Let's find the limit of f(g(x)) as x approaches 1. The graph of g has a hole at (1,2), so g(1) is undefined and we can't just plug in 1. But for a limit, we only care about what happens as x gets close to 1.
Remember how composition works. The input x goes into g, and the output g(x) becomes the input for f. So f never sees x. It only receives the outputs from g.
f(g(0))=f(1)=2
Find limx→1f(g(x)) from the graphs of g and f, and evaluate limx→1lnx−1x2−1
Text solution
Find where the inputs to f goFrom either side of 1, the heights on g approach 2.
x→1limg(x)=2
Use that f is continuous at 2There's no hole or jump in f at 2, so as its inputs approach 2, its outputs approach f(2).
x→1limf(g(x))=f(2)=3
The formula: simplify the inside, then plug into lnThe inside fraction gives 00 at 1, so we factor the top and cancel the x−1. The inside approaches 2, and ln is continuous at 2.
x→1limx−1x2−1=x→1lim(x+1)=2
x→1limlnx−1x2−1=ln2
Both answers came from what continuity means. If the inputs approach L, and f is continuous at L, the outputs approach f(L). That gives us our rule.
If x→alimg(x)=L
and f is continuous at L,
then x→alimf(g(x))=f(L)
2
When the outside function has a jump
Now keep the same g, but let f jump at 2. As x approaches 2 from the left, f(x) approaches 3. From the right, f(x) approaches 5. At 2, the solid dot gives f(2)=1.
We can't just plug in 2 and get f(2)=1, because the rule only works when f is continuous at 2. The inputs g(x) still approach 2, but now it matters which side of 2 they arrive from, so we look at x approaching 1 one side at a time. We'll do that for this g, and then for a parabola, g(x)=(x−1)2+2, with the same open circle at (1,2).
Find limx→1f(g(x)) when f has a jump at 2
Text solution
First g: check which side of the line y=2Left of 1, the points on g stay below the dashed line, so f gets inputs from the left of 2 and we use its left piece. Right of 1, they stay above, so we use its right piece.
x→1−:g(x)→2−⇒f(g(x))→3x→1+:g(x)→2+⇒f(g(x))→5
Different sides, no limitThe two sides approach different numbers. And f(2)=1 is never used, because near x=1 the input g(x) is never actually 2.
3=5⇒x→1limf(g(x)) does not exist
The parabola: above 2 on both sidesNear 1, the parabola stays above the line y=2 on both sides, so both sides use f's right piece.
x→1−:g(x)→2+⇒f(g(x))→5x→1+:g(x)→2+⇒f(g(x))→5
x→1limf(g(x))=5,
although x→2limf(x) does not exist
Look at the parabola's left side. There, x approaches 1 from the left, but g(x) approaches 2 from the right. Which piece of f we use depends on whether g(x) is above or below 2, not on which side x approaches from. And we don't even need the graph to see that the parabola stays above 2. For any x=1, the square (x−1)2 is positive, so adding 2 always gives a number greater than 2.
(x−1)2>0(x=1)
⇒g(x)=(x−1)2+2>2
If the parabola opened downward instead, with its vertex at (1,2),g(x) would stay below 2 on both sides of 1, and both sides would use f's left piece and give 3.
3
When the inside function stays constant
Same jumping f, new inside function. This time, g(x)=2 for every x between 0.5 and 1.5.
Find limx→1f(g(x)) when g(x)=2 for x near 1
Text solution
The input is exactly 2Near 1,g(x) doesn't approach 2 from the left or the right. It is 2, so f only ever receives 2.
f(g(x))=f(2)=1for x near 1
The limit of a constantWe never plug in x=1, but every nearby x already gives 1.
x→1limf(g(x))=1
Here the value f(2) is what counts, unlike in section 2. There, g(x) only got close to 2 and was never equal to it. Here it equals 2 for every x near 1.
How does the value of g(x) determine which part of f we use?
Limits of a composite function
4
When g has no limit
Keep the same f, but now let g jump at 1. As x approaches 1 from the left, g(x) approaches 2, and from the right, it approaches 4. So limx→1g(x) doesn't exist, and our rule, which needs g(x) to approach a single number, can't be used. But we can still check each side of 1 on its own.
Find limx→1f(g(x)) when limx→1−g(x)=2 and limx→1+g(x)=4
Text solution
Follow each side through fFrom the left, g(x) approaches 2 from below, and f jumps at 2, so the side matters and we use f's left piece. From the right, g(x) approaches 4, where f is continuous, so the side doesn't matter and f(g(x)) just approaches f(4).
x→1−:g(x)→2−⇒f(g(x))→3x→1+:g(x)→4⇒f(g(x))→f(4)=3(f continuous at 4)
Both sides give 3
x→1limf(g(x))=3,
although x→1limg(x) does not exist
5
When the inside and outside swap
Now let's find the limit of g(f(x)) as x approaches 6, using the same two graphs. This time f is on the inside, so we start with the graph of f, and whatever comes out of f goes into g.
