2 © LIMITS AND DERIVATIVES 2.1 The Tangent and Velocity Problems 1. (a) Using P(15, 250), we construct the following table: (c) From the graph, we can estimate the slope of the tangent line at P to be “ 2 =e = 333 5 (5,694) |e So \ REE en 5304 approximate 2 J 0.111) oe aan pm 25 | (25,28) an) = 500] 30, (30,0) 2904 ‘1904 (b) Using the values oft that comespond tothe points closest ‘ to P(t = 10 and t = 20), we have 9 15 2 25 30 aes) 38.8 + (—27.8) 2 (a) Slope (b) Slope = 2048 # uso — 2048 aaa (©) Slope (2) Slope = From the data, we see that the patient's heart rate is decreasing from 71 to 66 heartbeats minute after 42 minutes. After being stable fora while, the patient's heart rate is dropping. 3. For the curve y = ¢/(1 +2) and the point P(1, 4) . ° a a ma] @ | 05 (0.5, 0.333333) 0.333333 (ii) | 0.9 (0.9, 0.473684) 0.263158. ii) | 0.99 | (0.99,0.497487) |} 0.251256 6) | 0289 | (0.909, 0496750) | 0.250125 (vy) | 15 (1.5, 0.6) 0.2 wi | ia (1.1,0.523810) — | 0.238095, (vin | 1.01 | (1o1,050248e) | oasras (viii) | 1.001 | (1.001, 0.500250) | 0.249875 (©) The slope appears to be 566 CO CHAPTER2 LIMITS AND DERIVATIVES 4, For the curve y = In and the point P(2, In2): @ (b) The slope appears to be 3 z Q meg @] 15 | (.5,0.405465) | 0.575364 Gi) | 19 | (1.9,0.641854) | 0.512933 yom Giiy | 1.99} (1.99,0.688135) | 0.501254 (iv) | 1.999 | (1.999,0.692647) | 0.500125 (| 25 | (2.5,0.916291) | 0.446287 (wi | 21 | (2.1,0.741937) | 0.487902 (vii) | 2.01 | (2.01,0.698135) | 0.498754 (viii) | 2.001 | (2.001,0.693647) | 0.499875, B(x —2) or 1+in2 @ secant line at x= 15 tangent line at x=2 secant line atx = 2.5 yeinx m2 secant fine ar 25 4 os ‘secant line Ing ares ys 0 15 2 2 5. (a) y= v(t) = 40t — 1602. Att = 2, y = 40(2) — 16(2)? = 16. The average velocity between times 2 and 2 + h _ w2+h) —y(2) _ [4002 +h) ~ 162+ hp] —16 — -24h- 16h? istwe = “OTe? _ = I = 24 = 16h ith #0. (i) [2.2.5]: h = 0.5, vay = —32f1/s Gi) [2.2.1]: bh = Odeo = -25.6 ft/s (ii) [2.2.05]: h = 0.05, vave = ~24.8 1/5 Civ) [2,201]: h = 0.01, vave = ~24.16 f/s (b) The instantaneous velocity when ¢ = 2 (h approaches 0) is ~24 ft/s. 6. The average velocity between t and t + h seconds is S8(L h) = 088+ H)*— (68. OSE) _ Hh 1.60 —OSBN _ 551 gor osdhith 40 h (a) Here t = 1, so the average velocity is 58 ~ 1.66 — 0.83h = 56.34 ~ 0.83h. G [1.2]: k= 1,55.51 m/s (ii [1,15]: k= 0.5, 55.925 m/s (ii) [1, 1.1}: h = 0.1, 56.257 m/s (iv) [1, LO: A = 0.01, 56.3317 m/s (9) {1,1.001]: f = 0.001, 56.33917 m/s (©) The instantaneous velocity after 1 second is 56.34 m/sSECTION 21. THE TANGENT AND VELOCITY PROBLEMS Cl 67 1. s = s(t) = t°/6. Average velocity between times 1 and 1 +h is — s(1+h)~s(1) _ (1+ h)/6-1/6 _ h°+3h7 43h _ WP +3h43 ith go. Yoh) h ~ oh 6 @ @ [3k h=2 Gi) [1,2]: b= Levee = Ets (ii) (1.1.5): h Gv) (LL): h BI ns “angent s(2) _ 178-32 _ 146 @ @A=3 > vy = = HE x asrtys (ih=2 > ww Fa assnys Giih=1 = vy = $B)= 82) _ = _ gg, 3-2 T (b) Using the points (0.8,0) and (5, 118) from the approximate tangent line. the instantaneous 2is about U8=2 a 28 ft/s velocity at 8. For the curve y = sin(10x/z) and the point P(1,0): @ z Q meq z Q meg 2 [20 0 05 | (05.0) 0 15 | (1.5,0.8660) | 1.7321 0.6 | (0.6,0.8660) | -2.1651 1.4 | (1.4,-0.4339) | —1.0847 0.7 | (0.7.0.7818) | ~2.6061 1.3 | (1.3, -0.8230) | ~2.7433 08 | (08,1) -5 1.2 | (1.2,0.8660) | 4.3301 0.9 | (0.9, -0.3420) | 3.4202 11 | (.1,-0.2817) | -2.8173 ‘AS approaches 1, the slopes do not appear to be approaching any particular value.68 CHAPTER? LIMITS AND DERIVATIVES CO) ‘We see that problems with estimation are caused by the frequent oscillations ofthe graph, The tangent is so steep at P that we need to take a-values much closer to 1 in order to get accurate estimates of its slope. (o) If we choose « = 1.001. then the point @ is (1.001, ~0.0314) and mpg © ~31.3794. If « = 0.999. then Q is (0.999, 0.0314) and mpg = ~31.4422, The average of these slopes is —31 4108. So we estimate thatthe slope of the tangent line at P is about ~31.4. 2.2 The Limit of a Function 4. As approaches 2. f(r) approaches 5. [Or. the values of f (2) can be made as close to 5 as we like by taking 2 sufficiently close to 2 (but x # 2).) Yes. the graph could have a hole at (2,5) and be defined such that f(2) = 3 2, Asx approaches 1 from the left. fc) approaches 3; and as x approaches 1 from the right, f(x) approaches 7. No, the limit does not exist because the left- and right and limits are different. 3. (@) im, f(@) = 0 means tat the values of f(2) ean be made arbitrary large (as large as we pease) Py taking 2 sufficiently close to ~3 (but not equal to ~3) (b) im, f(2) = ~o0 means tha the values of f(x) ean be made arbitrarily large negative by taking 2 sufficiently close to 4 through values larger than 4 (6) tim f(x) =4 (© tim, f(a) =2 4. (a) lim fle) = (2) lirn f(x) does not exist because the limits in part (b) and part (c) are not equal © f(3) 5. (a) f(z) approaches 2 as x approaches | from the left.so tim f(c) =) 3 (6) f(«) approaches 3 as 2 approaches 1 from the right so lim, f(«) (© lim, f(x) does nt exist because he Limits in part) and par (bare not equal (a) fe) approaches 4 as approaches 5 from the left and from the right so Tiny f(2e) = 4 (e) (5) is not defined. so it doesn't exist. 6. (@)_lim_g(2)= (b) tim g(a) =1 (©) lim, g(4) doesn’t exist (@ g(-2) =1 (Sim o(2) =1 (6) Jim, gle) =2 (2) lim, g(2) doesn't exist ) 92) =2 (im, o(e) doesn’t exist () lim g(x) =2 (&) g(0) doesn’t exist () Tim o(2) 2 7. (@) fim ft) = () sim, 9(6) (©) lim g(t) does not exist because the limits in part (a) and part (b) are not equalSECTION22 THELIMITOFAFUNCTION O69 fim g(t) (©) tim, g(t) = 0 ant (8) lim 9(¢) does not exist because the limits in part (4) and part (e) are not equal (@) 9(2)=1 () lim a(t 8. (@) lim R(c) () lim RE (©) lim _R(2) (@) lim, R(@) at (©) The equations of the vertical asymptotes are e = —3, x = 2. and x = 5, 9 (a) —00 (b) lim, f(x) = 00 (©) lim f(z) = co (@) lim (eo 00 (e) lim. f(z) = 00 (£) The equations of the vertical asymptotes are a -Tn=-3.2=0.ande 10, lim _ f(¢) = 150mg and lim, f(t) = 300 mg. These limits show that there is an abrupt change in the amount of re ont does noe since f(z) -+ —00 ast + ~1~ and f(z) + 00 asa -1* 18. For f(x) = xln(x + 2”) = | fe) 7 0.693147 0.5 0.143841 OL 0.220727 0.05 —0.147347 oor | -0.015952 0.005 | —0.026467 0.001 | —0.006907, Ie appears that im, x In(e-+2#) = 0 mae) = EE £ f(z) 40.2 | 0.439279 40.1 | 0.566236 £0.05 | 0.591893 £0.01 | 0.599680 £0,001 | 0.599997 tanSe 9g 3 Ttappears that lim appears that 2 Yan 5SECTION22 THELIMITOFAFUNCTION O71 2. For f(a) = = x f(z) x J(z) z f(z) z f(x) 05 | ovsssa7| [5 | oissas0 05 | 1sereea| [—05 | o227e1 09 Jomoser | | 11 | o4ssig 01 Jozi] | -01 | 0485084 0.95 | 60186 | | 1.05 | 0.540783 0.05 | 0.646496 | | 0.05 | osaasa7 099 | ors | | 101 | o.ssso22 0.01 | 0599082 | | -o.01 | 0.576706 0.999 | 0.601200 | | 1.001 | 0.598800 0.001 | 0.588006 | | —0.001 | 0.586669 ao1 ost Heappears that Jim == = 06 = § Ie appears tat Jim, °° —*" — 059, Later we will be able to show that the exact value is In(9/5), 2B. lim, G5 >Otorr > 5, = co since (« — 5) — Oas z+ 5+ and 24. lim = 00 since (2-5) + O.as 2+ 5~ and —°— Ofor0 <2 < x. Hy Ser = him, (1/082) = —oo since cos —+ 0 as 2 —» (—/2)" and cos. < O for —n 0 for x <—1 and i fore > 2.s0_ lim y= lim, y = 00. Also, as > —1” or -2 zr 2° the denominator approaches 0 and y <0 for —1 Base 1‘SECTION 23 CALCULATINGLIMITS USING THEIMITLAWS O77 h(B+h) tim (18+ 8h +1?) = lim Fi mb h = jim(8+h) =8+0=8 (e@-V(e? +241) @- Dery mm (tah + 6h? + ah8 +h) ~ = h h 2 AS = fim MAOH SAN +H) _ isn (4.4 6-4 4h? +43) = 4404040=4 mh ar) , (8+12h+6h? +h’) —8 12h + 6h? +h Jim = im POE oh uy h meh h = jim (12 + 6h +h?) =12+04+0=12 (3+ vt) (3~ vt) . jig SE = Jim (3+ Vi) = 3 + VO =6 Jim sa = tiny Sov (34 vt) =3+ V0 m MUFR=1 YIFR+I_ y MeAY“1 h A T+h+1 s0h(ViFh+1) 0h(VIFR+1) 1 1 = tim 1 1 mo Viphel Vivi = tim YEF2=3 VEFB+3_, (2+2)~9 2. lim = er ET VetO48 ete (yerd a) o-7 1 1 1 “241 (@—7(Ver2+3) 2" Jet243 Vora 6 = tim HWE =2)(0? +4) mh Ba2 = (2+2)(2? +4) = 32 = Him, @ + 2)(2* +4) = lim (2 +2) im (2? +4) aya red tim, 4 jim Be = im 244 yi LL ee weds te Earay 7 6 1 1 (P+t)-t e L L 26. |i = - = lim — ——— = lim —__ = ——=-——= tia (tae) = ba =m ae aye = gh = a? -81 in (2S) (2 +9) im (VE 3) VE +3) (2 +9) factor z~ 9 asa 27. lim SS = dim ee 2h Ye—3 3 iy VvE-3 difference of squares, = lim [(V@ + 3) (x + 9)] = (V9 +3) (9 +9) =6- 18 = 108 : 14 (3+h)"!~ 37! 3B+h_3 3=(@+h) 28. im BEN AS" sign SHR any Re ee 1 B+ AB 1 1 = fn [: 3@+h)78 C CHAPTER2 UMITS AND DERIVATIVES 2 im ( L ~4) =m et yp Ao VEFE) (14 VEE) _ tiny =t Ea iyree 0) OR Tyree erie viet) berets VIF 8) = Lim = a1 MS TT ED TT) = lim unveils ver difference of cubes} = Tim [ve (1+ ve +2)] = lim +140) =3 Another method: We “add and subtract” 1 in the numerator. and then split up the fraction: (e-+(1-2) G@-20+2) Jy CF DAER 2) — pm [1 CSE] vay + ve\+2) Ga ve)0+ VEO+9) ay evi Uo 14 (14 viya+0) a) 1s OO) z f(z) —o.001 | 0.666166 =0.0001 | 0.6606167 \ 0.00001 | 0.666617 0.000001 | 0.666662 ” 0.000001 | 0.6666672 tim 2 3 0.00001 | 0.666717 OES Se 0.0001 | 0.6667167 0.01 _ | 0.671663 “The Limit appears to be 5 1 2 AEE +1) hin TERE +N) _ HOVERS +) © 19 (FET Vipsesi) 2) (¥3e)—1 Se lim, (VIF BE +1) [Limit Law 3) =3{ Tit 35) + tim | {and 11) 1 =4( /imiesin and 7 =4( [imi + Timms +1) (1.3.and 71 {7 and 8]SECTION 23 CALCULATING LIMITS USINGTHELIMITLAWS O79 32. @) il ) x f(a) —0.001 | 0.2886092 =0.0001 | 0.2886775 0.00001 | 0.2886754 -1 1 0.000001 | 0.2886752 0.000001 } 0.286751 lim YEFE—VB 4 99 o.00001 | 0.286749 * 0.0001 | 0.286727 0.001 | 0.286511 ‘The limit appears to be approximately 0.2887. = lim 20 SF a4 V3 [Limit Laws 5 and 1) = ee (Tand 11) [Tim (3+ 2) + V3 1 1. 7.and 8 340+ V3 ae 1 “28 3. Let f(x) = -2, g(a) = 2? cos 20m2 and h(x) = 2?. Then 1S c0s20nr <1 = 2? < 2% c0s20nr S(2) < az) < h(x). So since fim f(2) = lim h(e) = 0. by the G Squeeze Theorem we have lim g(r) = 0 H. Let f(z) = —VaF FH g(x) = VeT Fa? sin(n/x). and A(z) = VOTH Then —1 < sin(n/2) <1 Vs Fa? < VF Fa sin(x/z)< Vee = Ile) < ax) < h(x). So since lim f(x) = lim, (2) = 0. by the Squeeze Theorem we have lim g(r) = 0 BAS fle) < 2942042 forall, Now tim, 1 = 1 and aim, (28 +2042) = limo? +2 + lim, 2=(-1)? +2(-1) +2 = 1. Therefore. by the Squeeze Theorem, lim, f(z 18) Cl CHAPTER2 UIMITS AND DERIVATIVES 36. 