Explicit solution of the SVIR (Susceptible-Vaccinated-Infectious-Recovered) epidemic model ========================================================================================== * Norio Yoshida ## Abstract An explicit solution of an initial value problem for the Susceptible-Vaccinated-Infectious-Recovered (SVIR) epidemic model is obtained, and various properties of the explicit solution are investigated. It is shown that the parametric form of the explicit solution satisfies some linear differential system including a positive solution of an integral equation. In this paper integral equations play an important role in establishing the explicit solution of the SVIR epidemic model, in particular, the number of infected individuals can be represented in a simple form by using a positive solution of an integral equation. Uniqueness of positive solutions of the SVIR epidemic model is also investigated, and it is shown that the explicit solution is a unique solution in the class of positive solutions. Keywords and phrases * SVIR epidemic model * integro-differential equation * explicit solution * integral equations * initial value problem * linear differential system ## 1 Introduction Recently there is an increasing importance in establishing exact (or explicit) solutions of epidemic models. It goes without saying that a vast literature and research papers, dealing with epidemic models has been published so far (cf. [2, 3, 7, 12]). However, very little is known about exact (or explicit) solutions of the epidemic models (cf. [1, 8, 10, 18, 22–25]). An *explicit* solution of an ordinary differential equation (or system) *D*(***y***) = 0 is any solution that is given in the explicit form ***y*** = ***y***(*t*). An *implicit* solution is any solution that is not in explicit form. An *exact* solution is a representation of a function which solves a given differential equation or system exactly. An exact solution may be explicit or implicit. The Susceptible-Vaccinated-Infectious-Recovered (SVIR) epidemic models have been an important and interesting subject to study (cf. [5, 11, 13– 17,19–21]). There appears to be no known results about explicit solutions of Susceptible-Vaccinated-Infectious-Recovered (SVIR) epidemic models even for a simple model. The objective of this paper is to obtain an explicit solution of SVIR differential system, and to investigate various properties of the explicit solution. Moreover, uniqueness of positive solutions is studied, and it is shown that there exists a unique solution of an initial value problem for SVIR differential system in the class of positive solutions. We are concerned with the SVIR differential system ![Formula][1] ![Formula][2] ![Formula][3] ![Formula][4] for *t >* 0, where *β, γ* and *σ* are positive constants. The initial condition to be considered is the following: ![Formula][5] The above simple model assumes that the class *S*(*t*) is decreased by infection following contact with infectious individuals at a rate *β*, and is decreased by vaccination at a rate *σ*. The class *I*(*t*) of infectious individuals is generated through contact with infectious individuals at a rate *β*, and is decreased by recovery at a rate *γ*. In this model it is assumed that vaccinated people will not be infected. The model is illustrated below (cf. [16]). It is assumed throughout this paper that: (A1) ![Graphic][6]; (A2) ![Graphic][7] and ![Graphic][8] satisfies ![Formula][9] (A3) ![Graphic][10]; (A4) ![Graphic][11]; (A5) ![Graphic][12] (positive constant). We note that the assumption (A2) is equivalent to the following ![Formula][13] in view of the fact that ![Graphic][14]. In Section 2 we derive a parametric form of a solution of an initial value problem for SVIR differential system, and we discuss the existence and uniqueness of positive solutions of an integral equation in Section 3. In Section 4 we establish an explicit solution of an initial value problem for SVIR differential system, and we show that the number of infected individuals can be represented in a simple form by using a positive solution of an integral equation. Section 5 is devoted to various properties of the explicit solution of SVIR differential system. In Section 6 we show that there exists one, and only one, solution of an initial value problem for SVIR differential system in the class of positive solutions. ## 2 Parametric form of a solution of SVIR differential system In this section we show that a positive solution of the initial value problem (1)–(5) can be represented in a parametric form. Since ![Formula][15] by (1)–(4), it follows that ![Formula][16] for some constant *k*. In view of the fact that ![Formula][17] we conclude that ![Formula][18] We state the following important lemma. Lemma 1. *Let* (*S*(*t*), *V* (*t*), *I*(*t*), *R*(*t*)) *be a solution of the SVIR differential system* (1)–(4) *such that S*(*t*) *>* 0 *for t >* 0, *then R*(*t*) *satisfies the following integro-differential equation* ![Formula][19] *Proof*. Since *S*(*t*) *>* 0 for *t >* 0, we see from (1) that ![Formula][20] Combining (4) with (7) yields ![Formula][21] Integrating (8) on [*ε, t*](*ε >* 0) and then taking the limit as *ε* → +0, we obtain ![Formula][22] Or ![Formula][23] which implies ![Formula][24] Substituting (9) into (2), we get ![Formula][25] Integrating the above on [*ε, t*](*ε >* 0) and then taking the limit as *ε* → +0, we arrive at ![Formula][26] Since *S*(*t*) + *V* (*t*) + *I*(*t*) + *R*(*t*) = *N*, we see from (4) that ![Formula][27] which, combined with (9) and (10), yields ![Formula][28] which is the desired equation (6). By a *solution* of the SVIR differential system (1)–(4) we mean a vectorvalued function (*S*(*t*), *V* (*t*), *I*(*t*), *R*(*t*)) of class *C*1(0, ∞) ∩ *C*[0, ∞) which satisfies (1)–(4). Associated with every continuous function *f* (*t*) on [0, ∞), we define ![Formula][29] Lemma 2. *Let* (*S*(*t*), *V* (*t*), *I*(*t*), *R*(*t*)) *be a solution of the SVIR differential system* (1)–(4) *such that S*(*t*) *>* 0 *and I*(*t*) *>* 0 *for t >* 0. *Then there exist the limits S*(∞)(= 0), *V* (∞), *I*(∞) *and R*(∞). *Proof*. Since *I*(*t*) *>* 0 for *t >* 0, it follows from (4) that *R**′*(*t*) = *γI*(*t*) *>* 0 for *t >* 0, and therefore *R*(*t*) is increasing on [0, ∞). Since ![Graphic][30], we observe that *R*(*t*) *>* 0 for *t >* 0. We see from (2) that *V**′* (*t*) = *σS*(*t*) *>* 0 for *t >* 0 since *S*(*t*) *>* 0 for *t >* 0, and therefore *V* (*t*) is an increasing function on [0, ∞). Hence *V* (*t*) *>* 0 for *t >* 0 because of ![Graphic][31]. It is easy to see that *R*(*t*) is bounded from above in light of ![Formula][32] We deduce that *R*(*t*) is increasing and bounded from above, and therefore there exists the limit *R*(∞). Analogously we conclude that there exists the limit *V* (∞) because *V* (*t*) is increasing and bounded from above in view of ![Formula][33] Taking the limit as *t* → ∞ in (9), we find that *S*(∞) = 0. Using the relation ![Formula][34] and taking the limit as *t* → ∞ in the above, we obtain *I*(∞) = *N* − *V* (∞) − *R*(∞). Theorem 1. *Let* (*S*(*t*), *V* (*t*), *I*(*t*), *R*(*t*)) *be a solution of the initial value problem* (1)–(5) *such that S*(*t*) *>* 0 *and I*(*t*) *>* 0 *for t >* 0. *Then the solution* (*S*(*t*), *V* (*t*), *I*(*t*), *R*(*t*)) *can be represented in the following parametric form*: ![Formula][35] ![Formula][36] ![Formula][37] ![Formula][38] ![Formula][39] *with ψ*(*u*) *satisfying the integral equation* ![Formula][40] *Moreover, ψ*(*u*) *satisfies the following conditions* ![Formula][41] *Proof*. We define the function *u*(*t*) by ![Formula][42] It follows from (4) that *R**′*(*t*) = *γI*(*t*) *>* 0 for *t >* 0, and hence *R*(*t*) is increasing on [0, ∞) and ![Graphic][43]. Therefore *u* = *u*(*t*) is decreasing on ![Graphic][44] and lim*t*→∞ *u*(*t*) = *e*−(*β/γ*)*R*(∞). Since *e*−(*β/γ*)*R*(*t*) ∈ *C*1(0, ∞), we see that *u*(*t*) is of class *C*1(0, ∞). Hence, there exists the inverse function *φ*(*u*) ∈ *C*1(*e*−(*β/γ*)*R*(∞),![Graphic][45] of *u* = *u*(*t*) such that ![Formula][46] *φ*(*u*) is decreasing in ![Graphic][47] and ![Graphic][48]. We easily find that the relation ![Formula][49] holds for ![Graphic][50]. Substituting *t* = *φ*(*u*) into (6), we have ![Formula][51] Differentiating both sides of (17) with respect to *u*, we get ![Formula][52] and therefore ![Formula][53] From (17) we see that ![Formula][54] By the change of variables *s* = *φ*(*v*) we obtain ![Formula][55] in light of (17). It is easy to check that ![Formula][56] Combining (21) with (22) yields ![Formula][57] Substituting (19), (20), (23) into (18) and using (17), we are led to ![Formula][58] or equivalently ![Formula][59] We define the function *ψ*(*u*) by ![Formula][60] If we define ![Formula][61] by using (4) and (19), then *ψ*(*u*) is continuous in ![Graphic][62]. Noting that *t* = *φ*(*u*) *>* 0 for ![Graphic][63], we see from (4), (19) and (25) that ![Formula][64] for ![Graphic][65]. Consequently we observe that *ψ*(*u*) is a positive continuous function in ![Graphic][66]. We observe, using (25), that ![Formula][67] and therefore (15) holds. Combining (24) with (25), we deduce that *ψ*(*u*) satisfies the integral equation (16). Now we derive the parametric form of the solution (*S*(*t*), *V* (*t*), *I*(*t*), *R*(*t*)). Substituting *t* = *φ*(*u*) into (9) yields ![Formula][68] in light of (17), and therefore (11) holds. We see from (20) that ![Formula][69] which is the relation (14). Letting *t* = *φ*(*u*) in (10), we obtain ![Formula][70] because of (23), and hence (12) is obtained. Since *I*(*t*) = *N* − *S*(*t*) − *V* (*t*) − *R*(*t*), we observe, using (26)–(28), that ![Formula][71] which is the desired relation (13). Since ![Graphic][72], it is necessary that ![Graphic][73]. Taking account of (13) and (16), we get ![Formula][74] which means *I*(∞) = 0. We conclude that ![Formula][75] in view of the fact that ![Graphic][76]. Since *S*(∞) = *I*(∞) = 0, we are led to the relations ![Formula][77] by (1)–(4). Corollary 1. *Let* (*S*(*t*), *V* (*t*), *I*(*t*), *R*(*t*)) *be a solution of the initial value problem* (1)–(5) *such that S*(*t*) *>* 0 *and I*(*t*) *>* 0 *for t >* 0. *Then, S*(*t*), *V* (*t*) *and I*(*t*) *can be written in the