2.3 Sum rule
Now suppose that you know the formulas for the derivatives of two functions, and you want to know the formula for the derivative of their sum. For example, you already know the formulas for the derivatives of the functions and , but suppose that you want to know the formula for the derivative of the function .
When you add two functions, the -coordinates of the points on the two graphs are added. For example, Figure 27 shows the graphs of , and , and the point with -coordinate 1 on each of the three graphs.
The effect that adding the functions has on the gradients is quite complicated to think about, but in fact it’s just what you might expect: if you add two functions, then the gradient of the sum function at any particular -value is the sum of the gradients of the two original functions at that -value.
For example, Figure 28 shows the tangents at the three points marked in Figure 27. You can see that they have gradients 2, 3 and 5 (the sum of 2 and 3), respectively.
The general rule can be stated as follows.
Sum rule (Lagrange notation)
If , where and are functions, then
for all values of at which both and are differentiable.
For example, since the derivative of is , and the derivative of is , it follows from the sum rule that the derivative of is
The sum rule extends to sums of larger numbers of functions. For example, if , then
for all values of at which all of , and are differentiable.
It also follows from the sum rule, together with the constant multiple rule, that if , then
for all values of at which both and are differentiable. You can see this by writing the equation as
Then, since the derivative of is by the constant multiple rule, it follows by the sum rule that
as claimed. Again, this fact extends to sums and differences of larger numbers of functions. For example, if , then for all values of at which all of , and are differentiable.
Here is the sum rule expressed in Leibniz notation.
Sum rule (Leibniz notation)
If , where and are functions of , then
for all values of at which both and are differentiable.
Like the constant multiple rule, the sum rule can be formally proved by using the idea of differentiation from first principles, and you’ll see this done at the end of this section. First, however, you need to practise using it.
The next example shows you how to do this. In particular, it illustrates that you can sometimes differentiate a function that isn’t expressed as a sum of functions by first rearranging its formula so that it is. It also illustrates how you can use the sum rule together with the constant multiple rule.
Example 4 Using the sum rule
Differentiate the following functions.
a.
b.
c.
d.
Solution
a.
By the sum rule, you can differentiate each term individually (and add the results).
, so
b.
First write the function as a sum of functions, by multiplying out.
, so
Differentiate each term individually.
c.
First write the function as a sum of functions, by expanding the fraction.
, so
Differentiate each term individually. Simplify your answer.
d.
By the constant multiple rule, the derivative of is 5 times the derivative of . To differentiate , differentiate each term individually.
, so
Alternatively, multiply out the brackets and then differentiate each term individually.
, so
Activity 10 Using the sum rule
Differentiate the following functions.
a.
b.
c.
d.
e.
f.
g.
h.
i.
j.
k.
Answer
a., so
b., so
c., so
(Notice that in general the derivative of is , as you’d expect.)
d., so
- e.
so
- f.
so
g., so
h., so
(Note that this function can’t be differentiated in a similar way to the function in part ((g)), because the multiplier of the brackets, , isn’t a constant, and hence the constant multiple rule doesn’t apply.)
- i.
so
(The final answer can also be expressed as
- j.
so
(The final answer can be expressed in various ways. For example, you could combine the fractions and simplify the result like this:
- k.
so
In this section you’ve seen how to differentiate any power function, and you’ve met the constant multiple rule and the sum rule. So, in particular, you can now differentiate any polynomial function, that is, any function of the form
Of course, you can also differentiate many other functions too, such as some functions involving negative or fractional powers of .
You’ve seen that every function of the form , where is a positive integer, is differentiable at every value of , and that so is every constant function. These facts, together with the constant multiple rule and the sum rule, give the following useful fact.
Every polynomial function (with domain ) is differentiable at every value of .
To finish this section, here’s a proof of the sum rule, using differentiation from first principles. It uses the Lagrange notation form of the sum rule, which is repeated below.
Sum rule (Lagrange notation)
If , where and are functions, then
for all values of at which both and are differentiable.
A proof of the sum rule
Suppose that and are functions, and that the function is given by . Let denote any value at which both and are differentiable. To find , you have to consider what happens to the difference quotient for at , which is
(where can be either positive or negative, but not zero), as gets closer and closer to zero. Since , this expression is equal to
that is,
which is equal to
The expression in the first pair of large brackets is the difference quotient for at , and the expression in the second pair of large brackets is the difference quotient for at . So as gets closer and closer to zero, the values of these two expressions get closer and closer to and , respectively. Hence the whole expression gets closer and closer to . So
which is the sum rule.