30 < f(x) < 2° +2 for0 et se cel = vale s yee < Yee. Since Jim, ( /2/e) = Oand im, (ye) = 0.we have lin, [veenr> | = 0 by the Squeeze Theorem, 8. We > —4.then|2 +4) =2+4.so_ lim, je+4|= lim (2 +4)=—4+4=0. Wer < ~4.then [e+ 4] =—(@44).s0 tim fot] = lim —(e-+4) =-(-444)=0. Since the right and lef Himits are equal, lim, |x + 4| = 0. lim, 1 = 1. Ife < 2.then ent =1. The right and left limits are different, 0 lim e=21 does not exist 42. tha > 3, then [2x — 3] = 2x — 3,0 3. Qn? — 32 2x? — 32 «(22 ~3) 3 lin im 22 B® ji ZC2=3) — jim 2 = 1.5. < 5th wim, paaay 7 =the Bes wlim, pea = lim ea 1S a < 9 then pe—3|=8—22,s0 tim 2 —¥ a lim -15 2h pe ay At s- — @r—3) ate 2a? — Br ‘The right and lft limits are diferent, so lim, “>=—ay- does not exist: ta ie e| =e era 0, we have tim G ) (a Te =o 46. (a) (b) @ Since sgn = 1 fore > 0. lim, sgn = tim, 1 (id Since sgn = —1 fore <0, lim sgn x= tim (i) Since fim sanz 4 lim, sgn. lin sam. does not exist, (iv) Since [sgn el = 1 forx # 0, lina [sen-2|SECTION 23 CALCULATING LIMITS USING THE LIMITLAWS <8 ee ao ena tn4=o op 1) = ep, (@— 0) = ig == in 1 2-1-1 (©) No, Jim f(2) doesnot exist since lim fe) # tim, f(0) 47. (a) (i) lim, Jim, salt [B=] ant 2 Gi) tim, = tim er fe-1] ats @—D) () No, lim F(x) does not exist since linn, F(x) # lim F(2) 48. (a) (i) lim, Ae fim, «? P=0 i) Jim h(x) = im 2 = 0, so lim h(z) = 0. it) iy (x) = tim 2? iv) lim h(z) = aah n, (8-2) =8-2=6 (i) Since him A(x) A lim, h(x). lim, A(x) does no exis. 49. (9) (0) fo] = —2for—2 c. 53, Since p(.e) is a polynomial, p(x) = ap + arr + az2® +--+ aya". Thus, by the Limit Laws, tim pla) = lim (a0 + axe + azz” +--+ an2") = ay +04 fim x +a fim a? +--+ am Tim 2" = ay + aya + a207 +--+ ana" = pla) ‘Thus. for any polynomial p, lim p(x) = pa). SA Letr(2) =F : where p(.c) and g(x) are any polynomials, and suppose that q(a) # 0. Thus, = [Exercise 53] = (a). lim r(2) 5 58, Observe that 0 < f(2) < 2° for alls. and lim 0 = 0 = lim 2°. So, by the Squeeze Theorem. ling f(z) = 0. ant) 56. Let f(2) = [2] and g(2) = —[2]. Then ling f(2) and lim, g(r) do not exis (Example 10) but fel) = limo=0, i [J (2) + (2) = Jim (Ce enh 57. Let f(z) = H(2) and g(x) = 1 ~ H(x), where H is the Heaviside function defined in Exercise 1.3.59. but ‘Thus. either f or gis 0 for any value of x. Then im f(z) and lim, g(r) do not exis Jim [f(@)g(2)| = fim 0 = 0 58. Ii vO=z-2 tin, (YEE? V6—=+2 et) lim, gra s1 se Va-e-1 ve-e+2 V8q241 _ (Vea) -2 Va-et1 =I | (ges) n Voqe#2 (@2-a(VE=F+1) _ 4, V=EHL = lim, 1m, Qa a)(vena+2) se VO—a+2 2 58, Since the denominator approaches 0 as z —+ —2, the limit will exist only ifthe numerator also approaches 0 as p> —2, In order for this to happen, we need lim, (32* +ar+a+3)=0 3-2)? 40-2) +a+3—=0 4 12-2a+a+3=0 4 a= 15. With a ~ 15, the limit becomes fim BEE ISEH NS jig SEF MEFD © hy HES) 3243) _ 3 ali, “pe s2 eee (@—1et2) eo? e- 2-1SECTION 24 THEPRECISE DEFINTIONOFALMIT O83 60. Solution J: First, we find the coordinates of P and Q as functions of r. Then we can find the equation of the line determined by these two points, and thus find the «-intercept (the point 2), and take the limit as r — 0. ‘The coordinates of P are (0.1). The point Q is the point of intersection of the wo circles x? + y? = r? and (x ~ 1)? + y? = 1. Eliminating y from these equations, we get r? ~ 2? =1-(¢-1)? & 1422-1 ++ © = }r°, Substituting back into the equation ofthe shrinking circle to find the y-coordinate, we get ry +e 72g yP=r(1— 414) y=ry/1— 32? (the positive y-value) So the coordinates of Qare (412,ry/1 = $72). The equation ofthe line joining P and @ is thus ~ 0). We set y = 0 in order to find the 2-intercept. and get -i (fT (+) itn) 1 lim 2(v1+1) =4. = Now we take the limitasr —+0*: im, 2 = lim 2( = 472 +1) raot poe So the limiting position of is the point (4,0) Solution 2; We add a few lines to the diagram. as shown, Note that Z PQS = 90° (subtended by diameter PS), So ZSQR = 90° = ZOQT (subtended by diameter OT). It follows that ZOQS = ZTQR. Also 1.98 < x < 2.02. Thus, x must be within 0.02 units of 2 so that Sar +3 is within 0.1 of 13, (b) Use 0.01 in place of 0.1 in part (a) to obtain 0.002. 2. (@) To have 6x — 1 within a distance of 0.01 of 29, we must have 28.99 <62~1< 29.01 = 29.99 < 6x < 30.01 = 4.9983 100 whenever 130, +1) (@—1F a 0.93 < a < 1.07, So since [1 — 0.93} = 0.07 and 1 ~ 1.07] = 0.07. we can take 5 = 0.07 (or any smaller positive number) os, 2286 © CHAPTER2 LIMITS AND DERIVATIVES. = 100, we need —0.0997 < a < 0or0 < x < 0.0997. Thus. we choose 6 = 0.0997 (or any smaller 12. For A positive number) so that if 0 < Ja| < 6, then eot?.x > 100, 200 _y=cot?x “02 L =—0,0997 0 = 0.0997 a For M = 1000, we need —0.0316 < x < 0or0 1000. 20 1 =-0.0316 9 = 0.0316 43. (a) A= xr? and A= 10000m? = mr? = 1000 => 1000 > r= [p> Oo] = 17.8412em. (b) [A 1000] <5 —5< ar? 1000<5 + 1000-5< xr? < 100045 > Bi srs fe = 17-7006 0) 7 =200 T=199 iy wat) (b) From the graph. 199 0, we need 6 > 0 such that if 0 < Je — 1] < 6.then [Qx+3)-5| 0, we need 6 > 0 such that if 0 < x ~ (—2)| < 6. then |(Gx+8) 2] 0, we need § > 0 such that if 0 < |x — (—3)] < 6, then (1 ~ 42) ~ 13] <¢, But |(1 42) 13) 0, we need § > O such that if 0 < Ja — 4] < 6, then \(7 32) — (-5)| 0. we need 6 > 0 such that if -< | ~ 3] < 6. then |® — 3 5 ln ~ 3] <5e. So choose § = Se. Then 0 < 2-3) <6 = |x —3] <5e z 3 z_3 5 — 5| <& By the detinition of a limit, tim = = 3. jf the definition ofa timit, tim, F = 3 =88 © CHAPTER2 LIMITS AND DERIVATIVES 20. Given e > 0, we need 6 > 0 such that if 0 < |x ~ 6] <6, then |(j +3) ~ 3] |G +3) 8] |z-6|<4e + n (= +3) =2. dim (G +3) =3 21. Given € > 0. we need § > O such that if 0 < |x ~ (—5)| < d.then |(4— $x) —7] ) = 7] 0, we need 6 > O such that if 0 < [ar — 3] < 6. then 0, we need 6 > 0 such that if 0 < |x — al < 6 then |x al <£, Sos =e willl work. 24, Given € > 0, we need 6 > O such that if 0 < | < 6: then je — | <. But Je ~ ¢| = 0. so this will be true no matter what 6 we pick. 25. Given € > 0. we need 6 > O such that if0 < Jar — 0] <8.then fz*—O| 0, we need 6 > O such that if 0 < Ja ~ 0] < 6. then | Take 6 = ¥2. Then 0 < | 3-0 <6 -0<6 > =e. Thus. lim aay 21. Given ¢ > 0. we need 6 > such that if < | — 0] < 6. then || pick § =e, Thus. lit Ja] = Oby the definition ofa limit | 0, we need 5 > Osuch that if 9-6 <2 <9.then|YT=z-Ol 0. we need § > O such that if0 < Je —2] < 6.then |(z?—4e +5) —1] 0, we need 6 > O such that iO < fe 3] < 8.then (x? +2—4)—8) Ge rtdc8 = [etd] <8 Sotake d= min {I,¢/8}. Then0 <|e—3| 0, we need § > 0 such that if 0 < Jz — (—2)| < 6. then |(x? ~ 1) — 3] < or upon simplifying we need | -5<2-2<-3 = —A| 2| <5. Sotake 6 = min {e/5,1}. ThenO <|z+2|<6 = |x—2| 0. we need > 0 such that if 0 < |x ~ 2} <6. then |z® ~8| <¢. Now 8] =|(e—2)(e? + 2x +4) |. Mle ~ 2) < 1 thatis.1 <2 <3,thena? +2044 < 3742(3)44=19 and so. -8|=|2 = 2| (2? + 2x +4) < 19 |x — 2), Soi we take 5 = min {1, 5}. then O<|e <6 > |2*-3| 2| (0? + 2x +4) < 19 =e. Thus. by the definition ofa limit 8 33. Given € > 0, we let 6 = min {2, §}. 1f0 < jr 3) < 6.then|2—3)<2 + -2<2-3<2 3 A |e+3/ <8, Alsole—3] < §.s0|2?—9] = je 43]| Jim 2? 34. From the figure, our choices for 6 are 8, = 3 — VIE and 62 = VOTE — 3, The largest possible choice for 8 is the minimum value of {61,62}; that is, 6 = min{61.62} = 62 = VOTE — 3. 35. (a) The points of intersection in the graph are (2.2.6) and (22.3. 4) ao with x1 * 0.891 and ara ~ 1.093, Thus, we can take 6 to be the 7 smaller of 1 ~ ry and 2 ~ 1. $05 = 22 1 0.093, 4 2 (b) Solving 2? +24+1=346 gives us two nonreal complex roots and one real root. which is a(e) — Q2IG+ 108 + 12356 BDte Bie)” — 12 (6) = a ee oe) 6 (216 + 108 + 123304 Bde + Bie) (©) He = 04. then 2(e) = 1.093272 342 and 6 = 2(c) — 1 ~ 0.093, which agrees with our answer in part (a) ‘Thus. 6 = a(e) — 190 CHAPTER? LIMITS AND DERIVATIVES 36. 1. Guessing a value for 8 Let e > 0 be given, We have to find a number 8 > 0 such that r 2 | je) nek St <0 |2— alan we can make C2 2 11>} = suitable, Thus, we should choose § = min {1,2e} 2, Showing that 6 works Given e > 0 we let 6 min {1,2e}. 1f0 < [x —2| < 6,then jr ~2| <1 = 1 \z=2| [2a] de €. This shows l<2e3 > m< < } (asin prt), Also ~ 2} < 26.80 ~ fea] 2 that Him (1/2) = 5 31. 1. Guessing a value for § Given € > 0, we must find 6 > 0 such that | y= — V/a| < © whenever 0< |x—a| <6. But|yz— val ‘<¢ (from the hint). Now if we can find a positive constant C such than E+ Ya > Cthen EAL < Alc «, and we take [2 ~ a| < Ce, We ean find this number by verve © restricting 2 to lie in some interval centered at a. If|z—a|<4a.then—ha Vat vavand so 6 eal < ($a vale This suggests that we let 6 = min { 5a, a, (\/fa+ va)e} 2, Showing that 5 works Given e > 0, we let 6 = min {$a ( [ya+ va)e} 10 < |r —al <6.then je—al < da > VE + Va> y/ha+ Va (as in par). Also je — |< (Jat va)eso \vE-val= ert < a =, Therefore, lim y= V@ by the definition ofa Limit 38. Suppose that linn H(t) = L. Given ¢ = 3. there exists 6 > O such that 0 < IHi<é > HO-Li<3 ¢ LAH) < b+} For0 b> Zp Por-s J. Therefore jim. 17(2) does no exist 38, Suppose that in f(2) = L. Given = 3. there exists 6 > O such that 0 < [el < 6 > |f(e)—L< 4. Take any rational number r with 0 < |r| <6. Then f(r) = 0.50 0 ~ LI < $.80.L S|L) < 3. Now take any irrational number s with 0 < |s| <6. Then f(s) = 1.50 [1 — LI < - Hence, 1 - L < $.80 Z > 3, This contradicts 1 < }.s0 lim, f(#) does not exist.SECTION 24 THE PRECISE DEFINTIONOFAUMIT 91 40, First suppose that lim f Then. given € > O there exists 6 > 0 so that 0 <|e—al <6 + f(z) -L| 0<|x—al<6so|f(x)—L] Obe given. Since lim_ f(x) = L. there exists 61 > 050 thata— 6 Osothata <2 [fle)— L| <¢, Let be the smaller of 6; and 63. Then 0 <|z—al<8 + ab: <2 10.000 + (r+3)t< & |rtal< 1 1 10.000, & le-(-3< 10 42. Given M > 0, we need 6 > O such that 0 <|x+3|<8 = 1/(e+3)'> M. Now Gott > M s @4< eo le+8l< oe Souke = = Then 0 < |je+3) <6 > 1 eae > Mes tim, AG = co 43, Given M <0 we need 6 > 0so that In < M whenever 0 < x < &: thats = e!"* < eM whenever O such that 0 f(z) > M +1 ~e. Since lim g(x) = c. there exists 62 > 0 such that 0 < |z-al<62 = |9(z)-el <1 = az) > ¢~ 1. Let 6 be the smaller of 6; and 82. Then 0 < |x—al <6 = Jl) + 9x) > (M+ 1~ 6) + (e~1) = M, Thus, lim [f(2) + (2) (©) Let M > O be given. Since lim g(x) =e > 0, there exists 5: > O such that 0 < Je —al <8, > I9(2)—el g(a) > e/2. Since tim f(x 0. there exists 5p > 0 such that 0 < Ja — al < 69 = (2) > 2M/e. Ler 5 = min {64,62}. Then <|e—al <5 + fle) g(x) > a M.so Tim f(2) o(2 (©) Let N < O be given, Since lim g(x) = ¢ <0. there exists 6: > 0 such that O < je —al <6) Io(e)—el<—c/2 = g(z) Osuch that O < le — a < by + f(@) > 2N/c. (ote thate 0) Lets = min {61,62}. Then O<|e-al2NW/e + f(2)9(2)$2 0 CHAPTER2 LIMITS AND DERIVATIVES 25 Continuity 1. From Definition 1. lim, f(x) = (4), 2. The graph of f has no hole, jump. or verti asymptote, 3. (a) The following are the numbers at which f is discontinuous and the type of discontinuity at that number: 4 (removable), —2 (jump), 2 (jump), 4 (infinite). (b) Fis continuous from the left at 2 since lim _ f(x) = f(—2). f is continuous from the right at 2 and 4 since lim, 2) = £2) and im, f(2) = f(4). Wis continuous fom nether side at—4 sine f(—4) is undefined 4. gis continuous on [—4, ~2). (—2,2). 2.4). (4,6). and (6,8). 5. The graph of y = f(r) must have a discontinuity 6 y atx =3.and must show that lim_ f(x) = f(3) : at .2,3,and 4. A person parking in the lot would want to keep in mind 2@ cot (b) There are discontinuities at times t = that the charge will jump at the beginning of each hour, Time abouts) 8, (a) Continuous; at the location in question, the temperature changes smoothly as time passes, without any instantaneous jumps from one temperature to another. (©) Continuous; the temperature ata specific time changes smoothly asthe distance due west from New York City increases. without any instantaneous jumps. (©) Discontinuous: asthe distance due west from New York City increases, the altitude above sea level may jump from one height to another without going through al ofthe intermediate values — ata cliff. for example (2) Discontinuous: asthe distance traveled increases, the cost of the ride jumps in small increments, (c) Discontinuous: when the lights are switched on (or off). the current suddenly changes between O and some nonzero value, without passing through all ofthe intermediate values. This is debatable, though. depending on your definition of current. 