Residuals show how far the data is from the line of best fit.
If a line is a good fit for the data, then the residuals will be about half below and half above the x axis.  The residuals will also be close to the x-axis which means that their values are fairly small.
If the residual plot shows a pattern (see below), then a linear will not be a good fit.  See the link from Math Bits below:
http://mathbitsnotebook.com/Algebra1/StatisticsReg/ST2Residuals.html
Looking at the 4 choices, which looks most like the the left most graph that says “Random, No Pattern.  Linear Appropriate”?

Look at the graph above the minimum y value is -7 at the point (-3,-7).
We are looking for the y = that goes even  lower!!

Above is answer choice (1).  You can see the minimum y value is -2 (not less than -7)

Above is answer choice (2).  You can see the minimum y value is -6 (not less than -7)

Above is answer choice (3).  Whoa this one goes low!! Lower than the y value of the given graph way up above

Above is answer choice (4).  This graph does not go very low — it doesn’t even pass through the x-axis as its minimum y value is 2.