form* ![Formula][78] *by using R*(*t*). *Proof*. First we note that *t* = *R*−1(−(*γ/β*) log *u*) in view of *R*(*t*) = *R*(*φ*(*u*)) = −(*γ/β*) log *u*. It is easily seen that ![Formula][79] in the proof of Theorem 1. Substituting the above two relations into (11)– (14) yields the desired representations. **Remark 1**. If *I*(*t*) *>* 0 for *t >* 0, then *R*(*t*) is increasing on [0, ∞) in light of (4). Since ![Graphic][80], it follows that *R*(*t*) *>* 0 for *t >* 0. Analogously, *V* (*t*) *>* 0 for *t >* 0 if *S*(*t*) *>* 0 for *t >* 0. **Remark 2**. If *σ* = 0, ![Graphic][81] and *V* (*t*) ≡ 0, the SVIR epidemic model studied here reduces to the SIR epidemic model, and (16) reduces to ![Formula][82] which plays an important role in deriving an explicit solution of the SIR epidemic model (see e.g. [23]). ## 3 Existence and uniqueness of positive solutions of an integral equation In this section we study the existence and uniqueness of positive solutions of the integral equation ![Formula][83] where *K**i* (*i* = 0, 1, 2, 3) are positive constants, *x* ≥ 0, *x*1 *> x* and *K* *>* 2*x*. Theorem 2. *The integral equation* (29) *has at most one positive solution. Proof*. Let *y**i*(*x*) (*i* = 1, 2) be two positive solutions of (29), and let *c* := min{*a, b*}, where ![Formula][84] Then we see that ![Formula][85] It is easy to check that ![Formula][86] The mean value theorem implies that there exists a number *τ* ∈ (0, 1) such that ![Formula][87] If *u* ≤ 0 and *v* ≤ 0, then ![Formula][88] and therefore ![Formula][89] Letting ![Formula][90] we see from (30) and (31) that ![Formula][91] Choosing *L >* 0 large enough so that (*K*1*D*)*/L*2 *<* 1, we observe, using (32), that ![Formula][92] and hence ![Formula][93] Therefore we obtain ![Formula][94] which yields ![Formula][95] in view of (*K*1*D*)*/L*2 *<* 1. Consequently we conclude that |*y*1(*x*)−*y*2(*x*)| ≡ 0 on [*x*, *x*1], i.e. *y*1(*x*) ≡ *y*2(*x*) on [*x*, *x*1]. Theorem 3. *Assume that the hypothesis* ![Formula][96] *holds. Then there exists a positive solution y*(*x*) *of* (29). *Proof*. We easily see that the hypothesis (H) implies ![Formula][97] and therefore ![Formula][98] Defining ![Formula][99] we see from (33) that ![Formula][100] Letting ![Formula][101] and defining ![Formula][102] for *y* ∈ *C*[*x*, *x*1], we observe that *C*[*x*, *x*1] is the Banach space with norm ∥ · ∥. We define *Y* to be the set of functions *y* ∈ *C*[*x*, *x*1] satisfying ![Formula][103] We note that the set *Y* is a nonempty closed convex subset of *C*[*x*, *x*1]. Let φ denote the mapping defined by ![Formula][104] for *y* ∈ *Y*. Then it is easy to see that Φ [*y*] is non-decreasing for *y*, i.e., Φ [*y*1](*x*) ≤ Φ [*y*2](*x*) if 0 *< y*1(*x*) ≤ *y*2(*x*). Hence we deduce that ![Formula][105] holds for *y* ∈ *Y* by taking account of the inequality ![Formula][106] based on the mean value theorem and using (33). A simple computation shows that ![Formula][107] for *y* ∈ *Y*. Therefore, Φ maps *Y* into *Y*, i.e. Φ [*Y*] ⊂ *Y*, where ![Formula][108] It is easily seen that ![Formula][109] for *y*1, *y*2 ∈ *Y*. Since the inequality (31) holds for *c* replaced by *K**/*2, we get ![Formula][110] Where ![Formula][111] Hence we obtain ![Formula][112] which means that Φ : *Y* → *Y* is a continuous function. It is easy to check that Φ [*Y*] is uniformly bounded in view of the fact that ![Formula][113] for any *y* ∈ *Y*. Since ![Formula][114] we have ![Formula][115] Hence, for any given *ε >* 0 there exists a positive number *δ* = *δ*(*ε*) = *ε/*(1 + *K*1) such that |*x*1 − *x*2| *< δ* implies | Φ [*y*](*x*1) − Φ [*y*](*x*2)| *< ε* for any *y* ∈ *Y*, showing that Φ [*Y*] is equicontinuous. It follows from AscoliArzel’a theorem that Φ [*Y*] is a relatively compact subset of *C*[*x*, *x*1] because Φ [*Y*] is uniformly bounded and equicontinuous. Hence all the hypotheses of Schauder’s fixed-point theorem (see e.g. [6]) are satisfied, so φ has a fixed point *y* ∈ *Y*. Clearly *y*(*x*) is a positive solution of (29). Combining Theorem 2 with Theorem 3 yields the following theorem. Theorem 4. *Under the hypothesis* (H) *the integral equation* (29) *has a unique positive solution*. **Remark 3**. The hypothesis (H) is equivalent to ![Formula][116] where the right hand side of the above is greater than *x* in light of the condition *K* *>* 2*x*. ## 4 Explicit solution of an initial value problem for SVIR differential system In this section we derive an explicit solution of the initial value problem (1)– (5) based on the result of Section 3, and we show that the parametric solution obtained in Section 2 can be obtained by solving some linear differential system. First we investigate the integral equation ![Formula][117] which is equivalent to (16). Lemma 3. *If ψ*(*u*) *is a solution of the integral equation* (34) *such that ψ*(*u*) *>* 0 *in* ![Graphic][118]] *for some θ >* 0, *then w*(*x*) := (1*/β*)*ψ*(*e*−(*β/γ*)*x*) *is a solution of the integral equation* ![Formula][119] *such that w*(*x*) *>* 0 *in* ![Graphic][120]. *Conversely, if w*(*x*) *is a positive solution of* (35) *in* ![Graphic][121], *then ψ*(*u*) := *βw*(−(*γ/β*) log *u*) *is a positive solution of* (34) *in* (*θ*,![Graphic][122]]. *Proof*. Let *ψ*(*u*) be a positive solution of (34) in (*θ*, ![Graphic][123]]. Letting *u* = *e*−(*β/γ*)*x* in (34), we obtain ![Formula][124] It can be shown that ![Formula][125] by the change of variables *v* = *e*−(*β/γ*)*s*. Since ![Formula][126] by the transformation *ξ* = *e*−(*β/γ*)*z*, we get ![Formula][127] by substituting (38) into (37). We combine (36) with (39) to obtain ![Formula][128] or ![Formula][129] which is equivalent to (35). It is obvious that *w*(*x*) *>* 0 in [![Graphic][130], −(*γ/β*) log *θ*). Conversely, let *w*(*x*) be a positive solution of (35) in [![Graphic][131], −(*γ/β*) log *θ*). Putting *x* = −(*γ/β*) log *u* into (35) yields ![Formula][132] By the change of variables *z* = −(*γ/β*) log *ξ* we have ![Formula][133] Combining (40) with (41), we arrive at ![Formula][134] We observe, by the transformation *s* = −(*γ/β*) log *v*, that ![Formula][135] We see from (42) and (43) that ![Formula][136] and defining *ψ*(*u*) := *βw*(−(*γ/β*) log *u*) yields ![Formula][137] which is the integral equation (34). It is easy to check that *ψ*(*u*) *>* 0 in ![Graphic][138]. Lemma 4. *Under the assumption* (A2), *the transcendental equation* ![Formula][139] *has a unique solution x* = *ω such that* ![Formula][140] *Proof*. It follows from the assumption (A2) that ![Formula][141] We define the sequence ![Graphic][142] by ![Formula][143] It is easy to see that ![Formula][144] If *a**n*+1 ≥ *a**n*, then ![Formula][145] Therefore we see that *a**n*+2 ≥ *a**n*+1, and hence the sequence {*a**n*} is nondecreasing by the mathematical induction. We observe that the sequence {*a**n*} is bounded since ![Formula][146] Consequently there exists lim*n*→∞ *a**n*, and we define *ω* := lim*n*→∞ *a**n*. Taking the limit as *n* → ∞ in (45), we obtain ![Formula][147] The uniqueness of *ω* follows from the fact that the straight line ![Graphic][148] and the exponential curve ![Graphic][149] has only one intersecting point in ![Graphic][150] in view of the inequality ![Graphic][151], and hence the solution *ω* of the transcendental equation (44) is unique. The inequality ![Graphic][152] follows from the inequalities ![Formula][153] Lemma 5. *If we define f* (*x*) *by* ![Formula][154] *then we find that* ![Formula][155] *and that* ![Graphic][156] *and f* (*ω*) = 0. *Proof*. It is clear that ![Formula][157] and *f* (*ω*) = 0 by Lemma 4. A direct calculation shows that ![Formula][158] and that *f* *′*(*x*) = 0 if and only if ![Graphic][159]. Since *f* *′*(*x*) *>* 0 if ![Graphic][160] and *f* *′*(*x*) *<* 0 if ![Graphic][161], we observe that *f* (*x*) is increasing in ![Graphic][162] and decreasing in ![Graphic][163]. In view of the fact that ![Graphic][164] and *f* (*ω*) = 0, we conclude that *f* (*x*) *>* 0 in ![Graphic][165] (cf. Figure 2). ![Figure 1:](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2024/09/18/2024.09.16.24313772/F1.medium.gif) [Figure 1:](http://medrxiv.org/content/early/2024/09/18/2024.09.16.24313772/F1) Figure 1: Flow diagram of the SVIR epidemic model. ![Figure 2:](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2024/09/18/2024.09.16.24313772/F2.medium.gif) [Figure 2:](http://medrxiv.org/content/early/2024/09/18/2024.09.16.24313772/F2) Figure 2: Variation of ![Graphic][166] for *N* = 1000, ![Graphic][167], *β* = 0.3*/*1000, *γ* = 0.05. In this case we find that ![Graphic][168] and ![Graphic][169]. Theorem 5. *If w*(*x*) *is a positive solution of* (35) *in* [![Graphic][170], *r*) *for some* ![Graphic][171], *then w*(*x*) *is a positive solution of the initial value problem* ![Formula][172] ![Formula][173] ![Formula][174] *Conversely, if w*(*x*) *is a positive solution of the initial value problem* (46)– (48), *then w*(*x*) *is a positive solution of* (35) *in* ![Graphic][175] (cf. Figure 3). ![Figure 3:](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2024/09/18/2024.09.16.24313772/F3.medium.gif) [Figure 3:](http://medrxiv.org/content/early/2024/09/18/2024.09.16.24313772/F3) Figure 3: Variations of *f* (*x*) (dashed curve) and *w*(*x*) (solid curve) obtained by the numerical integration of the initial value problem (46)–(48) for *N* = 1000, ![Graphic][176], *β* = 0.3*/*1000, *γ* = 0.05 and *σ* = 0.008. In this case we obtain *ω* = 997.48 …, *α* = 795.93 …, ![Graphic][177] and ![Graphic][178]. *Proof*. Let *w*(*x*) be a positive solution of (35) in ![Graphic][179]. A direct calculation shows that ![Formula][180] and a differentiation of (49) yields ![Formula][181] Combining (49) with (50), we obtain ![Formula][182] and therefore ![Formula][183] which is equivalent to (46). It is clear that ![Formula][184] We note that ![Graphic][185] in light of the assumption (A1). Conversely, let *w*(*x*) be a positive solution of the initial value problem (46)–(48). Dividing (46) by *w*(*x*), we get (51), which is equivalent to ![Formula][186] Multiplying (52) by ![Graphic][187] yields, ![Formula][188] and therefore ![Formula][189] for some constant *C*. We observe, using (48), that ![Formula][190] Or ![Formula][191] Integrating (53) over ![Graphic][192], we