8, Since f and g are continuous functions, fle) — 9(0)] time) thy Limit Laws 2and 3) =24(3) ~ 9(3) [by continuity of f and g at 2 = 3) =2-5—g(3) = 10-9(3)SECTION25 CONTINUITY © 93 Since itis given that lim, [2f(2) ~ g(2)] = 4. we have 10 ~ g(3) = 4.0 9(3) = 6 P+ V7—4 = 16 + V3 = f(A). Jim 2? +, [Tim 7— Tim & ay uy my 40. lim, f(z) = lim (2? + VF=z) By the definition of continuity. f is continuous at a = 4. (2m, 2,1m, 2) = By the definition of continuity, f is continuous at -1+2(-1)"]* = (-3)* = 81 = s(-1) m, (x +225) M1. lim, f(e) = Jim x + fim 1 z+ 4 441 im g(x) = lim at et is continuous at 4, 12. fim, ata) = lim 5 2 him, — Tim T ~ 3a? —1 31 — $4) Sog is continuous at lim 2+ lim 3 4 Fora > 2. weave in f(a) = Jin BF = FE himicLaw Tein (1,2,and3] = zox8 (7and 8] = f(a). Thus, fis continuous at z= a for every a in (2,00): that is, fis continuous on (2, 00) 14 Fora <3, we have jim g(2) = Jim 2V3=F = 2 lim VE=B (LimitLaw3} = 2\/Tm(e—z) (01) [Yim 3—Tim = [2] =2V3=a_ [Tand 8] = g(a). so. is continuous atx = a for every a in = g(8). 80 g is continuous from the left at 3. Thus, gis continuous on (—o0, 3. (-00,3). Also, fim g(r ce 15. f(e) = In|x ~ 2] is discontinuous at 2 since f(2) = In0 is not defined is discontinuous at 1 because Jim f(x) does not exist 2 ifz>0 ‘The left-hand limit of f ata =Ois lim f(z) = lim e* = 1. The em mser=f * if21 lim f( (1422) = 141? =2and in f(x) lim, (4-2) =4-1=3. ‘Thus. f is discontinuous at 1 because lim, f() does not exist. a. F(x) is a rational function. So by Theorem 5 (or Theorem 7), F is continuous at every number in —2hor Pr5e+o is domain. {x | 2? + 5 +6 £0} = {2 | (w+ 3)(x+2) #0) U(-3.-2) U (-2,00). 22. By Theorem 7, the root function ¥ and the polynomial function 1 + a are continuous on R. By part 4 of Theorem 4, the product G(x) = ¥z (1 + 2°) is continuous on its domain. R. lee- (-00. = ZB. By Theorem 5, the polynomials 2? and 2x ~ 1 are continuous on (~00, 26). By Theorem 7, the root function Vs continuous on [0. 00). By Theorem 9. the composite function v2 —T is continuous on its domain, [3,90). By part 1 of Theorem 4, the sum R(z) = x? + V2 —T is continuous on (3,00) 24, By Theorem 7. the trigonometric function sin and the polynomial funetion z+ 1 are continuous on R. By part $ is continuous on its domain, (2 |" # 1} of Theorem 4. h (x) = z+ 25. By Theorem 5. the polynomial Sz: is continuous on (—90, 00). By Theorems 9 and 7, sin 5z is continuous on (00, 2¢). By Theorem 7. eis continuous on (90, 20). By part 4 of Theorem 4 the product of e* and sin 5x 1s continuous at all numbers which are in both of their domains. that is, on (20,20). 26. By Theorem 5. the polynomial x — 1 is continuous on (—00, 00). By Theorem 7, sin” is continuous on its domain, [—1, 1]. By Theorem 9, sin~? (2? — 1) is continuous on its domain, which is {z|-1s27-1<1} 2 [0 0} = {t] > 1} = {tt > 1) = (00, -1) U0), 28. By Theorem 7, \/@ is continuous on (0, 00). By Theorems 7 and 9, €V is continuous on [0, 00). Also by ‘Theorems 7 and 9. cos(ev*) is continuous on (0, 00) 28. The function y = 27; is discontinuous a 2 = O because the T+ left- and right-hand limits at 2 = 0 are different. 30, The function y = tan? x is discontinuous at x = 3 + xk, where kis ; any integer. The function y = In(tan® 2) is also discontinuous where J ] i tan? xis 0, thats, ate = mk. Soy = In(tan?®.x) is discontinuous at 2 dM 1 = $n.nany integer: \ i i i | 31. Because we are dealing with oot functions. 5 + Ys continuous on (0,06). V5 is continuous on [—5, 00). $0 Ste Vara cominuous on [0, 0). Since f is continuous atx = 4, lim, f(z) = f(4) the quotient f(2) 42. Because 2 is continuous on R. sin x is continuous on R. and 2 + sin x is continuous on R, the composite function J (x) = sine + sinzz) is continuous on R. so lim f(x) = f(x) sin =0. sin(a + sin) 33, Because 2? — a is continuous on R, the composite function f(x) =e Jim f(a) =f) =e" = 68 is continuous on R, so 34. Because arctan is a continuous function, we can apply Theorem 8, dren (SESE) = an (1 EEE 2) — a (422) arctan 2 ~ 0.588 32? ~ 6x #2 3a(x — 2) 22 32 2 ifen/4 36. f(x) By Theorem 7. the trigonometric functions are continuous. Since f(r) = sin on (~o0, 7/4) and f(2) = cos on (2/4,00). fis continous on (—o0, 7/4) U(n/4,00)- lim, f(2) = lim, _ sina = sin | = 1/V2since sata antl the sine function is continuous at x/4. Similarly. lim f(z) = lim, cos = 1/2 by continuity of the satn/ayt art cosine function at 7/4. Thus, Jim 7 J(a) exists and equals 1/-/2, which agrees with the value {(7/4). ‘Therefore, f is continuous at 1/4, so f is continuous on (~c0, 00) 1+2? ife2 3. f(c) is continuous on (—00,0), (0,2). and (2, 00) since itis a polynomial on each of these intervals. Now lim f(z) = lim (1+ 2°) and lim, f(2) = im, (2~ 2) =2, so fis discontinuous at 0, Since (0) = 1 J continuous from the left at 0 Also, lim f(2) = fim (2—<#) =0, lim, f(a) = Jim (@ — 2)? = 0, and f(2) = 0, so f is continuous at 2 ‘The only number at which. f is discontinuous is 0. z+1 ifs BB f(z)=ii/e if l<2<3 v-8 ifz23 {is continuous on (~o0, 1), (1,3). and (3,00), where itis a polynomial. a rational function, and a composite of a root function with a polynomial respectively. Now lim f(x) = lim (x +1) = 2and lim, f(a) Since f(1) = 2. f is continuous from the left at 1, Also, Tim f(z) = lim (1/2 lig, 02) = Jp, V8 = lim, (1/2) = 1, so f is discontinuous at 1 aatt 1/3. and f(3). 0 F is discontinuous at 3, but i is continuous from the right at 3 et2 ife<0 3. fiz)=e if OSe<1 Q-2 ife>1 {fis continuous on (—00,0) and (1, 20) since on each of these intervals it is ‘polynomial; itis continuous on (0, 1) since iti an exponential. Now -80 fis lim (+2) =2and Jim, f(z) = Jim, e* = et discontinuous at 0. Since f(0) = 1. f is cor lim_e* =e and inuous from the right at 0. Also tim. f(@ lim. f(2) — lim, (2 ~«) = 1,0 fis discontinuous at 1. Since f(1) = e, f is continuous from the left at 3 at autSECTION25 CONTINUITY © 97 40. By Theorem 5, each piece of F is continuous on its domain. We need to check for continuity at r = R. GMr _ GM GM _ GM tig PO) = Me aa = ae tnd Him PO) = inn, Sar = Fe GM so lim F(r) = >. Since F(R) = SM ris continaous at R. Therefore, F is a continuous function of 41. Fis continuous on (00,3) and (3,00). Now lim f(z) = tim (er + 1) = Se + 1 and lim, f(x) = lim, (cx? —1) = 9c~1. So fis continuous ¢ 3¢+1=9e-1 @ 6e=2 c=4 at at ‘Thus, for f to be continuous on (90, 00).¢ = 4 42. The functions * ~ c* and ex + 20. considered on the intervals (—oe,4) and [4, 00) respectively. are continuous for any value of ¢, So the only possible discontinuity is at « = 4 For the function to be continuous at = 4, the left-hand and right-hand limits must be the same. Now lim g(z) = tim (x? — ¢2) = 16 — c? and lim, g(x) = lim, (ex + 20) = de +20 = (4). Thus. 16-c? = 4e +20 & ce +4ce+4=0 & 2. w= 2r-8 _ (x~4)(0+2) $8. (a) f(e) = = = SAF hs a removable discontinuity at ~2 because g(2) = 2 ~ dis continuous on R and f(x) = g(c) for x —2. [The discontinuity is removed by defining f(-2) = —6.] (b) f(x) = Land lim, f(e) = 1. Thus. lim, f(2) does not exist, s0 the discontinuity is not removable. (It is a jump discontinuity.) (© fle) = 21 A(T 42 +19) asa remouable discontinuity at ~4 because a(2) = 2? ~ 4x + 16s continuous on R and (2) = g(z) for-x # —4, [The discontinuity is removed by defining f(—4) = 48.) @ f@)= 3- ve 2 > Tae Va Mareen scominaty a9 becase g(a) = =i continuous on R and f(z) = g(x) for x # 9. [The discontinuity is removed by defining f(9) = £.] 4, » 3 a[w=2 1 F does not satisfy the conclusion of the 4 does satisfy the conclusion of the Intermediate Value Theorem, Intermediate Value Theorem, 45. f(x) = 2° — 2* + xis continuous on the interval (2,3). f(2) = and f(3) = 21. Since 6 < 10 < 21. there isa ‘umber c in (2,3) such that f(c) = 10 by the Intermediate Value Theorem, 46. f(z) = 2° is continuous on the interval [1.2]. f(1) = 1. and f(2) = 4. Since 1 < 2 < 4, there is a number cin (1,2) such that f(c) = ¢? = 2 by the Intermediate Value Theorem,98 (0 CHAPTER2 LIMITS AND DERIVATIVES 41. f(x) = x4 + x — 3s continuous on the interval (1, 2), f(1) = —1, and f(2) = 15. Since -1 <0 < 15, there is a number cin (1,2) such that f(c) = 0 by the Intermediate Value Theorem. Thus, there isa root of the equation (in the interval (1,2) 4% + x — 1is continuous on the interval [0, 1], f(0) = —1. and f(1) = 1. Since —1 < 0 < 1, there isa umber c in (0, 1) such that f(c) = 0 by the Intermediate Value Theorem. Thus. there is a root of the equation YE +2 —1=0.0r YF = 1 — 2. in the imerval (0,1) 49. f(x) = cos.x — 2 is continuous on the interval [0,1]. f(0) = 1. and f(1) = eos 1 — 1 = -0.46. Since 0.46 < 0 < 1. there is a number c in (0,1) such that f(c) = 0 by the Intermediate Value Theorem. Thus, there is 4 root ofthe equation cos.x ~ x = 0, or cos = =. in the interval (0, 1) 50. f(t) = Inx — e~* is continuous on the interval (1,2), f(1) = ~e~! = -0.37, and f(2) = In2~ e°? = 0.56. Since —0.37 <0 < 0.56, there is a number ¢ in (1, 2) such that f(c) = 0 by the Intermediate Value Theorem. ‘Thus, there is a root of the equation In x — e~ sf 0, or In: = e~*. in the interval (1,2) 51. (a) f(x) = e* +a — 2is continuous on the interval (0, 1], f(0) = —1 < 0.and f(1) = e ~ 1 © 1,72 > 0, Since 1 <0 < 1.72, there is a number c in (0,1) such that f(c) = O by the Intermediate Value Theorem. Thus, there is a root of the equation e® +2 ~ 2= 0. or e? = 2~ a, inthe interval (0,1) (b) (0.44) = -0.007 < 0 and f(0.45) ~ 0.018 > 0, so there is a root between 0.44 and 0.45. $2 (a) f (ae) = 2° — 2? + 2x +3 is continuous on [-1,0], f(=1) = —1 < ,and f(0) = 3 > 0. Since -1 <0 <3, there is a number c in (1,0) such that f(c) = 0 by the Intermediate Value Theorem. Thus, there is a root of the equation 2° ~ 2 + 2x +3 =O in the interval (~1,0) (b) (0.88) ~ -0.062 < 0 and f(—0.87) ~ 0.0047 > 0, 50 there is a root between —0.88 and —0.87. 53. (a) Let f(x) — x? — 4. Then f(1) = 1° — 1? 4 = —4 < Oand f(2) Intermediate Value Theorem, there is a number ¢ in (1,2) such that f(c) 28 — 2? —4= 24> 0. Soby the &-8-4=0. (b) We can see from the graphs that, correct to three decimal places, the root is = 1.434. | LA -05 1 < 0 and f(6) = § > 0. and fis continuous on [5, 00). So by 1 5A.) Let fle) = VE=B ~ Sy Then f(6) the Intermediate Value Theorem. there is @ number c in (5, 6) such that f(c) = 0. This implies that 1 = ve-5. ors (b) Using the intersect feature of the graphing device, we find that the root of the equation is « = 5.016. correct to three decimal places,SECTION25 CONTINUTY © 99 55. (=) If f is continuous at a, then by Theorem 8 with g(h) = a+ h, we have fim f(a+h) = F( im (a+ 4) = F(a) F(a), there exists § > Osuch that 0< |h| <5 = (@)Lete > 0. Since fim f(a+h) [fla-+h) = f(a)| <=. Soif0 < |x ~ aj < 6, then [f(x) — f(a) = [flat (@ —a)) — fla) 0, the interval 1 iis irational (a 6.a-+ 8) contains both infinitely many rational and infinitely many irrational numbers. Since f(a) there are infinitely many numbers x with 0 < |x — al <6 and |f(«) — f(a)| = 1. Thus, lim (2) # f(a). Un fact. tim, f(«) does not even exist © if wis rational 60. g(x) = { “P iffzis tational ©COMUUOUS AO. To see why. note that ~ jx] < g(r) < [xl.so by the Squeeze ‘Theorem lim g(x) = 0 = (0). But gis continuous nowhere else. For if a # 0 and 6 > 0, the interval (@~ 6,04 8) contains both infinitely many rational and infinitely many irrational numbers. Since g(a) ~ 0 or a, there are infinitely many numbers x with 0 < lar ~ a < 6 and |g(e) ~ g(a)| > lal /2. Thus. lim 9(z) ¥ g(a). 