Find limx→6g(f(x)) from the graphs of g and f
Text solution
The inside firstf is continuous at 6 and f(6)=1. But g jumps at 1, from 2 on the left to 4 on the right, so we need to know whether f(x) approaches 1 from above or from below.
x→6limf(x)=1
f is decreasing near 6Just left of 6,f(x) is just above 1, so g's inputs come from the right of 1. Just right of 6, they come from the left.
x→6−:f(x)→1+⇒g(f(x))→4x→6+:f(x)→1−⇒g(f(x))→2
4=2⇒x→6limg(f(x)) does not exist
Summary
When the outside function is continuous
x→alimg(x)=L,f continuous at L
⇒x→alimf(g(x))=f(L)
x→1limlnx−1x2−1=ln(x→1lim(x+1))=ln2
Find the limit of the inside function first. If that limit is L and the outside function is continuous at L, plug L into the outside function. For lnx−1x2−1 as x→1, the inside approaches 2, and since ln is continuous at 2, the limit is ln2.
When the outside function has a jump
g(x)→L+
⇒f(g(x))→x→L+limf(x)
g(x)→L−
⇒f(g(x))→x→L−limf(x)
When f has a jump at L, heights above L on g's vertical axis become inputs to the right of L on f's horizontal axis, so g(x)→L+ uses the right-hand limit of f. Heights below L become inputs to the left, so g(x)→L− uses the left-hand limit. The side of L depends on whether g(x) is above or below L, not on whether x approaches a from the left or the right.
If g approaches different numbers from the two sides of a, check each side on its own. If f is continuous at one of those numbers, it doesn't matter which side g(x) approaches it from.
When the inside function stays constant
g(x)=L for x near a
⇒x→alimf(g(x))=f(L)
If g(x)=L for every x near a, then f receives only L, not numbers approaching L. The limit is f(L), even if f has a jump at L.
Swapping the inside and outside functions
In g(f(x)),f is on the inside, so look at the graph of f first. Find the limit of f(x). If g has a jump at that limit, check each side of a separately. When f(x) approaches that value from above, use the right-hand limit of g, and when f(x) approaches from below, use the left-hand limit.
6
Practice
Enter exact values, or write DNE if a limit does not exist.
1
Evaluate x→0limexsin3x.
ex is continuous everywhere, so we can move the limit into the exponent. What is the limit of the exponent as x approaches 0?
3xsin3x approaches 1 as x approaches 0. How can you write xsin3x with a 3x in the denominator?
If you got e Check the limit of the exponent. The angle inside the sine is 3x, not x, so xsin3x approaches 3, not 1.
ex is continuous everywhere, so we can move the limit into the exponent.
Multiply the top and bottom by 3 so the bottom matches the 3x inside the sine. As x approaches 0,3x approaches 0, so 3xsin3x approaches 1.
xsin3x=3⋅3xsin3x→3⋅1=3
The exponent approaches 3, so the limit is e3.
x→0limexsin3x=e3
2
The graphs of f and g are shown above. Find (a) x→−1limf(g(x)), (b) x→2+limf(g(x)), (c) x→2limf(g(x)).
As x approaches −1, does g(x) approach 1 from above or from below?
If you got DNE in (a) Even though limx→1f(x) does not exist, g(x) stays greater than 1 on both sides of −1, so we only use f's right-hand limit at 1.
If you got 3 in (a) Check whether g(x) equals 1 when x is close to −1 but isn't −1. For x near −1 with x=−1,g(x) stays greater than 1, so f never receives 1 itself.
For x just to the right of 2, is g(x) less than 1 or greater than 1?
If you got 2 in (b) Check whether g(x) is above or below 1 to the right of 2. The graph of g is decreasing through (2,1), so g(x) approaches 1 from below.
In part (b) we found the limit from the right. What does g(x) approach as x approaches 2 from the left?
If you got 3 in (c) Check whether the input to f equals 1 for nearby x other than 2. For x near 2 with x=2,g(x) is never equal to 1, so the value f(1)=3 is not used.
Limit at −1The graph of g has a local minimum at (−1,1), so from both sides of −1,g(x) approaches 1 from above, where f approaches 2.
x→−1−:g(x)→1+⇒f(g(x))→2
x→−1+:g(x)→1+⇒f(g(x))→2
x→−1limf(g(x))=2
Right-hand limit at 2For x just to the right of 2,g(x) is less than 1 because the graph of g is decreasing through (2,1). As the input approaches 1 from the left, f approaches 4.
x→2+:g(x)→1−⇒f(g(x))→4
Two-sided limit at 2As x approaches 2 from the left, g(x) is greater than 1, so the input approaches 1 from the right and f approaches 2. The one-sided limits differ, so the two-sided limit does not exist.
x→2−:g(x)→1+⇒f(g(x))→2
2=4⇒x→2limf(g(x)) does not exist
3
The graphs of f and h are shown above. Find x→0limf(h(x)).