obtain ![Formula][193] which, combined with (47) and (54), yields ![Formula][194] Since ![Graphic][195], we see that (55) is equivalent to (35). We need the following continuation proposition to prove Lemma 6. Proposition 1. *We consider the ordinary differential equation* ![Formula][196] *where F* (*x, w, y*) *is a continuous function defined in a domain D* ⊂ ℝ × ℝ2, *and let w*(*x*) *be a solution of* (*) *in some interval* ![Graphic][197]. *Assume that the limits* ![Formula][198] *exist. If* ![Graphic][199], *then w*(*x*) *can be continued to the right of r*. *Proof*. Let ![Graphic][200]. Since *D* is a domain, there exist *a >* 0, *b*1 *>* 0 and *b*2 *>* 0 such that ![Formula][201] It follows from Peano’s existence theorem (see e.g. Hartman [9, Chapter II, Theorem 2.1]) that the initial value problem ![Formula][202] possesses a solution *g*(*x*) defined on some interval *r* ≤ *x* ≤ *r* + *δ*. Letting *h*(*x*) := *g**′*(*x*) (*r* ≤ *x* ≤ *r* + *δ*), we obtain ![Formula][203] Let *x* be a point in ![Graphic][204], and let *y*(*x*) := *w**′*(*x*) (*x* ≤ *x < r*). Then we get ![Formula][205] If we define *W* (*x*) and *Y* (*x*) by ![Formula][206] then it is easy to check that *W* (*x*) and *Y* (*x*) are continuous on *x* ≤ *x* ≤ *r*+*δ*, and satisfy the relations ![Formula][207] Therefore we conclude that *W* (*x*) is a solution of (*) on the interval [*x*, *r*+*δ*] by differentiating the above relations, and that *W* (*x*) = *w*(*x*) (*x* ≤ *x < r*), showing that *W* (*x*) is a continuation of *w*(*x*). Lemma 6. *There exists a positive number α* ≤ *ω such that the integral equation* (35) *has a unique positive solution w*(*x*) *in* ![Graphic][208] *with the property that* lim*x*→*α*−0 *w*(*x*) = 0. *Proof*. It follows from Theorem 4 that there exists a unique positive solution *w*(*x*) of (35) on ![Graphic][209] for some ![Graphic][210] because ![Graphic][211] is satisfied by the assumptions (A3) and (A5). Then we easily see that ![Graphic][212] and −∞ *< w**′*(*x*1) *<* ∞. Moreover, Theorem 5 implies that *w*(*x*) is a positive solution of the initial value problem for the equation ![Formula][213] subject to the initial conditions (47) and (48). Then the solution *w*(*x*) of (56) can be continued to the right of *x*1 by applying Proposition 1 with *r* = *x*1 for ![Formula][214] And ![Formula][215] Then, the continuation of *w*(*x*) is unique. Let ![Graphic][216] be the maximal interval of existence of the continuation of *w*(*x*). We represent the continuation of *w*(*x*) by the same *w*(*x*). Then it is obvious that ![Formula][217] Since ![Formula][218] we see from (35) that ![Formula][219] Lemma 5 implies that ![Formula][220] and therefore *x*1 ≤ *α* ≤ *ω*. We observe, using Theorem 5, that *w*(*x*) satisfies (35) in ![Graphic][221]. Letting *x* → *α* − 0 in (35), we obtain ![Formula][222] Taking the limit as *x* → *α* − 0 in (49) yields ![Formula][223] We note that there exist *w*−(*α*) and *w**′* (*α*) even if either ![Graphic][224] or ![Graphic][225]. Then 0 ≤ *w*−(*α*) *<* ∞, −∞ *< w**′* (*α*) *<* ∞ and *w*(*x*) is a solution of ![Formula][226] If *w*−(*α*) *>* 0, then Proposition 1 with *r* = *α* implies that *w*(*x*) can be continued to the right of *α*. This contradicts the fact that ![Graphic][227] is the maximal interval of existence of *w*(*x*). Hence we conclude that *w*−(*α*) = 0, i.e. ![Formula][228] Theorem 6. *There exists a positive number α* ≤ *ω such that the integral equation* (34) *has a unique positive solution ψ*(*u*) *in* ![Graphic][229] *with the properties that* ![Graphic][230] *and* ![Formula][231] *Proof*. It follows from Lemma 6 that there exists a unique positive solution *w*(*x*) of (35) in ![Graphic][232] such that lim*x*→*α*−0 *w*(*x*) = 0. Then Lemma 3 implies that *ψ*(*u*) := *βw*(−(*γ/β*) log *u*) is a unique positive solution of (34) in ![Graphic][233]. It is clear that ![Formula][234] Theorem 7. *If ψ*(*u*) *is a positive solution of* (34) *in* ![Graphic][235], *then ψ*(*u*) *is a positive solution of the initial value problem* ![Formula][236] ![Formula][237] ![Formula][238] *Conversely, if ψ*(*u*) *is a positive solution of the initial value problem* (58)– (60), *then ψ*(*u*) *is a positive solution of* (34) *in* ![Graphic][239] (*cf*. Figure 4). ![Figure 4:](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2024/09/18/2024.09.16.24313772/F4.medium.gif) [Figure 4:](http://medrxiv.org/content/early/2024/09/18/2024.09.16.24313772/F4) Figure 4: Variations of ![Graphic][240] (dashed curve) and *ψ*(*u*) (solid curve) obtained by the numerical integration of the initial value problem (58)–(60) for *N* = 1000, ![Graphic][241], *β* = 0.3*/*1000, *γ* = 0.05 and *σ* = 0.008. In this case we get *α* = 795.93 …, *e*−(*β/γ*)*α* = 0.00843 …, *ω* = 997.48 …, *e*−(*β/γ*)*ω* = 0.00251 …, ![Graphic][242] and ![Graphic][243]. Moreover, *P* (*e*−(*β/γ*)*ω*) = 0 and ![Graphic][244]. *Proof*. Let *ψ*(*u*) be a positive solution of (34) in ![Graphic][245]. Differentiating (34) with respect to *u*, we obtain ![Formula][246] and hence ![Formula][247] A combination of (61) and (62) yields ![Formula][248] and therefore ![Formula][249] which is the equation (58). It is obvious that ![Formula][250] and therefore (59) and (60) are satisfied. We note that ![Graphic][251] in view of the assumption (A1). Conversely, let *ψ*(*u*) be a positive solution of the initial value problem (58)–(60). Dividing (58) by *ψ*(*u*), we are led to (63) which is rewritten in the form ![Formula][252] Since ![Formula][253] it follows from (64) that ![Formula][254] Or ![Formula][255] Hence we get ![Formula][256] and therefore ![Formula][257] for some constant *C*. Using (60), we obtain ![Formula][258] and hence ![Formula][259] Consequently we find that (61) holds. Integrating (61) over ![Graphic][260] and then taking the limit as *ε* → +0, we are led to ![Formula][261] and therefore ![Formula][262] Since ![Graphic][263], we see that (65) is equivalent to (34). Lemma 7. *Let ψ*(*u*) *be the unique positive solution of the integral equation* (34) *in* ![Graphic][264], *and we define the function φ*(*u*) *by* ![Formula][265] *Then there exists the inverse function u* = *φ*−1(*t*) ∈ *C*1(0, ∞) ∩ *C*[0, ∞) *of t* = *φ*(*u*) *such that φ*−1(*t*) *is a decreasing function on* [0, ∞), ![Graphic][266] *and* lim*t*→∞ *φ*−1(*t*) = *e*−(*β/γ*)*α*. *Proof*. It is easily seen that ![Graphic][267], *φ*(*u*) is decreasing in ![Graphic][268] and ![Graphic][269]. Since ![Formula][270] by use of (61), we have ![Formula][271] and therefore ![Formula][272] in view of (57). Hence, there exists the inverse function *φ*−1(*t*) which has the desired properties. Now we state our main theorem. Theorem 8. *The function* (*S*(*t*), *V* (*t*), *I*(*t*),*R*(*t*)) *defined by* ![Formula][273] ![Formula][274] ![Formula][275] ![Formula][276] *is a solution of the initial value problem* (1)–(5), *where φ*(*v*) *and φ*−1(*t*) *are given in Lemma 7*. *Proof*. First we note that ![Formula][277] by means of (34) and (68). Differentiating (66) and using (70), we get ![Formula][278] which is the equation (1). We differentiate (68) to obtain ![Formula][279] by (70), and hence *I*(*t*) satisfies (3). We observe, utilizing (70), that ![Formula][280] which is the desired equation (4). A direct calculation shows that ![Formula][281] and hence (2) holds. It is easy to see that ![Formula][282] in light of ![Graphic][283], and therefore (5) is satisfied. Theorem 9. *Let* (*S*(*t*), *V* (*t*), *I*(*t*), *R*(*t*)) *be the explicit solution* (66)–(69) *of the initial value problem* (1)–(5), *and let* ![Formula][284] *Then*, ![Graphic][285] *satisfies the linear differential system* ![Formula][286] ![Formula][287] ![Formula][288] ![Formula][289] *For*![Graphic][290], *and the initial condition* ![Formula][291] ![Formula][292] ![Formula][293] ![Formula][294] *Proof*. It follows from (70) that ![Formula][295] and hence we obtain ![Formula][296] Since *S*(*t*) satisfies (1), we get ![Formula][297] and therefore ![Formula][298] by taking account of (80). Thus (71) is satisfied. We observe, using (3) and (80), that ![Formula][299] and hence (73) holds. Utilizing (4) and (80) yields ![Formula][300] which is the equation (74). Using (2), we get ![Formula][301] which is the desired equation (72). It is obvious that ![Formula][302] ![Formula][303] and therefore (75)–(78) are satisfied. Theorem 10. *Solving the initial value problem* (71)–(78), *we obtain the parametric solution* (11)–(14) *for* ![Graphic][304]. *Proof*. Since ![Formula][305] we see that ![Formula][306] and therefore ![Formula][307] for some constant *k*. It is readily seen that ![Formula][308] We find from (75) that ![Formula][309] by taking account of ![Graphic][310]. Therefore we conclude that ![Graphic][311], and consequently ![Formula][312] which is the relation (11). Solving (74), we obtain ![Formula][313] for some constant *k*. The initial condition (78) implies ![Formula][314] and hence *k* = 0. Therefore ![Formula][315] thus (14) is obtained. Employing (81), we get ![Formula][316] Integrating (82) over ![Graphic][317], we have ![Formula][318] Or ![Formula][319] in light of (77). Since ![Graphic][320], (83) is equivalent to (13). We observe, using (81), that ![Formula][321] Integrating the above over ![Graphic][322] yields ![Formula][323] and therefore we get ![Formula][324] which is equal to (12). **Remark 4**. The formula (79) implies that the number of infected individuals *I*(*t*) can be represented in the simple form ![Formula][325] ## 5 Various properties of the explicit solution In this section we investigate various properties of the explicit solution (66)– (69) of the initial value problem (1)–(5). Theorem 11. *Let R*(*t*) *be given by* (69). *Then we find that R*(∞) = *α, and that R*(*t*) *is an increasing function on* [0, ∞) *such that* ![Formula][326] *where α is the positive number defined in Lemma 6*. *Proof*. It can be shown that ![Formula][327] In view of the inequality ![Graphic][328], we observe that ![Formula][329] Or ![Formula][330] Since *φ*−1(*t*) is decreasing on [0, ∞), we see that *R*(*t*) is increasing on [0, ∞). Theorem 12. *Let S*(*t*) *be given by* (66). *Then we see that S*(∞) = 0, *and that S*(*t*) *is a decreasing function on* [0, ∞) *such