61. Ithere i such a number, it satisfies the equation a3 41=2 ¢> 2*—241=0. Let the left-hand side ofthis equation be called f(«). Now f(—2) = —5 <0. and f(—1) = 1 > 0. Note also that f(2) is a polynomial, and thus continuous. So by the Intermediate Value Theorem, there is a number ¢ between —2 and —1 such that S(e) = 0. so that e = c8 +1100 9 CHAPTER? LIMITS AND DERIVATIVES 62, (a) lim, F(z) =O and linn (=) = 0.so lim F(x) =0, which is F(0)-and hence F is continuous atx = a i Fora > 0. lim P(e) = Jim 2 =a = F(a). Fora <0, lim P(e) = fim (—2) = ~a = F(a). Thus, F is continuous at r = a; that is, continuous everywhere. (b) Assume that f is continuous on the interval I. Then for a € I, lim | f(z) $(2)| = sledl by ‘Theorem 8. (If ais an endpoint of Fuse the appropriate one-sided limit.) So || is continuous on 1 ife>0 (c) No. the converse is false. For example, the function f(x) = is not continuous at x = 0. but -l ife<0 |f(2)| = 1 is continuous on R. 63. Define u(t) to be the monk’s distance from the monastery, as a function of time, on the first day nd define d(t) to be his distance from the monastery. asa function of time. on the second day. Let D be the distance from the monastery to the top of the mountain. From the given information we know that u(0) = 0, (12) = D. d(0) = D ‘and d(12) = 0. Now consider the function u — d. which is clearly continuous. We calculate that (u ~ d)(0) = and (u ~ d)(12) = D. So by the Intermediate Value Theorem. there must be some time fo between () and 12 such that (u—d)(to) =0 4 u(to) =d(to). So at time to after 7:00 A.M., the monk will be at the same place on both days, 2.6 Limits at Infinity; Horizontal Asymptotes 1. (a) Asx becomes large. the values of f(x) approach 5. (by As a becomes large negative, the values of f(x) approach 3 2. (a) The graph of a function can intersect a ‘The graph of a function can intersect a horizontal asymptote vertical asymptote in the sense that it {can even intersect its horizontal asymptote an infinite can meet but not cross it ‘number of times. =e (b) The graph of a function can have 0, 1, or 2 horizontal asymptotes. Representative examples are shown, — No horizontal asymptote One horizontal asymptote Two horizontal asymptotesSECTION 26 LIMITS AT INFINITY, HORIZONTAL ASYMPTOTES 101 3. @) lim f(x) = 00 (b) © im, f(@) @ Jim f(z)=1 ()_lim_ f(x) =2 (f) Vertical: x = 1, ¢ = 2; Horizontal: y = Ly = 2 4.(@) Jim g(z) =2 () tim g(2) = -2 (©) lim g(z) = 00 (4) lim 92) = -00 (f) Vertical: x = -2, ¢ = 0,2 = 3: Horizontal: y = ~2,y = 2 5. f(0) =0. Sisodd J) =1, im f(x) =0. & Jim, f(2)=c0. lim f(z) = 2 a fle)=1 7. tim f(2) = ~00. im, f(x) = 00. lim f(z) = 3. 8. If F(2) = 27/2", then a calculator gives f(0) = 0. f(1) = 0.5. f(2) = 1. (8) = 1.125. f(4) =1 £0) = 0.78125. (6) = 0.5625. (7) = 0.3828125, f(8) = 0.25. f(9) = 0.158203125. f(10) = 0.09765625. F(20) = 0.00038147, (50) ~ 2.2204 x 10-2. (100) = 7.8886 x 10-27, Mappears that lim (2°/2*) = 0, 10. (a) From a graph of f(x) = (1 ~ 2/)* ina window of [0, 10.000} by [0.0.2]. we estimate that lim f(x) = 0.14 (to two decimal places.)102 OCHAPTER2 LIVITS AND DERIVATIVES o From the table, we estimate that lim f(z) = 0.1353 = Je) (to four decimal places.) _ 10,000 | 0.135308 100,000 | 0.135333 1,000,000 | 0.135335, , : , {divide both the numerator and denominator by 2? th tim 32 =2*4 «jg CP = 8+ Ole dim gayes se 7 AM, Geta be 8)? (the highest power of z that appears in the denominator)} Jim (3-1/2 + 4/2") ~ Tim OF 572-87) [Limit Law 5] lim 3— lim (1/2) + lim (4/2”) = OF ie GT me as [Limit Laws 1 and 2] im 2+ Tim (6/2) ~ Jim 8/2") 3— Jim (1/2) + 4 Jim (1/2?) ~ TFS lim (Je) 8 jim (17) [Limit Laws 7 and 3) 3-0+400) ~ 2+5(0) — 8(0) 3 [Theorem 5 of Section 2.5] 2 1225 — 5a +2 12a” — 5a +2 [rat ar eae Limit Law 11 12 Jim VT pat raat ~ Va yaa oe L = Pim 25/2 + 2a 3 A Te +443 [eivide by =") Tim (12 — 5/27 +2/2°) Laws © \ Tim O72 ¥ a7 +3) (Limit Law 5) lim 12— ‘lim (6/27) + Jim @/2*) Bae) + i /=) + 9 12-5 Jim (1/2) +2 im (0/2) swe 7 and 3 =| Rs [Limit Laws 7 and 3] _ fi2=5(0) + 210) a =a [Theorem 5 of Section 2.5] te dm) lim (1/2) m 243 lim (Ja) 2+3(0) 2 {Limit Laws 1 and 2} 1 = —___ 18. Jim. 5553 Get Se Jim 2+ 3/a)SECTION 26 LIMITS AT INFINITY, HORIZONTAL ASYMPTOTES C108 1 lim 345 lim = S245 Bets/e_, s+5/e_ jms+Siim > 3450) “tn aad 7 i Gaye ~ ape im 124 im 2 1=4(0) ar lim (1/2? — fa 1) Tim_@— 17) Cee 1 _ o-0-1_ 1 ~ lim _2—7 Tim (1/27) ~~ an? 2 lim @/v?—3) 2 im (1/4?) ~ tim 3 _ $6. tim 22387 = ji, 2 = 80 V/y? yt wise? _ 2(0)-3 yatte By? + dy yt (Sy? + dy)/y? ine) ~ dim 54 im (17a) 5 +400) 17. Divide both the numerator and denominator by 2° (the highest power of 2 that occurs in the denominator). lim 145 jim eee 1450) _ igo in Fe im Se TEAOD P42 (P+ 1t+2/8 40 tim, oO A2 ty LEHANE yy tay? 0-0 lin rae ~ lee olin, PE = EO 19. First, multiply the factors in the denominator. Then divide both the numerator and denominator by u dut +5 5 Bim py $5 a, Ae WFD) Bae bute A Gees = lim py — Qa ts 5 1 tim (44 ares - din, a) He 44500) tm (2-542) ~ Jig? -5.hm Ee ig EHO wim (2- gate) li? - 5 tim 5 +2 Lim J 20. tim —2+2_ lim, EAP i 1+2/e _1+0 1 2 YOST oh acta Vat aie VOrO Je VEER Ej? lim Jae lina = im, VO elt OT since? = VB fore 7 i el > eras Tim (1+ 175) Isince 2? = Va8 for x > Oo} lim (O-175 _ VE oe) fm + Tim (7) 0 =V=0 3104 CHAPTER? LIMITS AND DERIVATIVES varayet _ im, VORA 2. Ii = lim 5 ate aT Ti a De lim +178) Usince 2° = —Va for 2 < 0] lim, - VOT =f Tim 9 Tim 79) > Tim 1+ Tim (729) 1+0 =-vo=0 2. lim (V9 Fa — 32) = lim (VEPER — s0)( VTE +82) _ 5, (VO FA)"— 2)? Viet te + Se on Yea y at oe = jim Oz) = 92" Ye ~ Vora ate = Irate Ve afr 1 = tim ——— 2/2 _____ = tim ——4+__ = J, Sears a dale JOR ETS Whim. (e+ VF BE) = lim, (x + VT FB) [See vee | = oy Vere 20 =2 -2 = — 2 = tim —S— = — = site go Vartde tlt Vitae 1+ V1+20) Nore: In dividing numerator and denominator by 2. we used the fact that for 2 < 0,2 = ~via? 2B. im, (Ve Faw ~ VT oe) = im, eee vere ee aes vert be) = tim tae) = (2? + bz) hi, [(a — bal /a. a ea Fart ae Wit part Ve eta) Ve a-b a-b lim, © ae rae aa = iso vivo 26, lim cos:r does not exist because as z increases cos does not approach any one value, but oscillates between and =1 27. Sais targe when 2 is large. so lim, V% = 90 28. 3/2 is large negative when is large negative. 80 28. im (@ ~ VE) = lim, VE (VF ~ 1) = oosince Ya + oo and. — 1 —+ 00 a8 2 — 00. iy, (PEE vey eee poner ane emi = —0o because 2 — 2/2 + 3/2? + 00 and 5/2? — 2+ ~2asz + 00. —o0 because 2° + 00 and 3 tim (et +2") lim_ 2°(2 + 1) [factor out the largest power of 2] Ae 1 = Las 2 + —00.‘SECTION 26 LIMITS AT INFINITY, HORIZONTAL ASYMPTOTES 105 32, im tan" (2? — 2") = him tan“! (22(1 ~ 24). Ife lett = 2(1 — 22). we know that t+ —90 a8 2 + 00, since x? — oo and 1 — 2” + —oo, So lim tan”! x?(1— 27) im tan-?¢ = —. Peat) Tafa 1a? +Afeee _ "a Vee because (1/a° + 1/a+ x) — oo and (1/2* ~ 1a? +1) + Las 2 +00. [divide by the highest power of z in the denominator] 34 If we let ¢ = tan z, then as © + (n/2)*. t+ oo. Thus. Him, e0"*= tim et 35. (a) ~100 ° ) z F(x) ~10,000 | —0.4999625 —100.000 | ~0.4999962 ~1.000.000 | 0.499996 From the graph of f(x) = Va? Fet1 +2. we From the table, we estimate the limit to estimate the value of lim_ f(x) to be —0.5 be ~0.5. eta+l—a ©) lim, (VERE +2) = tim (VEFEET +2) eee im + VOe) te (Vata FI —a)(l/a) _ 140) ~ SisF0=1 2 Note that for x < 0, we have Va? 1+ (1/2) —V1+ (72) + (172) —1 ii = a, so when we divide the radical by ©. with x < 0, we get 1 1 gVU 4EF1 =~ tet =-/1 tay FU). 36. (a) Ls. (b) z f(z) 10.000 | 1.44359 100.000 | 1.44338 1,000,000 | 1.44338 oy 100, From the graph of From the table, we estimate (to four decimal S(@) = V8FF REE ~ VIF FBEFT, we estimate __PISCES) the limit wo be 1.4434. (to one decimal place) the value of lim f(.r) to be 1.4.106 CHAPTER? LIMITS AND DERIVATIVES (©) tim fle) = tim (VERE EEO Veer Be FT) (VEE Be tO + Vat + Se 1) eee aoe V3e? +82 +6 + V8x? +3x+1 (30? + 8x +6) — (32? + 3041) ues Vie? + 8246+ V3x? +32 +1 = tim et) 2 (Viste Bo 0+ V3x" + Se + 1) (La) lim, S+5/z 5 _ 5 58. aass76 Breet oe+ Vst3jerie Va+Vv3 2Vv3 oe 3. lim 1,80 y = Lis a horizontal 2 = lim ol g4d ~ etfs 14 4/e 140 wan asymptote, lim = ——> = lim, 2, = -00.s0n = ~4is | ymptote. im = = oo and Tim 55 sor=—dis P jl, ‘a vertical asymptote. The graph confirms these calculations. 4 38. Since 2? — 1 + Oas a — +1 andy <0 for —1 << Landy > 0 for a <-Land 2 > 1.webave tim 24 = -o0, tim, G*4 =o, ; aim a 7 6 6 wr ara lim covand tim 54" = sai a1 aati e are vertical asymptotes. Also era L44/o* _ 140 im 148 <1 soy=1is eRe gt 1 2tkeo = 1a! aad “ 1-0 horizontal asymptote. The graph confirms these calculations, 2 im pe ti aay = E00: Ho there is 38. lin age cio 7 «lie TEs) aloes) ~ BO 40 10 horizontal asymptote 7 a a im, Ee = im, Ee = 00-8 - wim 3730700 ~.. Gra@-y 12 0 a - 2 similarly. lim, ——S— I ere Tae > 2, Similarly. jim ay3e>10 Lu lin a eeap = Nara t= 2 and x = —5 are vertical asymptotes. The graph confirms these calculations,SECTION 26 LIMITS AT INFINITY; HORIZONTAL ASYMPTOTES © 107 2 rtd L+1/2* 0 Jina Type? e+1_ e+ Pye a@+1) =1.s0y = Lisa horizontal > O fore > Oand y <0 for asymptote. Since y 2 2 B41 e4l _ lim, coand lim ses 0 pe sat he a vertical asymptote. 00. sor = 0is 80 y = +1 are horizontal asymptotes. There is no vertical asymptote. @ tim —2=9__ ue ante eT % Ji Gla) +O/e) VaFOTO 2 Using the fact that Vz? wv for 2 < 0, we divide the numerator by —2 and the denominator by V2, Thus, lim —=222 a4 9/e 140 1 Me Jia 8043 2h TET la ae © VEOH 2 The horizontal asymptotes are y = 4. The H——— 1 s polynomial 4z* + 32 + 2is positive for all 2, 40 *” so the denominator never approaches zero, l2| and thus there is no vertical asymptote 48. Let’s look for a rational function, ot aoe F( sign change at = 0). and the function is negative near a = 0, {(e)=0 = degree of numerator < degree of denominator 2 00 => there i a factor of 2? in the denominator (not just, since that would produce a @) tim f(e) = coand lim, f(z) 00 = vertical asymptote at x = 3: there is a factor of (« — 3) in the denominator ( F2)=0 => isan a-intercept; there is atleast one factor of (2 ~ 2) in the numerator. Combining all of ae information and putting in a negative sign to give us the desired left- and right-hand limits a 8 one possibility aives us f(z) = 2108 CHAPTER? LIMITS AND DERIVATIVES. 44, Since the function has vertical asymptotes « = 1 and « the denominator of the rational funetion we are looking for must have factors (x ~ 1) and ( ~ 3). Because the horizontal asymptote is y = . the degree of the numerator must equal the degree of the denominator, and the ratio of the leading coefficients must be 1. One possibility : SO = Gea = f(x) = 2%(w — 2)(1 — 2). The y-intercept is f(0) = 0. and the z-intercepts occur when y=0 => 2 =0. 1,and 2. Notice that, since 2° is always positive, the ‘graph does not cross the 2-axis at 0, but does cross the z-axis at 1 and 2 lim 2% 2)(1- 2) oo, since the first two factors are large positive and the 100 because 2a third large negative when = is large positive. Tim <7 the first and third factors are large positive and the second large negative as > —00. 