These eight properties let a definite integral be simplified — often to zero, or to twice a simpler integral, or to another definite integral that is easier to handle — purely by symmetry or by a change of variable, without ever writing down the antiderivative explicitly. In every proof below, g denotes any primitive of f, i.e. ∫f(x)dx=g(x)+c, and the Fundamental Theorem (Section 4.2) is used to convert each bracketed evaluation.
Property I.∫aaf(x)dx=0.
Proof: ∫aaf(x)dx=[g(x)+c]aa=(g(a)+c)−(g(a)+c)=0 — the upper and lower limit are the same point, so the bracket has nothing to measure.
Property II.∫abf(x)dx=−∫baf(x)dx.
Proof: ∫abf(x)dx=g(b)−g(a)=−(g(a)−g(b))=−∫baf(x)dx. Swapping the limits of a definite integral flips its sign.
Worked examples: ∫31xdx=[2x2]31=21−29=−4, while ∫13xdx=[2x2]13=29−21=4 — exactly the negative of each other, confirming the property.
Property III.∫abf(x)dx=∫abf(t)dt.
Proof: both sides reduce, via the Fundamental Theorem, to g(b)−g(a) regardless of whether the dummy integration variable is called x or t. This says definite integration is independent of the name given to the variable — the definite integral is a fixed number determined only by f, a, and b, not by the symbol used inside. Worked example: both ∫π/6π/3cosxdx and ∫π/6π/3costdt evaluate to [sin(⋅)]π/6π/3=sin3π−sin6π=23−21=23−1.
Property IV (additivity over sub-intervals).∫abf(x)dx=∫acf(x)dx+∫cbf(x)dx, for any c∈[a,b].
Proof: the right side is (g(c)−g(a))+(g(b)−g(c))=g(b)−g(a), which is exactly the left side. Worked example: for ∫−15(2x+3)dx with c=3, both the direct evaluation [x2+3x]−15=(25+15)−(1−3)=42 and the split evaluation [x2+3x]−13+[x2+3x]35=[(9+9)−(1−3)]+[(25+15)−(9+9)]=20+22=42 agree.
Property V.∫abf(x)dx=∫abf(a+b−x)dx.
Proof: substitute t=a+b−x in the right side, so x=a+b−t, dx=−dt; as x runs from a to b, t runs from b down to a. So ∫abf(a+b−x)dx=∫baf(t)(−dt)=−∫baf(t)dt=∫abf(t)dt (using Property II), which by Property III equals ∫abf(x)dx. This is the single most useful property in the chapter: reflecting x about the midpoint of [a,b] never changes the value of the integral.
Worked example: for I=∫π/6π/3sin2xdx, reflecting via x→6π+3π−x=2π−x turns sin2x into sin2(2π−x)=cos2x, so I=∫π/6π/3cos2xdx too. Adding the two expressions for I: 2I=∫π/6π/3(sin2x+cos2x)dx=∫π/6π/31dx=3π−6π=6π, so I=12π.
Property VI.∫0af(x)dx=∫0af(a−x)dx — the special case of Property V when the lower limit is 0 (so a+b−x becomes a−x).
Worked example: for I=∫0π/4log(1+tanx)dx, reflecting via x→4π−x and using tan(4π−x)=1+tanx1−tanx gives 1+tan(4π−x)=1+tanx2, so I=∫0π/4[log2−log(1+tanx)]dx=(log2)⋅4π−I. Solving, 2I=4πlog2, so I=8πlog2.
Proof: in the second term on the right, substitute t=2a−x (dx=−dt; x:0→a gives t:2a→a), so ∫0af(2a−x)dx=∫2aaf(t)(−dt)=∫a2af(t)dt. Adding ∫0af(x)dx+∫a2af(x)dx gives exactly ∫02af(x)dx by Property IV. This is the tool used whenever the upper limit is twice the useful reflection point.
Property VIII (odd/even functions).
∫−aaf(x)dx=⎩⎨⎧2∫0af(x)dx,0,f even (f(−x)=f(x))f odd (f(−x)=−f(x)) …