that* ![Formula][331] *Proof*. Taking account of *φ*−1(∞) = lim*t*→∞ *φ*−1(*t*) = *e*−(*β/γ*)*α*, we deduce that ![Formula][332] Since ![Graphic][333], we obtain ![Formula][334] and therefore ![Formula][335] We conclude that *S*(*t*) is decreasing on [0, ∞) because both *φ*−1(*t*) and *e*−*σt* are decreasing on [0, ∞). Theorem 13. *Let I*(*t*) *be given by* (68). *Then it follows that I*(∞) = 0, *I*(*t*) *>* 0 *on* [0, ∞), *and that I*(*t*) *has the maximum* ![Formula][336] *at t* = *T, where T is the unique solution of the equation* ![Formula][337] *Moreover, it can be shown that T* = *S*−1(*γ/β*), *I*(*t*) *is increasing in* [0, *T*) *and is decreasing in* (*T*, ∞). *Proof*. It is easy to check that ![Formula][338] by means of (57) and (79). Since ![Graphic][339] for *t* ≥ 0 and *ψ*(*u*) *>* 0 for ![Graphic][340], we find that ![Formula][341] Differentiating both sides of (68) and taking account of (66) and (70), we obtain ![Formula][342] ![Formula][343] It is clear that *φ*−1(*t*)*e*−*σt* is a decreasing function on [0, ∞) such that ![Graphic][344] and lim*t*→∞ *φ*−1(*t*)*e*−*σt* = 0 by using the fact that lim*t*→∞ *φ*−1(*t*) = *e*−(*β/γ*)*α*. Since ![Formula][345] by the assumption (A1), there exists the unique solution *T* (*>* 0) such that ![Formula][346] Then, it follows from (85) that *I**′*(*T*) = 0. Taking account of the fact that (*φ*−1(*t*))*′* *<* 0 and *φ*−1(*t*)*e*−*σt* is decreasing on [0, ∞), we see from (85) that *I**′*(*t*) *>* 0 [resp. *<* 0] if and only if *t < T* [resp. *> T*]. Hence, *I*(*t*) is increasing in (0, *T*) and is decreasing in (*T*, ∞). Utilizing (86), we get *T* = *S*−1(*γ/β*). Employing (87), we obtain ![Formula][347] We easily see that ![Formula][348] and that ![Formula][349] by the transformation ![Graphic][350]. Combining (88)–(90) yields the desired relation (84). Theorem 14. *Let V* (*t*) *be given by* (67). *Then we deduce that* ![Formula][351] *and V* (*t*) *is an increasing function on* [0, ∞) *such that* ![Formula][352] *Proof*. In view of the fact that *S*(∞) = 0, *I*(∞) = 0 and *R*(∞) = *α*, it follows that ![Formula][353] Since *V* (*t*) satisfies (2) and *S*(*t*) *>* 0 for *t >* 0, we observe that *V* *′*(*t*) = *σS*(*t*) *>* 0 for *t >* 0, and hence *V* (*t*) is an increasing function on [0, ∞). Theorem 15. *The following relation holds:* ![Formula][354] *Proof*. Since *S*(∞) = *I*(∞) = 0, we conclude from (1)–(4) that ![Formula][355] **Remark 5**. It follows from Theorems 11–14 that *S*(*t*) *>* 0, *I*(*t*) *>* 0 for *t* ≥ 0, and that *R*(*t*) *>* 0, *V* (*t*) *>* 0 for *t >* 0 (cf. Figure 5). ![Figure 5:](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2024/09/18/2024.09.16.24313772/F5.medium.gif) [Figure 5:](http://medrxiv.org/content/early/2024/09/18/2024.09.16.24313772/F5) Figure 5: Variations of *S*(*t*), *V* (*t*), *I*(*t*) and *R*(*t*) obtained by the numerical integration of the initial value problem (1)–(5) for *N* = 1000, ![Graphic][356], *β* = 0.3*/*1000, *γ* = 0.05 and *σ* = 0.008. In this case we obtain ![Graphic][357], *R*(∞) = *α* = 795.9 …, *S*(∞) = *I*(∞) = 0 and *V* (∞) = 204.0 …. **Remark 6**. Noting ![Formula][358] we conclude that ![Graphic][359] is equivalent to ![Graphic][360]. If ![Graphic][361], it follows that ![Formula][362] Using (85) and (91), we find that ![Formula][363] ![Formula][364] in view of the fact that ![Graphic][365] and (*φ*−1(*t*))*′* *<* 0. Therefore, *I*(*t*) is a decreasing function on [0, ∞) such that ![Graphic][366] and *I*(∞) = 0 (cf. Figure 6). ![Figure 6:](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2024/09/18/2024.09.16.24313772/F6.medium.gif) [Figure 6:](http://medrxiv.org/content/early/2024/09/18/2024.09.16.24313772/F6) Figure 6: Variations of *S*(*t*), *V* (*t*), *I*(*t*) and *R*(*t*) obtained by the numerical integration of the initial value problem (1)–(5) for *N* = 1000, ![Graphic][367], *β* = 0.1*/*1000, *γ* = 0.1 and *σ* = 0.008. In this case we find that ![Graphic][368], *I*(*t*) is decreasing on [0, ∞) and *I*(∞) = 0. **Remark 7**. Let R denote the basic reproduction number defined by ![Formula][369] Under the hypotheses *γ >* 0 and ![Graphic][370], the basic reproduction number ℛ *>* 1 is equivalent to ![Formula][371] and ℛ ≤ 1 is equivalent to ![Graphic][372]. When R *>* 1, the number of infected individuals *I*(*t*) reaches a peak and then decreases to 0 (cf. Theorem 13). When ℛ ≤ 1, *I*(*t*) decreases on [0, ∞) and the disease vanishes at infinity, i.e. *I*(∞) = 0 (cf. Remark 6). It is obvious that the assumption (A1) is equivalent to ℛ *>* 1. **Remark 8**. The function *R*(*t*) given by (69) is a positive and increasing solution of the initial value problem for (6) with the initial condition ![Graphic][373]![Graphic][374]. In fact, Theorem 11 implies that *R*(*t*) is an increasing function on [0, ∞) such that *R*(*t*) *>* 0 for *t >* 0. It is obvious that ![Graphic][375]. It is readily verified that ![Formula][376] by (70), and that ![Formula][377] in view of (34). Since ![Formula][378] And ![Formula][379] by the change of variables *v* = *φ*−1(*s*), we get ![Formula][380] Combining (92)–(94) yields ![Formula][381] which is