46. y = (2 +.2)°(1 — 2)(3 — 2). As a — 00, the first factor is large positive, and the second and third factors are large negative. Therefore, lim f(s) = 00. AS z+ ~oo, the first factor is large negative, and the second and third factors are large postive. Therefore, lim f(2) F(0) > p= -21and 0, Now the y-intercept is (2)8(1)(3) = 24 and the «-intercepts are the solutions to f(x) = 0 and the graph crosses the z-axis a all of these points 41. y = fle) = (e+4)5(a— 3). The y-intercept is f(0) = 4°(—3)* = 82.944. ‘The z-intercepts occur when y= 0 => = ~4,3, Notice that the graph does not cross the x-axis at 3 because (x ~ 3)* is always positive, but does cross the z-axis at —4, lim (2 + 4)°(a — 3)* = 00 since both factors are large positive when zis large positive, _lim_(x + 4)®(ar ~ 3)* = ~co since the first factor is large negative and the second factor is large positive when zis large negative. 48, y = (1 —2)(2 — 3)*(@ — 5)?. As. —+ 00, the first factor approaches —o0 while the second and third factors approach oo, Therefore. lim (x) = —o0. As + ~00, the factors all approach oo. Therefore, (a) = 00. Now the intercept is f(0) = (1)(~3)°(-8) to f(2) =0 = 2 =1.3,and 5, Notice that f(r) does not change sign at = 225 and the z-intercepts are the solutions a= Bor = 5 because the factors (a ~ 3)? and (x ~ 5)? are always positive, so the graph does not cross the z-axis at = 3 or x = 5, bat does cross the a-axis ate =1. Losing 21 ince —1 < sing < 1 forall x, > < =" <7 49, (a) Since -1 < sin < 1 forall 2, > <== <7 Squeeze Theorem, (sin:z)/2—> 0. Thus, Jim, “= forz > 0. AS + 00, -1/z — Oand 1/a — 0, so by theSECTION 26 LIMITS AT INFINITY, HORIZONTAL ASYMPTOTES 109 L (b) From part (a), the horizontal asymptote is y = 0. The function 1 = (sin) /« crosses the horizontal asymptote whenever sin x = 0; that is, atx = an for every integer n. Thus. the graph crosses the ~2s 2 asymptote an infinite number of times. 05 50. (a) Inboth viewing rectangles, fim P(x) = lim Q(z) = oo and_lim_ P(x) =_lim_Q(z) = -oo. Inthe larger viewing rectangle, P and Q become less distinguishable. P and Q have the same end behavior. 51. Divide the numerator and the denominator by the highest power of z in Q(z). (@ Ideg P < deg Q. then the numerator — 0 but the denominator doesn't, So lim [P(2)/Q(z)] = 0. () Ifdeg P > deg. then the numerator — too but the denominator doesn't, so lim [P(2)/Q(z)] = too (depending on the ratio of the leading coefficients of P and Q). aN (n=0 Gi) > 0 (modd) (iii) > O (even) (iv) n < 0 (nod) From these sketches we see tha () n < O(neven) ee 1 ifn=o0 0 itn @ lim2"=40 ifn>o (b) lim 2" = Li aoe co wen a =o ifn <0. nodd co ifn <0. neven 1 ifn=o 1 ifn=o0 co ifn >0, nodd oo ifn>o (tim 2” sate co if n>0, neven 0 ifn ay.thene*/? < 0.1 (by en

#1 and y = 0-1 intersect at 1 = 23.03. =2/10 > Win = —10n 10" = 101n10 24 be — 57. <02 & 28< SH t8-3 3250 WaT wwe graph the three pats of this inequality on the same sereen, and 6x7 + 52-3 ind that the c S274 52 — 8 coms to lie between the lit find that the curve y = SF +52 = seems tie between the ines y = 28 and y = 3.2 whenever 2 > 12.8, So we ean choose N’ = 13 (or any larger number) so that the inequality holds whenever « > N. 58, Fore = 0.5, we must find N’ such that whenever: > NV. we have aes | 3, So we choose N= 3 (or any larger Viet FT 2.1, and rth a number). For € = 0.1, we must have 1.9 < the graphs show that this holds whenever x > 19. So we choose N = 19 (or any larger number). 7 = 80. So the salt concentration approaches that of laSECTION 26 LIMITS AT INFINITY, HORIZONTAL ASYMPTOTES C411 Via 59, For = 0.5. we need to find N such that al = (-2 <05 -20 0 y 25 < SET <-1.5 whenever 2500. So we choose 60. We need N’ such that 100 whenever > N. From the N= 2500 (or any larger number) fi 5000 61. (@) 1/2? < 0.0001 < a? > 1/0.0001= 10.000 = 2>100 (r>0) (b) Ife > Ois given, then I/a?1/e a> I/Ve.LetN 1 Tnz>N = a> s | ve 4| 4 1/0.0001=10' & 2> 10° (b) Ife > Ois given, then /yaIfe x>1fe.LetN = 1/e? <..s0 lim ve soo VE Trena>N = 2>4 = 68 Fore <0. l/r —0| = ~1/x. fe > Ois given. then 1/2 2<-1fe + |(I/x)—0| -lfe 0.weneed.N > Osuchthatz>N = 2°>M.Now29>M © 2> YM.souke N= VM.Thena>N= YM > 2°>M.so lim 2? ©. $5. Given M > 0, weneed N > Osuch that > N = e® > M.Nowe™>M 4 >In Aso take N= max(1,In M). (This ensures that N > 0.) Then 2 > N= max(1.iuM) =e > max(e,M) > M, s0 Jim €* = oo.172 CHAPTER2 LIMITS AND DERIVATIVES 66, Definition Let f be a function defined on some interval (—o0,a). Then lim _ f(z) = 0 means that for every negative number MI there is a corresponding negative number N such that f(r) < M whenever 2 < N, Now we use the definition to prove that lim_ (1 +.2°) = —co, Given a negative number IM, we need a negative number Nesuch that 0 there is a corresponding positive number 1V such that fla) — L| N.It=A/z.thene>N e O<1/e O there is a corresponding 6 > 0 (namely 1/N) such that |f(1/t) ~ L| <€ whenever 0 0 there is a corresponding negative number 1V such that lw) — L| 0 there isa corresponding § > 0 (namely —1/NN) such that |f(1/¢) ~ L| mes jiya=0 = &) @ 1 > Tiss 772 (i) =m 2 4 Given m= Gee e ng 12. (a) Using (1). (l42+2%) -(1+ata’) a+r?-a lim Trt (e=a)eta) in eg =a eo — cn = tim S2OU+244) im 142 a) = 1420 fg aaa ca & @e=-1 = m=142(-1) © 3s (ie +2(-2)= yeltate Giy2=1 > m=142) 2 £125SECTION 27 TANGENTS, VELOCITIES, AND OTHER RATES OF CHANGE 115 13. (a) Using (1), (2 4241) = (0 4041) (a —a)(z? + ax +) an za 1a? — 4 +ar+a’—4)= = dm 6 -z-1) B(e-2) = —1. so an equation of the tangent line is y ~ (~2) 3(2)? ~ 4 = 8, so an equation of the tangent line is y — (b) At (1, -2): m = 3(1)? 1. At(2, 1): m_ 14. (a) Using (1). = tim YE_Va Vit _ 5, (Ja~ v8) (a+ ve) ee ena za @—a ssa Van(@—a)(yat+ yz) 41 . 1 “IEA Yaee= var ve) ~ 1 Yasar ve) ~ VERVE) (b) At(L.1): m= 3. $0 an equation of the tangent line is y — At (4,3): m 3.0 an equation ofthe tangent line is y — © 2 15. (a) Since the slope of the tangent at t = 0 is 0. the car's initial velocity was 0 (b) The slope of the tangent is greater at C' than at B. so the car was going faster at C. (©) Near A, the tangent lines are becoming steeper as x increases, so the Velocity was increasing, so the car was speeding up. Near B. the tangent lines are becoming les step, so the car was slowing down. The steepest tangent near C'is the one at C. so at C the car had jus finished speeding up, and was about to start slowing down. (@) Between D and B. the slope of the tangent is 0, so the car did not move during that time.116 CHAPTER? LIMITS AND DERIVATIVES 16. Let a denote the distance traveled from 1:00 to 1:02, b from 1:28 to 1:30. and c from 3:30 t0 3: times are relative to . where all the distance in miles 200-+a+b +e 2040+ b time in hours 17. Let s(t) = 40 — 16¢?. mm 1982) tiny (40t — 164) — 16 =1642 +40t—16 _, —8(2t? ~ St +2) s)- 6) 5 ater + E16 v(2)=1 im, iim (2)= fim yaa bm G2 f= = tim TSE“ DEED = -8(3) = — = lim = -8 Jimy(2¢— 1) = -8(8) = ~24 ‘Thus, the instantaneous velocity when £ = 2is ~24 t/s 48. (a) v(1) = Jim H+ A= HO) nm h 58 + 58h — 0.83 — 1.66h ~ 0.83h?) ~ 57.1 = Jim SBSH = 08 = Lo OE) = ETT = Jim (56.34 — 0.88h) = 56:34 m/s H(a+h)- (6) v(a) = jim Hla+ = HO) vi ($84 58h — 080? 1.66h — 0.83) — (580 — 0.880") na h = jim, (58 ~ 1.660 ~ 0.838) = 58 — 1.66a m/s (©) The arrow strikes the moon when the height is 0. that is, 58¢— 0.8317 =0 > {(58— 0.834) =0 a 2 69.9 s (since t can't be 0). 58 m/s. Thus, the arrow will have a velocity 58 (2) Using the time from part ().» (%) of —58 m/s (a+ h)—s(a) _ yy, Math) + 6a +h) +2- (da? + 6a +2) h = eb h da? + 120%h + 12ah? + 4h® + Ga + 6h +2 ~ 4a° — 6a = 2 h 12a2h + 12ah? + 4h® + Gh = fig Pe EEE = fit (1202 4 12h + 4h? 4-6) = (12a? +6) m/s So v(1) = 12(1)? +6 = 18 m/s, v(2) = 12(2)? + 6 = 54 m/s, and v(3) = 12(8)? + 6 = Ld m/sSECTION 27 TANGENTS, VELOCITIES, AND OTHER RATES OF CHANGE 117 20, (a) The average velocity between times t and t+ his s(t+h) ~ s(t) _ (t+ h)? —8(¢ +h) +18 ~ (1? ~ 8+ 18) (+h) =t h P+ 2th +h? ~ 8t-8h+18—#+8t—18 _ Ces =(2t+h-8) m/s @ [B.4): C= 3.A=4-3=Lsothe average (il) (3.5.4): t= 3.5.4 velocity is 2(3) + 1-8 = -1 m/s. is 2(8.5) +05 — 0.5. so the average velocity -0.5 m/s. Gili) [4,5]: €= 4, b= 1, sorthe average velocity is iv) (4, 4.5]: t= 4, 2(4) 41-8 = 1 m/s. 2(4) + 0.5 — 8 = 0.5 m/s = 0.5, so the average velocity is (6) v() = fim mh son = fim Qt +h ~8) = 24-8, so (4) =0. o 4 21. The sketch shows the graph for a room temperature 2. The slope of the tangent (that is, the rate of of 72° and a refrigerator temperature of 38°, The change of temperature with respect 0 time) at initial rate of change is greater in magnitude than the #= Lh seems to be about rate of change after an hour, Temperature re “on 79-115, a) 90-115, 22-20 2-115, “Baap = 13 °C/h (b) In the figure, we estimate A to be (18, 15.5) and B as (23,6). So the slope is 23. (a) (i) 20,23} = ~12°C/h 1 i) (20, 23} =1.25°C/h (ii) (20, 29 0 2 We m2 a ot 1.9°C/hat 8:00 P.w.118 © CHAPTER 2 LIMITS ANO DERIVATIVES ~). P(1996) — P(19 1 = 24) 6 (192,195; PO) — PLU) _ 105210008 _ 6 ound 1996 — 1992, 4 P(1996) — P(1994) 10,152 — 10,109 Gi) (1994, 1996} fe ed 10.452 10.108 _ B — 21.5 thousand/year _ 10,175 = 10.152 2 2 2154115 2 = 11.5 thousand/year Gi) (1996, 1998]; POS) (b) Using the values from (ii) and (ii), we have (c) Estimating A as (1994, 10.125) and B as (1998, 10.182) the slope at 1996 is 182 — 1 Ey 10.182 = 10.125 _ ST _ 14.5 thousand year. 1998-1991 4 won WG oH BOF 201873 _ 1885 _ a4 aaa (1997) — N(1995) _ 25. (a) (i) (1995, 1997): To poe i (1995, 1996} Ncgeo) — NCO) ns = 640 thousand/year Gin (3994, 10955, M0908) = NUGHA) _ $78 572 — 501 ous yar (by Using the values rom Gi) and Gi), we have MESO _ 952 — 470.5 thousand/yea. (©) Estimating A as (1994, 420) and B as (1996, 1275), the slope 44m thousants) 1275 — 420 a 1995 is, = 427.5 thousand/year 1996 — 1994 2 oo Woo se . (1998) — N(1996) _ 1886 = 1015 _ 871 _ . 26. (@) 6 (1996, 198}; LOTR EE = SE = oy = 486.5 locations /year gogy, N1998) = NGU99T) _ 1886-112 _ 474 Igcations/ye a (2997, 1998}; MGR) I = EET = a7 locations /yeur N(1999) — N(1998) _ 2135 — 1886 a (ii) (1998, 1999]: 55998 = 249 locations /year 414-2280 _ PY _ p15 = gotoaton ye i) and (iii). we have (b) Using the values fromSECTION 27 TANGENTS, VELOCITIES, AND OTHER RATES OF CHANGE C119 (c) Estimating A as (1997, 1525) and B as (1999, 2250). the slope 2250 — 1525 _ 725 a0 Tiga7 = a = 3625 locations year. at 1998 is, 3% 97 Nw amo AC _ C(105) ~ C(100) _ 6601.25 ~ 6500 aa =e > $20.25 /unit. AC _ (101) ~ C(100) _ 6520.05 ~ 6500 _ ; (i) = Sots = 7 20.05 /unit (by C{100+ h) ~ C100) _ [5000 + 10(100 + h) + 0.05(100 + h)?] — 6500 _ 20h + 0.05h? z = = h h = 20+ 0.05h.h #0 C(100 + A) — C1100) So the instantaneous rate of change is im - = fim (20+ 0.05h) = $20/unit 2. AV = Vit +h) - v(t) = 100000 (1 - +)” _ r00.000 (1 £.)" ~ ~ 60 ° © 60 2 2 2 = 1ooo00 [ (1-64 Sao) ~(1~ 35+ segs) = 1000( 30 * 3605 * Stu) S 10. (120 +2+h)= 201 (120 + 20 +h) Dividing AV by hand then letting h ~+O. we se thatthe instantaneous ate of change is 28 (¢ — 60) gal/min, t_[ Flow rate (gal/min) | Water remaining V(2) (gab 0 333353 100, 000 ol 27777 9,444.1 20] 29022 4.444 30} ~ 16665 25,000 sof st nT 50 = 5555 2777 60 o 0 ‘The magnitude of the flow rate is greatest at the beginning and gradually decreases to 0.to © CHAPTER2 UMTS AND DERWATIVES 2.8 Derivatives ‘The line from P(2, f(2)) 0 Q(2 +h. f(2+h)) L2+h) = f2) h f+ h) =f) isthe line that has slope f2+h) ‘ / 2 tat 2. As h decreases, the line PQ becomes steeper. s0 its slope increases. So < LM=10) i. £@)= f2) ¢ 3-2 HOLE) thus.0< 4 (4) ~ £2)] < £8) - f2)< 3. g/(0) is the only negative value. The slope at = 4 is smaller than the slope at 2 = 2 and both are smaller than the slope at 2 = ~2. Thus. g/(0) <0 <9'(4) <'(2) —g.7q = 0:08 0.001 | 2.0020 |[ -0.001 | 1.9980 0.0001 } 2.0002 || 0.0001 | 1.9998 We estimate that g' () = 2 13, Use