For x near 0, does h(x) only approach 1, or does it equal 1?
If you got 2 or 4 Check the graph of h near 0. Because h(x) is always 1 near 0, we use the value f(1), not a one-sided limit of f.
If you got DNE Check whether f needs a limit at 1. Near x=0,h(x) is always 1, so f(h(x)) is always f(1).
For all x between −1 and 1,h(x) is exactly 1, so f(h(x)) is the constant f(1)=3.
h(x)=1 for −1≤x≤1⇒f(h(x))=f(1)=3 for x near 0
x→0limf(h(x))=3
4
The graphs of f and k are shown above. Find x→0limf(k(x)).
The graph of k has a jump at 0, so check each side of 0 separately. As x→0−, does k(x) approach 1 from above or below?
As x→0+, what does k(x) approach? Is f continuous at that value?
If you got DNE Even though limx→0k(x) does not exist, check the one-sided limits of f(k(x)) on each side of 0.
If you got 4 To the left of 0,k decreases toward 1, so k(x) approaches 1 from above.
As x approaches 0 from the left, k(x) decreases toward 1, so the inputs to f approach 1 from above.
x→0−:k(x)→1+⇒f(k(x))→2
As x approaches 0 from the right, k(x) approaches −1, where f is continuous.
x→0+:k(x)→−1⇒f(k(x))→f(−1)=2(f continuous at −1)
Both one-sided limits equal 2, so the two-sided limit is 2.
x→0limf(k(x))=2
5
The graphs of f and k are shown above. Find x→3−limk(f(x)).
In k(f(x)), which function is on the inside?
For x just to the left of 3, is f(x) positive or negative?
If you got 1f is inside, so find what f(x) approaches before looking at k. The graph of f is decreasing through (3,0), so f(x) approaches 0 from above, not below.
The function f is on the inside, and because it decreases through (3,0), its values are positive for x just to the left of 3.
x→3−:f(x)→0+
Now find the limit of k as its input approaches 0 from the right.
k(f(x))→x→0+limk(x)=−1
6
Let f(x)=x−2x2−4 for x=2, and f(2)=0. Find x→1limf(x+1).
As x approaches 1, what does x+1 approach? Can x+1 equal 2 when x=1?
Factor x2−4 to simplify the formula for f(x) when x=2.
If you got 0 In a limit as x→1,x gets close to 1 but never equals 1, so the input x+1 never equals 2.
As x approaches 1,x+1 approaches 2, but for x=1 it never equals 2.
x→1:x+1→2,x+1=2(x=1)
Because the input is never 2, we use the rule for x=2. The x−2 cancels, so f just adds 2 to its input.
f(x)=x−2(x−2)(x+2)=x+2(x=2)⇒f(x+1)=x+3(x=1)
x→1limf(x+1)=x→1lim(x+3)=4
7
Let f(x)=x+1 for x≤0, and f(x)=4−x for x>0. Find x→0limf(x2).
Can x2 ever be negative?
Because x2 approaches 0 from the right, which piece of f applies?
If you got DNE Check which side of 0 the inputs x2 approach from. For any nonzero x,x2 is positive, so f only receives inputs from the right of 0.
If you got 1 The inputs x2 are positive for all x=0, so f never uses the formula for x≤0.
For any x=0,x2 is positive, so f(x2) uses the piece for positive inputs.
x2>0(x=0)⇒f(x2)=4−x2
As x approaches 0,4−x2 approaches 4.
x→0limf(x2)=x→0lim(4−x2)=4
8
Let f(x)=3x for x<1, and f(x)=x+4 for x≥1. Find x→0limf(xsinx). (Challenge)
From the squeeze theorem, what does xsinx approach as x→0?
On the unit circle for 0<x<2π, is the vertical height sinx shorter or longer than the arc length x?
If you got 5 Check whether xsinx approaches 1 from values less than 1 or greater than 1.
From the squeeze theorem, the limit of xsinx as x approaches 0 is 1.
x→0limxsinx=1
On the unit circle, the vertical segment sinx is shorter than the arc x for 0<x<2π, so xsinx<1. Because sin(−x)=−sinx, the same inequality holds for negative x.
0<x<2π:0<sinx<x⇒xsinx<1
−2π<x<0:xsinx=−xsin(−x)<1
The inputs to f stay strictly less than 1, so f uses the piece 3x.
f(xsinx)=x3sinx(x near 0,x=0)⇒x→0limf(xsinx)=3