a line without a
computer or graphing calculator:

The general idea is to use the linear equation of the form y-y1=m(x-x1) to
get the equation of the line. In this case the y values are the log(molecular
weights) and the x values are the Rms. m is the slope of the line.

Here is how to do it: 
1.	Calculate the log(mw) for two values from the table. (Preferably the
  	values closest to the Rm of the unknown you are required to
  	calculate - unless the values are scattered (which should not happen
  	on an exam!), as opposed to curved.)
2.	Calculate the slope of the line using those two points (remember,
  	the slope is m= (y-y1)/(x-x1)).
3.	Plug the slope value into the equation of the line (in place of m).
4.	Choose a data point (an Rm and its corresponding log(mw)) from the
  	table.
5.	In place of x1 in your linear equation, plug the Rm value from the
  	data point.
6.	In place of y1, plug the log(mw).
7.	Solve the equation for y.
8.	You now have the equation of the line.

In order to find the molecular weight of your unknown, you must plug the Rm
value of your unknown into x and solve for y; you will get the log(mw) of
your unknown. Then take the antilog of the y value to get the molecular
weight of your unknown.

Be sure to check your final answer by comparing the molecular weight you
obtained from your equation with standards of similar Rms (e.g., the two
values from the table that you used).