Definition 2 with f(ie) = 3 ~ 22 + 4a" ni) — tim feat h) = fla) _ [3 2(a+h) + 4(a + h)?] — (3-20 + 4a") S10) = fam = fig (3-20 — 2h + 4a? + 8ah + 4h?) ~ (3-20 + 40?) —2h+ Sah + 4h? h(=24 80+ 4h) _ ag _. = im, i sim MORE REED = jim (—2 + 8a. dh) = 2+ Ba (a+ h)* — 5(a + h)] — (a* — 5a) 4. (a) = fim in, Hoth) fla) _ = jim Lt )* = 5a : )} = (a4 ) pe, (at + dah + 07h? + 4ah® + h* — 5a—5h) a) = i ah + 6a7h? + dah’ + h* —5h (4a? + 6a7h + dah? + h® — 5) =lim toe inl dat Ole tdci ms h i = fin (4a? + 602A + da? +H? —8) = Ae? — 5 15, Use Definition 2 with f(¥) = (2t-+ 1)/(t+3) Aath)+1 20+) lim fla+h— fe) mm EWES OFS mh ay h _ oat2h+ Nias) (a+ 1a+h+3) =n h(ath+3)(a+3) a (20? + 6a+2ah + 6h +a+3) ~ (2a"+2ah +6a+a+h+3) = eb h(a +h+3)(a+3) Sh 5 5 litt, @zhaayats) (+3? =| ier hs ayers)SECTION 28 DERIVATIVES 123 (a+hP +1 fy AL fo, GRA r (a? + 2ah +h? + 1)(a 2) - (a? +1)(a+h— 2) nb ha +h—2)(a—2) (a? ~ 2a? + 2a7h — dah + ah? ~ 2h? +a~2) ~ (a° + ah ~ 20 +a+h—2) dim, ha+h—2)(a—2) = im Sh = dal + ab? — 2h? h(a? ~4a+ ah ~ 2h 1) a) Tat h—2)(a—2) 16. f'( nto Rat h—2A(a—2) = lim aT 17. Use Definition 2 with f(r) = 1/42. om Tea es an a Var - Jatha3 = lim Vath radars _ vaF2-Vatht2 Vara+ ve ne 7 hvath+2Vat2 Va+2+Vathe2 (a+2)- anes ~ MS ieP REI Var3 (Vad Vat RTD) ah 0 hVat ht 3 Vata varie Vath+2) -1 failing valaies -1 1 ” Watt) Vara) ara? 18 #(@) = im fete f(a) _ /3(a +h) + V8lath)+1~ V8atT lim, sim ——— am h(VaatSh+ 1+ V3aFT) (3a + 3h +1) ~ (3a +1) enact 1 BTR] = lim S a T+ Vaa+l 2V3a41 "Note that the answers to Exercises 19-24 are nat unique, Geyer J'(1). where f(z) = 2° anda G+ hye ~1 h 18. By Definition 2. fim OF: By Definition 2, lm, = £'(0). where f(e) = (1+ 2)" anda128 0 CHAPTER2 LIMITS AND DERIVATIVES. VI6+h-2 "| 20. By Definition 2, iy SEEER—2 _ 716), where f(x) = YE and a = 16 Vie = Or: By Definition 2. fim, wine £'(0). where f(x) = IEF and a = 0. 2-32 z= 21. By Equation 3, im, = (5). where f(«) = 2% and 2. By Equation 3. lim SE —1 linn , SBETE =F (a/4). where f(2) = tans anda = 2/4 cos(a +h) +1 h cos +h) +1 a "(re). where f(z) = cos and a = 23, By Definition 2, Jim, mb Or: By Definition 2, lim = f'(0). where f(x) = cos(m +2) and a = 0. A +t-2 = f'(1). where f(t) = t+ tanda = 1 sgn LOI = 62+ 4) — 8] ~ 2" — 62) — 5] ho h jim (tae A? —12—Gh—5)~(-13) _ 4, WP 2h ho h ho 8 0(2) = f'(2) = fim, FAFA LO uh =tie) = tim LOTNE= Oth + 1 -Re =a = (2h? + 12h? + 24h + 16 -2—h+1) -15 = ie} he = ty 2A HDA? + 28H ae) h = jim(h—2) = -2 m/s 241) = im (2h? + 12h + 23) = 23 m/s 21. (a) f(z) isthe rate of change of the production cost with respect to the number of ounces of gold produced. Its units are dollars per ounce. (b) After 800 ounces of gold have been produced. the rate at which the production cost is increasing is $17/ounce. So the cost of producing the 800th (or 81st) ounce is about $17. (6) Inthe short term. the values of f"(2) will decrease because more efficient use is made of start-up costs as 2 increases. But eventually f"(2) might increase due to large-scale operations. 28. (a) 1"(5) is the rate of growth of the bacteria population when ¢ = 5 hours. Its units are bacteria per hour. (6) With unlimited space and nutrients. f” should increase as t increases; so f"(5) < f"(10). Ifthe supply of ultients is limited, the growth rate slows down at some point in time, and the opposite may be true. 28. (a) /'(v) is the rate at which the fuel consumption is changing wit respect to the speed: Its units are (gal/hy/(mmi/n) (b) The fuel consumption is decreasing by 0.05 (gal/h)/(mi/h) as the ear’s speed reaches 20 mi/h. So if you increase your speed to 21 mi/h, you could expect to decrease your fuel consumption by about 0.08 (gal/’n)/(mi/h).SECTION28 DERIVATIVES C125 30. (a) f’(8) isthe rate of change of the quantity of coffee sold with respect to the price per pound when the price is $88 per pound. The units for f"(8) are pounds /(dollars/pound). (b) f"(8) is negative since the quantity of coffee sold will decrease as the price charged for it increases. People are generally less willing to buy a product when its price increases. ‘31. (10) is the rate at which the temperature is changing at 10:00 a.m. To estimate the value of (10). we will average the difference quotients obtained using the times ¢ = 8 and t = 12. Let _ 28) =T(0) _ 72-81 TU2)-T(10) _ 88-81 = = 5 = 35. The a= 4.5 and B Sn Fo = 35. Then Ti) =TU0) A+B _ 45435 pny Ji, pe Fy 32, For 1910: We will average the difference quotients obtained using the years 1900 and 1920. £(1900) ~ E1910) _ 48.3— 51.1 1900-1910 70 (1920) — E(1910) _ 552-514 9 4s then Let A= = 0.28 and 1920 — 1910 10 _ E(t) ~ B(910) | A+B _ a re : F910) = hig aig © gy = 0.345. This means that life expectancy at birth was increasing at about 0.345 year/year in 1910. For 1950: Using data for 1940 and 1960 in a similar fashion, we obtain £"(1950) = [0.31 + 0.10]/2 = 0.205. So life expectancy at birth was increasing at about 0.205 year/year in 1950. 33, (a) S'(T) is the rate at which the oxygen solubility changes with respect to the water temperature. Is units are (mg/L)/°C. (b) For T = 16°C. it appears that the tangent line to the curve goes through the points (0, 14) and (32,6). So =u , sis) ~ 5 3 0.25 (mg/L)/°C. This means that as the temperature increases past 16°C. the ‘oxygen solubility is decreasing ata rate of 0.25 (mg/L)/°C. 34. (@) S'(T) isthe rate of change of the maximum sustainable speed of Coho salmon with respect to the temperature. Its units are (em/sy/°C. (b) For T = 15°C. it appears the tangent line to the curve goes through the po 32~25 (15) ~ F—3 = salmon is changing ata rate of 0.7 (em/s)/°C. In a simitar fashion for T= 25 °C, we ean use the points 10, 25) and (20,32). So 0.7 (cm/s)/°C. This tells us that at 7 = 15 °C. the maximum sustainable speed of Coho (20, 35) and (25,25) to obtain $"(25) = (cm/3)/°C. As it gets warmer than 20 °C, the 25 ‘maximum sustainable speed decreases rapidly. 35. Since fi sin(I/z) when x £ 0 and f(0) = 0. we have £0+h)=F00) _ ,,,, hsin(1/h) ~0 10) pm, FO48 iad takes the values —1 and 1 on any interval containing 0. (Compare with Example 4 in Section 2.2.) = fim sin(1/h). This timit does not exist since sin(1/h)12 CHAPTER? LIMITS AND DERIVATIVES 36, Since f(x) = 2? sin(1/2) when x # Oand f(0) = 0. we have $0) = fim LO*M = SO) ding anf) 0 mb h mo sim, fsin(1/h). Since 1 ssint < l.we have Ll 1 =[h| <[hlsin # < jh] = —|h| < Asin 7 <[h), Because fim (—[h) = O and Jim, |h| = 0. we know that 1 im (asin i) by the Squeeze Theorem. Thus. f"(0) = 0. 2.9 The Derivative as a Function 1. Note: Your answers may vary depending on your estimates. By y estimating the slopes of tangent lines on the graph of f. it appears that @ f( > (b) f'(2) #08 (co) f'(3) = (a) f'(4) = -0.5, 2. Note: Your answers may vary depending on your estimates. By estimating the slopes of tangent lines on the graph of f. i appears that @ fOr -3 () f(x 0 OF Q=15 @) £3) x2 ) f(a) 0 (f) f(5) -1.2 3. It appears that fis an odd function, so f* will be an even function—that is. f"(-a) = J”(a). (@) f'(-3) 215 (f(a) 21 © f(ayeo (@ fe -4 © s(eo ® fFQ)~1 (@) f'3) 21.5 4. (a= thesince from left to right the slopes of the tangents to graph (a) start out negative, become 0. then positive, then 0, then negative again, The actual funetion values in graph II follow the same pattern (0)'=1V. since from left to right the slopes of the tangents to graph (b) start out ata fixed positive quantity. then suddenly become negative. then positive again. The discontinuities in graph IV indicate sudden changes in the slopes of the tangents. {)'= Iesince the slopes of the tangents to graph () are negative for « < Oand positive for 2 > 0. as are the function values of graph | (6)/= IL since from left to right. the slopes ofthe tangents to graph (A) are positive then 0. then negative. then 0, then positive, then 0. then negative again, and the function values in graph II follow the same patternSECTION29 THEDERIVATIVEAS AFUNCTION C127 Hints for Exercises 513: First plot z-intercepts on the graph of f" for any horizontal tangents on the grap off Look for any corners an the graph of f —there will be a discontinuity onthe graph of f’. On any interval where f has @ tangent with positive (or negative] slope, the graph of f will be positive (or negative. If the graph ofthe function is linear, the graph of f will be a horizontal ine 5 , 6 5 7 : Ly fg 8 e 8 7 5 , j y , ; p 7 : —\ q : tn. ’ eI 7 3 U f n * a : ’ , , » 7 :12% CHAPTER? UiMITS AND DERIVATIVES 14, The slopes of the tangent lines on the graph of y = P(t) are always y positive, so the y-values of y = P(t) are always positive. These {0 yori values start out relatively small and Keep increasing, reaching a so maximum at about ¢ = 6. Then the y-values of y = P(t) decrease swt and get close to zero, The graph of P’ tells us that the yeast culture ‘grows most rapidly after 6 hours and then the growth rate declines. 15. It appears that there are horizontal tangents on the graph of M for > t= 1963 and ¢ = 1971. Thus, there are zeros for those values of oy void con the graph of M’, The derivative is negative for the years 1963 005 to 1971 003 1950 1960 1970 1980 1990 1 _ 1 y= fis) =Inx 16, Sce Figure | in Section 3.4 1. > 8 fo yas) re ‘As increases toward 1, f(z) decreases from The slope at 0 appears to be 1 and the slope tL very large numbers 1. Ast becomes large appears to be 2.7. Asir decreases. the slope gets Fe) getacloserto0, Asa guess, "(a) = 1/2? closer to 0. Since the graphs are so similar. we Deas right guess that f"(2) 49. (a) By zooming in. we estimate that j’(0) = 0. #"() = 1. #'() = 2. 25 and f'(2) = 4 (b) By symmetry. f/(=2) = — f(x). So f"( fy and f'(—2) =~ (6) appears that f(z) is twice the value of 2. so we guess that f(z) = 22. as20. (a) By zooming in, we estimate that f"(0) = 0. f’ (3) 0.75, © SQ) © 3, f'(2) © 12. and (3) = 27. a, 2.2) = jim FEN =) 28. @ f'(@) = (b) By symmetry. f'(—a) = f'(2). f'(-2) = 12, and f"(-3) = 2 (d) Since f’(0) = 0. it appears that Using f’(1) = 3, we have Fle+h) - f( h © f'(@) = lim anh 1) = jn LE*N = See) Domain of f = domain of f! = h = fim, h Domain of f = domain of f* Hert) fle) _ % lim mh Domain of f =Grh — 3n? fern 1 Je) = jim, Domain of f = domain of f’ = R. AO + 5h +3) i 12+ Te+Th—12 qe +2xh + h?)j — (1 — 32) h h(—6x ~ 3h) im, ACO 8H) lim 222 ee 2-527 ~3r42 jim “= Wek Bh Set Gh = 2 ~ 5a? — Se +2 _ SECTION 29 THEDERIVATIVEAS AFUNCTION jim (22 + h) = 20 So f'(—}) © 0-75. f(-1) 7 tf" may have the form f’(2) = a2’ 3.80 f'(x) = 32°, (2? + 327h + 3xh? + h3) i = fim (32? + anh + 42) = 32° 0 7 Jim el = lim fim, “> = jim = = fimo = 0 lim [12+ a +A)] - (12+ 72) moh 25 og Tho tyr him = 3(e + hy] (1 ~ 302 so h lim L232 = 6rh ~ 3h? -1 432? = h = jim (62 ~ 8h) = ~62 lin B(x +h)? + 3(e+ = 2] - (64? +3 ~ 2) im 1OZK + 5h? + 3h h ~ mh h = Jim (102 + 5h +3) = 102 +3 129130 CHAPTER2 LIMITS AND DERIVATIVES fle+h)— la) _ [(e+h)? - 3(@ +h) +5] - (2° - 30 +5) ee = im SS 25. f(x) h 32h + Beh? + h8 — 3h h(x? + 3xh +h? — 3) = tim SER Beh = tim fry h cy he (32? + 3h + h? — 3) = 32? -3 Domain of f = domain of f’ = R. ferns (eth+veFh) - (2+ ve) h 3% se) = Joy ao tn (#+ aa E vetht+ yz _(e+h)=2 aah Vesht vi. “iver hs va) 1 1 1 sin(}} gera) “ene tt Domain of f = [0,00). domain of f" = (0,00) g(e+h) h tim EERE) — VIF | YT RETA + VIER y+ vitae oh h Viv2@ th) + vie 2 1 (1+ 20 + 2h) ~ (1+ 22) _ Tors ah+ Vitae 2Vit2e Vi+2e = lim ® [Vite + VIRB] Domain of g 00). domain of g’ = (—4.00) 34 (eth) 1-3@th) = h (840-4 h)(1- 32) ~ (8 +2)(1~ 30 — 3h) (d= Bx — Bh)(T — 32) (3-92 +2 ~ 32? +h —Shz) — (3-92 ~ 9h + 2 ~ 30° ~ 3hz) am fe) = jin aL) = Jim, =e Wd = Bx — 3h) — Be) = 10h _ 10 ___10 = jim [aa de — 3h) — az) AY Be — BAYA — Be) (32)? 00, 3) U (3-90) a(t+h) at A(t AY +1) — att +h) (441 tT _,, (eth+ E+) hn os) h Domain of f = domain of f" = py, WE ttt) (42 4 4ne-+ 40) ~ nerhe es) = jim, aH = in aoe > ss lin aera pesy MM eEha Des CIP Domain of G = domain of G’ = (~90,~1) U(~1, 09).SECTION29 THE DERIVATIVEAS AFUNCTION D131 14 = im GEAR _ i Pale thy? ah h mo fie + hPa? ee wobble + ha? — aso (e+ Aya? Domain of g = domain of g' _ fet h (x) _ = fy SEA SPH AN (e084 ath 4 dah? 418) ? Domain of f = domain of f’ = R. 32. (a) (at + 4e%h + 6x7h? + doh? +h i ya Vie = 6) = Vox (b) Note that the third graph in part (a) has small negative values for its slope. f; but as. = 6.’ —+ —oe. See the graph in part (). fe+h)- = fla) /6 = Ea V6=(e +h) ~ Ve=a | VO +) + VOW SE Ve—e+h)+ Vo=2 = een ae FH) +ve=a] = Wrest +Vé=2) 1 SSS ht Ve-a 26-2 Domain of f = © F(z) = jim —00, 6). domain of132 CHAPTER2 LIMITS AND DERIVATIVES (b) Notice that when J has steep tangent lines. f"(x) is very large. 0 When f is flatter, /"(x) is smaller. 3 3 m4 On Bh a) (10) = jim LEE A=LO _ pig TEER ME TFP _ ig 6 +6? — 6 ~6(t +h)? mo h mo hm Mb nies + AP] + 8) tig att = 8h, 12 6h_ at mba + (t+ hyU+e) MO +EFAPFI+F) +e? (b) Notice that f has a horizontal tangent when t = 0. This corresponds 6 to J"(0) = 0. f' is positive when f is increasing and negative when f is decreasing. “2 2 =4 35, (a) U'(t) is the rate at which the unemployment rate is changing with respect to time. Its units are percent per year (Toma Uwe eg LUFA =U WEE = UE apa vaso (i992) — U(U991) _ 7.5 1992 — 1991 For 1992: We estimate U’(1992) by using h = —Land f= 1. and then average the two results to obtain a inal estimate, For 1991: U" (1991) = hand => U"(a992) w YC991) — UC1992) 1991 — 1992 hei = oto = 20M VEHH) _ 69-78 _ gop So we estimate that U" (1992) © 3[0.70 + (—0.60)] = 0.05, ¢ | 1901 1992 1993 1994 1995 19961997 __-1998__1999__2000 Ti) | 070 0.05 0.70 —065 -0.35 -0.35 -0.45 0.35 0.25 —0.20 36. (a) P(t) is the rate at which the percentage of Americans under the age of 18 is changing with respect fo time, Its units are percent per year (%/) 4 Pltth) = Plt) PEN) — PUY gop 6m, = (b) To find PY (1). we use fing FEE A ASS xs SE for small values of h P(1960) — P1950) _ 35.7— 31.1 _ 1960 — 1950 0 10 and h = 10. and then average the two results to obtain & 46 For 1950: P’(1950) = For 1960: We estimate P’(1960) by using h final estimate. h=—10 > P1060) 2980) — POSED) BLA — 36.7 = 1950 — 1960 =i 0.46SECTION 29 THEDERIVATIVEASAFUNCTION D133 P(1970) — P(1960) _ 34.0 ~ 35.7 1970 — 1960 10 So we estimate that P’(1960) ~ 3[0.46 + (—0.17)] = 0.148. h=10 = P'(1960) = 0.17 t | 1950 1960 1970 1980-1990 2000 P\(t) | 0.460 0.145 -0.385 0.415 —0.115 0.000 © 1950 1960 1970 19K0 1990. 2000 -05 (d) We could get more accurate values for P'(t) by obtaining data for the mid-decade years 1955, 1965, 1975, 1985, and 1995. 37. f is not differentiable at z = ~1 or at 2 = 11 because the graph has vertical tangents at those points; at 2 = 4, because there is a discontinuity there; and at x = 8, because the graph has a comer there 38, (@) gis discontinuous at « = —2 (a removable discontinuity) at x = 0 (g is not defined there). and at 2 = 5 (a jump discontinuity) (©) gis not differentiable at the points mentioned in part (a) (by Theorem 4), nor is it differentiable at (vertical tangent), or x = 4 (vertical tangent), 2 = —1 (comer). 2 38. As we zoom in toward (—1,0). the curve appears more and more like —— 8 straight line, so f(x) =a + /fa] is differentiable at x = —1. But ‘no matter how much we zoom in toward the origin, the curve doesn’t straighten out—we can’t eliminate the sharp point (a cusp). So f is = ! not differentiable ate = 40. As we zoom in toward (0, 1), the curve appears more and more like a 3 straight line, so f is differentiable at ‘we zoom in toward (1, 0) or (—1,0). the curve doesn’t straighten 0. But no matter how much ‘out—we can’t eliminate the sharp point (a cusp). So f is not +1 differentiable at 2194 CHAPTER2 LIMITS AND DERIVATIVES 4A. (a) Note that we have factored — a as the difference of two cubes in the third step. 1) = jm, =F) = mee in Goa a ST = lin opamp ~ garb?” (6) 710) = fim LO N= LO) _ = jm hy. This function nereases without bound, 0 the limit does not exist, and therefore f'(0) does not exist. (© im 7 (2) = lim, goby = co and fs cominuus at # = 0 (ot funtion, so fhas a veri tangent ate =0. <7: Which does not exist. ('* — a9)(a¥? + a"/3) Rai aa 4 a) as Jim Sarma pee = 2*/ is continuous at x = 0 and @ o4 ‘0. This shows that g has a vertical tangent line at 02 Y v2 4B fle) =|e-6| { yi =F sin LE) =H) — gy - \ aaa et 6 vat tin, HAO = im BEES = im = f2<6 so f'(6) = tim £2 LO goes not exis However. (0 {" ‘te _ 6 ~ |e= 6" Another way of writing the answer is f"(1) 44, fe) = differentiable at n by the contrapositive of Theorem 4 If @ is not an fx] is not continuous at any integer m, so fis not integer. then fis constant on an open interval containing a, 0 f(a) = 0. Ths, (2) 0, 2 not an integer.SECTION29 THEOERIVATIVEAS AFUNCTION 135 2 45. (a) f(a) = 2|a| = ie : : ‘ (b) Since f(2) = 2 fore > 0, we have f'(z) = 2x for x > 0. [See Exercise 2.9.19(d),] Similarly, since f(x) = —x? for » 2 <0.wehave f'(z) = —2r forx <0. Ata = 0, we have £E)= 10) — hig Zh 0 8 a + So f is differentiable at 0. Thus. f is differentiable for all 2. ki ee cee eee (©) From par (o,we have (2) =) 9° eo cg f = 2th 48.0) £08) = hg LEE FO gg SOMA ty tang L —iy-! (4) = f(4+h) = F(4) _ (4+h) 1-(1-h) _ ,—t AO nh = BR aay = lim goa 0 ite4 ‘These expressions show that f is continuous on the intervals (00,0). (0,4). (4,5) and (5,00). Since Jim, f(x) = tim, (5 sat =e Jn f(z), lim f(2) does not exist. so f is discontinuous (and therefore not dlifereniable) a0 Atdwe have lim f(s) = tim (6 and Jim, f(@) = lim, = 80 lim f(z) =1= f(4) and f is continuous at 4. Since f(5) is not defined, f is discontinuous at 5. (a) From (a), fis not differentiable at 4 since (4) 4 f/.(4) 47. (a) If f is even, then (2) = tim fo +h) = faz) _ flee — A) - Few f(z~=h)— FN y= fit RIEL CD tp He) = See) == jim LEH LE) the yy =~ fim, FE*89— 10 __ yy ), and from (c). f is not differentiable at 0 oF 5, Therefore, f" is odd. (b) If f is odd, then f'(-2) Scat h)= fr) _ fle A) - f-2) h “ h =f (e@—h) + fle) a h = km = fin SEDI) tiene 1) = gm, EAA) $8) _ py Therefore, f" is even,196 CCHAPTER2 LIMITS AND DERIVATIVES 8. (a) 1 © ami a 7 (b) The initial temperature of the water is close (0 room temperature because of the water that was in the pipes. When the water from the hot water tank starts coming out, dT'/dt is large and positive as T’ increases to the temperature of the water in the tank. In the next phase. dT/dt = 0 as the water comes out at a constant, high temperature. After some time. d7'/dt becomes small and negative as the contents of the hot water tank are ‘exhausted, Finally, when the hot water has run out. d7/dt is once again 0 as the water maintains its (cold) temperature. 49, , In the right triangle in the diagram, et Ay be the side opposite angle ¢ and ‘Ac the side adjacent angle 6. Then the slope of the tangent line € is m = Ay/Az = tang. Note that 0

0. lim. f(x) # st0e and J has no horizontal asymptote M1. False. Consider f(x) = {7 ~) ' : , ‘ 12. False, The function f must be continuous in order to use the Intermediate Value Theorem, For example, let 1 if0se<3 f(a) = ‘There is no number ¢ € [0,3] with f(c) = 0. -1 if2=3 13, True. Use Theorem 2.5.8 with a = 2, = 5, and g(x) = 42” ~ 11. Note that f(4) = 3 is not needed. 14, Truc. Use the Intermediate Value Theorem with a = —1.b = 1,and N =, sine 3 <7 < 4. 15, True, by the definition of a limit with ¢ = 1 if 2 #0 16. False. For example, let f(x) = al example, let f(x ‘Then f(z) > 1 for all, but lim f(x) = lim, (x? +1) 17, False. See the note after Theorem 4 in Section 2.9. 1B Tre. f(rjexists > fisdifferentiadleatr => fiscontinuousatr => Jim f(x) = f(r) EXERCISES 14.@@ Jim, f(e)=3 Gi) | tim, f@) Gi) im, f(@) does not exis since the left and eight (iv) tim, (2) limits are not equal. (The left limit is ~2.) (w) lim f(x) = 00 (vi) tim f(@) = (vii) lim f(z) viii) tim, f(z) = —1 (b) The equations of the horizontal asymptotes are y = —Land y = 4 (©) The equations of the vertical asymptotes are # = O and x = 2. 0,2, and 4. The discontinuities are jump, infinite, infinite, and removable, (@) f is discontinuous at © respectively.CHAPTER2 FEVIEW D139 4. Since the exponential function is continuous, 4. Since rational funetions ae continuous. lig FP = ya ; ; (e+8)w~3)_ |, 2-3 _-3-3_-6 3 & lim, aya M5 (e@+3)(@-1) ~ 235-1 > ~3=1~ ~a 9 cs + or x 6 inn, a Tgpng = noe since x? +28 Onset and = cone ce cs -1? h? — 3h? + 3h —1 8 — 3h? Te tin =P + WN BH gang 9 i 7h im fim, Another solution: Factor the numerator as a sum of two cubes and then simplify, him (HIP 41 Goa DP418 5 (A= 1) +1) [a1 1-1) +P] mh a) h lim [(h - 1)? ~h +2] =1-04+2=3 (e+ 2)(t=2) +2 242 4 1 ~QC +84) PETA T4413 = co since (r~ 9)* + Oasr + 9 and >Oforr £9, Yee [a of = HB. aay = lin, 4-V5 -1 -1 Jim, 1 SEARED Ea” le era ~ Tieea 8 = lim —@+O=2) ved aa 22 (V+ DV DOF HA ~ oA WFD HA ~ OPH EH ~ 16 if2-8>0 {: ita>s =@-8) 4 1 ife Land 2? + Os +0, sox = Dis the vertical asymptote, 24, From the graph of y = f(z) = Va? a+ 1 — Vz*— ait appears that there are 2 horizontal asymptotes and possibly 2 vertical asymptotes. To obtain a different form for J. let's multiply and divide i by its conjugate fila) = (Vee Fe+1 - VP) verrar + va Verotit+ve—2 _(@te41)-(e?=2) arti ~ aatis Vas Ve aati tee Now 241 dm f@)= MR erpasit vee = im 2) since Va? = a for = lm, Us (since Va? = a for x > 0) 2 “Ti so y = 1 isa horizontal asymptote. For < 0, we have \V/@ = |x| = —z, so when we divide the denominator by 2, with «<0, we get Veeari + VPae_ VEEP t VP —e rn rr iCHAPTER2 AEVIEW CO 1M Therefore, h fi lim Qe +1 in MO)= lm, Taperit Vee 2400 lim. 2+ (1/2) = [Vis ORs) + Via] OD soy = —1 isa horizontal asymptote. As x — 0. f () + 1, soe = Ois nota vertical asymptote, As 2 —+ 1*, F(x) + V3, so. = Lis not a vertical asymptote and hence there are no vertical asymptotes. Since 21 < f(z) < 24 for0 <2 < Band lim (22 ~1 lim, 2, we have lim (x) = L by the ‘Squeeze Theorem, 2. Let f(x) = —2?, g(x) = 2* cos(1/x*) and A(z) = 2?. Then since Jcos(1/2?)| < 1 fore # 0, we have f(x) < g(x) < h(x) for x £0, and so Jim f(2) = lim h(x) = 0 > Jim g() = 0 by the Squeeze Theorem. 2. Given ¢ > 0. we need 6 > 0 so that if < [2 ~5| < 6, then (7x ~27)-8| 0 we must find 6 > 0 so that if 0 < [x ~ O| < 6 then | Ye — 0| < €, Now| YF — | 2] =1¥2l* lve -0| =| y72| Therefore. by the definition ofa limit lim, Y= Wl 0. we need § > Oso that iO < [a ~ 2] < 6, then |2? — 32 ~ (~2)| <, Fist note that it | ~ 2| <1, then=L<@—2 0, we need 6 > 0 such that if 0 < 2-4 < 6.then 2/Z=4> M. Thisis true es Ve~4<2/M 4 2—4<4/M?. Soitwechoose = 4/M* then) <2—4<8 = 2VETA>M. Soy the definition of limit, lim, (2h/#=4) = oo, 4H. (a) f(x) = V~eifx <0. f(2)=3-2it0 <2 <3. f(x) = 3Y if2>3, @ Jim, f(2) = lim, (8-2) =3 Gi) tim fe) = tim =z Gi) Because of (i) and Gi), lim f(x) does not exist. 6) Jim f(2) = tim (3-2) =0 (9) lim, J(e) = lim, (e— 3)? (i) Because of iv) and (¥, lim f(142 CHAPTER2 LIMITS AND DERIVATIVES (6) fis discontinuous at 0 since lim f(x) does not exist. © r {is discontinuous at 3 since f(3) does not exist. : ys 32. (a) g(a) = 2x - a7 if0 <2 <29(z)=2-wif2<2<3, g(a) =2—4if3 <2 <4,9(2) =mifx > 4. Therefore, Tim_o(2) = tim (222°) = Oand lim, g(a) = lim, (2 ~ 2) = 0. Ts. lim, (2) g(2).sog iscontinuous at 2, lim_g(2) = lim (2~<) Land lim, g(2) = tim, (@ — 4) = —1. Ths, lim of) (3). og is continuous at 3 ast Jima} “vm (@=4) = Oand lim, g(2) = fim, =m Thus, tim, g(2) doesnot exist, 509 is (4), so g is continuous from the right at 4. discontinuous at 4, But lim, g(x) 33, sina is continuous on R by Theorem 7 in Section 2.5. Since e* is continuous on R, e*** is continuous on R by “Theorem 9 in Section 2.5. Lastly, « is continuous on R since it's a polynomial and the product zre""* is continuous ‘on its domain R by Theorem 4 in Section 25. 34, 2 — 9 is continuous on R since it is a polynomial and \/7 is continuous on {(, 00). so the composition Ve —9is continuous on {=r | x? ~ 9 > 0} = (—o0, ~3] U [8,00). Note that x? — 2 4 0on this set and so the quotient z function g(x) = — 3 i, continuous on its domain, (—o0, —3] U [3, 0). 35. j(2) = 22° +2? + 2isa polynomial, so itis continuous on [-2, ~I] and f(—2) = -10 <0. <1 = f(-1). So by the Inermediate Value Theorem there isa number ¢ in (~2,~1) such that f(c) = 0. thats, the equation 20° +a? + 2-=O has aroot in (—2,—1). 36. f(z) + is continuous on Bo itis continuous on (0,1). f(0) = 1 > 0 > 1/e~1 = f(A). So by the Intermediate Value Theorem. there is a number ¢ in (0; 1) such that f(c) = 0. Thus. —2=0,0re™ has root in (0,1) 31. (a) The slope of the tangent line at (2,1) is Fle) = $2) _ jy, 9-22” = i, AE = DUO +2) by yo eH? my e-2 = tim, [-2(@ + 2)] = (b) An equation of this tangent line is y ~ 1 = —8(x ~ 2) ory = ~8x + 17. 38, For a general point with z-coordinate a, we have 2/(i — 32) ~ 2/(1~3a) _ ,,,, 21 = 8a) ~ 20 — 32) eae) za 22a (1 —3a)(l — 32)(2— tim ———8@=4) = fim 8 = = lim Gayl —3z)@— a) se (1—3a)(1—3z) (1 ~ 3a)?CHAPTER? FEVIEW C143 For a = 0m = 6 and f(0) = 2. so an equation of the tangent line is y ~ 2 = 6(x — 0) or e+ Iory 6x +2. For bes a=-1,m so an equation of the tangent line is y — 3 38. (a) 5 = s(t) = 1+ 2 + £2/4. The average velocity over the time interval (1, 1 +h] is = S(L+h) = (1) _ 14+ 20+A) + (1 +4)7/4— 13/4 ohh? _ OA be" OFAy=T Ra So for the following intervals the average velocities are: (i) 1,3), 2. vave = (10 + 2)/4 = 3 m/s Gi) [1,2]: h=1. reve = (10-4 1)/4 = 2.75 m/s (ii) [1.1.5]: b= 0.5, vave = (10 + 0.5)/4 = 2.625 m/s Gv) (L.1.1]: k= 0.1, vave = (10 + 0.1)/4 = 2.525 m/s 40. (a) When V increases from 200 in* to 250 in’, we have AV = 250 ~ 200 = 50 in®, and since P = 800/V. AP = P(250) — P(200) = = ~ BO = 5.2 — 4 = ~0.8 fin’. So tne average rate of change is AP _ -08 Ibyin? BV > 50 = 016 5 (b) Since V = 800/P, the instantaneous rate of change of V with respect to P is lim 2Y. V(P +h) -V(P) 800/(P + h) — 800/P nab AP h a) h = jim SO0P— (P+) _ _ 80 RP +RYP ad (PHA)P which is inversely proportional to the square of P. $2) _ yy = 22-4 2 im a-2 © + son (27 2)(0? +20 +2) eb ze AL. (a) #'(2) = rs = lim (x? +22 +2) = 10 (b) y~4 = 10(2 —2) ory = 102 - 16 42. 2° = 64. s0 f(z) = 2° anda = 2 43. (a) J"(r) is the rate at which the total cost changes with respect to the interest rate. Its units are dollars/ (percent per year), (©) The total cost of paying off the loan i increasing by $1200/(percent per year) as the interest rate reaches 10% So if the interest rate goes up from 10% to 11%, the cost goes up approximately $1200. (©) Asr incteases. C increases. So f(r) will always be positive144 CHAPTER2 LIMITS AND DERIVATIVES For Exercises 44~6, see the hints before Exercise § in Section 2.9 41. (a) f(z) = Ii f(zth) — fl I f= ses —V3—5r V3 —5(@ +h) + V3 — 52 5(a +h) + V3 . (a) $2) = fim, h aE eaunaad im B= 5le+ W)= (3 52) 7 ens) mea STE” Pricer (b) Domain of f: (the radicand must be nonnegative) 3-52 >0 = 5rS3 = TE (-co, 3] Domain of /’: exclude $ because it makes the denominator 2er0: 2 € (—90. §) (©) Our answer to part (a) is reasonable because f’(x) is always negative and f is always decreasing 48. (a) As —+ boo. f(x) = (4 2)/(3-+ 2) + 1, sothere is a 1, Ase > ~3*, f(z) = 00. horizontal asymptote at y = and as at + —3~, f (2) + ~o0. Thus, there is a vertical asymptote at ¢ = 3.CHAPTER? REVIEW 145 (b) Note that f is decreasing on (—o0, 3) and (-3,00). 0 f” ’ is negative on those intervals. As 2 —> oo, f” + 0. As 2-3" and as ~3*, f' + 00, 4-(+h) 4-2 {@+h)- fe) _,,, SEER) 3¥e h k= h (3+ 2) (4 ~ (2 +h)] ~ (4—2) (3+ (@+h)} = hB+ (+h) @+2) jim G2= 32 = 3h+ 4x ~ 2? ~ he) ~ (12-4 do + 4h ~ 30 ao AB + (c+ h)B+a) =Th = -7 7 “DR ipretmGrs ES Bream ese 7 Gre © f'(@) = lim no = he) (@) The graphing device confirms our graph in part (b). 49, f is not differentiable: a J is not continuous, and at x = 5 because f has a vertical tangent 4 because f is not continuous, at I because f has a comer, at 2 = 2 because 50. (a) Drawing slope triangles, we obtain the following estimates: F”(1950) = 42 = 0.11, F'(1965) = = = -0.16, and F*(1987) ~ 92 = 0.02, (©) The rate of change of the average number of children born to each woman was increasing by 0.11 in 1950, decreasing by 0.16 in 1965, and increasing by 0.02 in 1987, (6) There are many possible reasons: * In the baby-boom era (post-WWID. there was optimism about the economy and family size was rising + In the baby-bust era, there was less economic optimism. and it was considered less socially responsible to have a large family. * In the baby-boomlet era, there was increased economic optimism and a return to more conservative atitudes 51. B°(1990) is the rate at which the total value of U.S. banknote in circulation is changing in billions of dollars per year. To estimate the value of B3'(1990). we will average the difference quotients obtained using the times # = 1985 and¢ = 1995. Ler A= 21985) ~ B(1990) _ 182.0 268.2 ea Be = 17.24 and — B(1995) ~ B(1990) __ 401.5 — 268.2 _ Cas ato = = 26.66. Then rn Bi) ~ BY990)_ A+C _ 17.24 426.66 5 B'(1900) = lig, P—ag9g AES = ERA 21.95 billions of dollars/year.446 CHAPTER2 LIMITS AND DERIVATIVES. rtl is — 52. The slope ofthe tangent to y = (eth)t1 2t1 (w@+h)-1 1 tim @—D@+h+)—(@+V@th—-Y ky i im, Tile —1)e+h—1) = lim ——S2*@_ = 2 Iie De +h-l) @-1 So at (2,3), m -aope? Ss y-3=-22-2) > 8 y Qa +7. At(—1,0).m > (ttl) > y=—3 =4 = lim ~9(2). Thus, by the Squeeze Theorem, 58. [f(2)| <9) = —9() < f(a) < g(x) and lim g(x) = lim f(z) = 0. 54, (a) Note that f is an even function since f(x) = f(—z). Now for any integer n, [n] + [-n] =n — n = 0, and for any real number k which is not an integer, (a) + [-4] = fh) + (- FA) - 0) = (and is equal to ~1) for all values of a. So lim f(a) exists (b) f is discontinuous at all integers.(PROBLEMS PLUS 1. Lett = ¥@,sox = ¢°. Thent + 1asz— 1,s0 tim CHD), tL ee et@-je@ti+) Ml erep1 ~ Peel 3 Another method: Multiply both the numerator and the denominator by (/% + (VF + YE+ 1). vartb-2 Varyb+2 artb-4 2 tionalizs ator: Timy SOP ON, VS = tim Sot Noy since the First rationalize the numerator: ling Var tb+3 ~ 2 xyarbs ay Now siete denominator approaches 0 as —+ 0, the limit will exist only if the numerator also approaches 0) as x — 0. So we require that (0) +b—4=0 = b= 4. Sothe equation becomes tim ———— = 1 = —*—=1 S a=4 Therefore,a=b=4. $3 aeTea Va+2 3. For 3 <2 < 3, wehave 22 — 1 < Oand 2 +1 > 0. so |2e— 1] = ~(2¢—1) and [2e +1) = 20 +1. ‘Therefore, ty P=N= Bel =O —GeHD) = z = z 22> 4. Let Re the midpoint of OP, so the coordinates of Rare (32, 322) since the coordinates of P are (2, 2°). Let 1 Q = (0,0). Since the slope mop = 2 mar = ~~ (negative reciprocal). But As z+ 0,a—+ 4, and the limiting postion of @ is (0, 3) * SeeblS tel etme << “a = 1s Gy <1 pore eb Ase on 1 ~ 00.80 F5 + Oand + rey + Thus. im ae = L by the Squeeze Theorem, 6 (@) [x] + 1. Since [z]? and [yJ? are positive integers or 0, there are only 4 cases: Case (i): fa] = 1. fy) = 1 fe+y=41 (@) Forn > 1s 0 and F(t) = f(i) —1 <0. Soby the Intermediate Value Theorem. there exists some number cin the interval (0,1) such that F(c) = J(c) — ¢ = 0. So f(c) = c. and therefore f has a fixed point 9. lim f(z) Aim, (2) = him (5 [F(@) + 9f2)] + LF) ~ o@)]) sim [f() + 9(2)] + Sim LF(2) ~ 92] Boog de. fand tim gfe) = im ({F(e) + a(@)] ~ FC) = fim Fle) + 9f@)] ~ Bm fe So tm (a)o(a] = [Lim 2)] [Jim 9(@)] = 8-4CHAPTER2 PROBLEMS PLUS 149 Another solution: Since lim [f(z) + g(x)] and lim (f(x) ~ g(2)] exist, we must have if (2) + ote)? = (im (2) + 9(2]) "an tm [7 (2) — (2? = (im (72) ~9(2])° 80 dim (f(z) 9(e)] = im 3 ([f(2) + 9(2)]? ~ [f (@) — g(x)J?) [because all of the f? and g? cancel) Jim [f(2) + 9(2)}? — lim [f(2) ~ g(z))*) = 4(2?-12) = 2 10. (a) Solution J: We introduce a coordinate system and drop a perpendicular from P. as shown, We see from ZINCP that tan 20 = -¥— and from ZN BP that tan 6 = y/2. Using the double-angle formula for 2 2 tangents, we get 4 — tan gg = 220 pte 1 2x ‘After a bit of simplification, this becomes y* = @ (3x ~ 2), As the altitude AM decreases in length. the point P will approach the a-axis, that is. y —~ 0. So the limiting location of P must be one of the roots of the equation «(3.2 ~ 2) = 0. Obviously it is not x = 0 (the point P can never be to the left of the altitude AIM, which it would have to be in order to approach 0) soit ‘must be 3a: — 2 Solution 2: We ada few lines tothe original diagram, as shown. Now note that BPQ = ZPBC (altemate angles: QP || BC by symmetry) and similarly ZCQP = ZQCB. So ABPQ and ACQP 9 e are isosceles, and the line segments BQ. QP and PC are all of equal length. As |A.M| — 0, P and @ approach points on the base. and the b c point P is seen to approach a position two-thirds of the way between B and C, as above. thats @ = 3 (b) The equation y? = (3x ~ 2) calculated in part (a) is the equation of ” the curve traced out by P. Now as |AM| + 00,20 + $.4 —+ = A Land since tan @ = y/x. y —+ 1. Thus. P only traces out the Part of the curve with 0 < y <1 P(x y) c. uP Te 11. (a) Consider G(x) = T(@ + 180°) ~ T(x). Fix any number a. If G(a) = 0, we are done: ‘Temperature at @ = Temperature ata + 180°. If G(a) > 0. then G(a-+ 180°) = T(a+360°) ~ T(a-+ 180°) = T(a) ~ T(a + 180°) = ~G(a) < 0. Also. G is continuous since temperature varies continuously. So, by the Intermediate Value Theorem. G has a zero on the interval [a,a + 180°]. If G(a) < 0, then a similar argument applies (b) Yes. The same argument applies, (€) The same argument applies for quantities that vary continuously. such as barometric pressure. But one could argue that altitude above sea level is sometimes discontinuous. so the result might not always hold for that quantity,180 CHAPTER? PROBLEMS PLUS 12g) = fi SEEN = ME) — hn +L 7 Ay = 2f(z) aan fe se af (a) , hfle+h) im, mn Let De s@) , Jim fle +m) = xf (xe) + f(x) because f is differentiable and therefore continuous. 4B. (a) Put x = O and y = 0 in the equation: f(0 +0) = (0) + f(0) +0? 040-07 = f(0) = 2f(0). Subtracting f(0) from each side of this equation gives f(0) = 0. a ©) s'0)= Jim LOA = £0) — yp [HO 109 + OH | — HO) = fin, £2 = tm LD in it yn © SC) = jim $E*N—LE) _ im pe)+ 10) sant sh) Fle) = py AOD-bath a py (LM) 402 nat] 12 ah i |a| > 0 for alla. The only conclusion is that < LO <1) But 44, We are given that |f()| < 2? for all x. In particular, |f(0)| < 100) =0.Now |£2=£0 Ml # tim (— ||) = 0 = lim |a|, so by the Squeeze Theorem, im fi f0. = 0. So by the definition of a derivative, f is differentiable at 0 and, furthermore, f"(0) = 0. £@) @