equal to (6). ## 6 Uniqueness of positive solutions In this section we study the uniqueness of positive solutions of the initial value problem (1)–(5). Our approach is based on that used in ordinary differential equations (cf. Coddington and Levinson [4] and Hartman [9]). For example we refer to Picard-Lindelöf theorem [9, Theorem 1.1]. We show that the explicit solution (66)–(69) is unique in the class of positive solutions. A solution (*S*(*t*), *V* (*t*), *I*(*t*), *R*(*t*)) of the SVIR differential system (1)–(4) is said to be *positive* if *S*(*t*) *>* 0, *V* (*t*) *>* 0, *I*(*t*) *>* 0 and *R*(*t*) *>* 0 for *t >* 0. Theorem 16. *Let R**i*(*t*) (*i* = 1, 2) *be solutions of the integro-differential equation* (6) *subject to the initial conditions* ![Graphic][382] *such that R**i*(*t*) *>* 0 *for t >* 0. *Then we conclude that* ![Formula][383] *Proof*. Integrating (6) with *R*(*t*) = *R**i*(*t*) over [*ε, t*] (*ε >* 0) and taking the limit as *ε* → +0, we obtain ![Formula][384] Since ![Formula][385] we get ![Formula][386] and therefore ![Formula][387] The above formula means that ![Formula][388] in light of the inequalities |*e*−*σs*| ≤ 1 and |*t* − *s*| ≤ *t*. Using the same arguments as in the proof of Theorem 2, we obtain ![Formula][389] Therefore it follows from (96) that ![Formula][390] Letting ![Formula][391] we have ![Formula][392] Or ![Formula][393] Where ![Formula][394] Multiplying (97) by ![Graphic][395] yields ![Formula][396] and hence ![Graphic][397] is nonincreasing on (0, ∞). Therefore ![Formula][398] which means ![Formula][399] Combining (97) with (98), we arrive at ![Formula][400] which implies ![Formula][401] Since ![Graphic][402], we conclude that ![Formula][403] Theorem 17. *Let* (*S**i*(*t*), *V**i*(*t*), *I**i*(*t*), *R**i*(*t*)) (*i* = 1, 2) *be solutions of the initial value problem* (1)–(5) *such that S**i*(*t*) *>* 0 *and I**i*(*t*) *>* 0 *for t >* 0. *Then we observe that* ![Formula][404] *Proof*. First we note that *R**i*(*t*) *>* 0 and *V**i*(*t*) *>* 0 for *t >* 0 (*i* = 1, 2) by Remark 1, and therefore (*S**i*(*t*), *V**i*(*t*), *I**i*(*t*), *R**i*(*t*)) (*i* = 1, 2) are positive solutions of the initial value problem (1)–(5). Corollary 1 implies that *S**i*(*t*) (*i* = 1, 2) can be written in the form ![Formula][405] Since *R*1(*t*) ≡ *R*2(*t*) on [0, ∞) by Theorem 16, we observe that *S*1(*t*) ≡ *S*2(*t*) on [0, ∞). Similarly we conclude that *I*1(*t*) ≡ *I*2(*t*) and *V*1(*t*) ≡ *V*2(*t*) on [0, ∞) by applying Corollary 1. Theorem 18. *The explicit solution* (*S*(*t*), *V* (*t*), *I*(*t*), *R*(*t*)) *given by* (66)– *is a unique solution of the initial value problem* (1)–(5) *in the class of positive solutions*. *Proof*. It follows from Remark 5 that the solution (*S*(*t*), *V* (*t*), *I*(*t*), *R*(*t*)) is a positive solution of the problem (1)–(5). Uniqueness in the class of positive solutions follows from Theorem 17. ## Data Availability All data produced in the present work are contained in the manuscript. ## Acknowledgements The author would like to thank Professor Manabu Naito for his great contribution to the results in Sections 3 and 4. * Received September 16, 2024. * Revision received September 16, 2024. * Accepted September 18, 2024. * © 2024, Posted by Cold Spring Harbor Laboratory The copyright holder for this pre-print is the author. All rights reserved. The material may not be redistributed, re-used or adapted without the author's permission. ## References 1. [1]. M. Bohner, S. Streipert and D. F. M. Torres, Exact solution to a dynamic SIR model, Nonlinear Anal. Hybrid Systems 32 (2019), 228–238. 2. [2]. F. Brauer, P. van den Driessche and J. Wu (Eds.) Mathematical Epidemiology, Lecture Notes in Mathematics, Vol. 1945, Springer-Verlag, 2008. 3. [3]. V. Capasso, Mathematical Structures of Epidemic Systems, Lecture Notes in Biomathematics, Vol. 97, Springer-Verlag, 1993. 4. [4]. E. A. Coddington and N. Levinson, Theory of Ordinary Differential Equations, McGraw-Hill, New York, 1955. 5. [5]. X. Duan, S. Yuan and X. Li, Global stability of an SVIR model with age of vaccination, Appl. Math. Comput. 226 (2014), 528–540. 6. [6]. R. E. Edwards, Functional Analysis: Theory and Applications, Holt, Rinehart and Winston, New York, 1965. 7. [7]. J. C. Frauenthal, Mathematical Modeling in Epidemiology, Springer-Verlag, Berlin, Heidelberg, 1980. 8. [8]. T. Harko, F. S. N. Lobo and M. K. Mak, Exact analytical solutions of the Susceptible-Infected-Recovered (SIR) epidemic model and of the SIR model with equal death and birth rates, Appl. Math. Comput. 236 (2014), 184–194. 9. [9]. P. Hartman, Ordinary Differential Equations, Birkhäuser, Boston, 1982. 10. [10]. W. O. Kermack and A. G. McKendrick, Contributions to the mathematical theory of epidemics, Part I, Proc. Roy. Soc. Lond. Ser. A 115 (1927), 700–721. 11. [11]. X. Liu, Y. Takeuchi and S. Iwami, SVIR Epidemic models with vaccination strategies, J. Theoret. Biol. 253 (2008), 1–11. 12. [12]. P. Magal and S. Ruan (Eds.) Structured Population Models in Biology and Epidemiology, Lecture Notes in Mathematics, Vol. 1936, Springer-Verlag, 2008. 13. [13]. T. T. Marinov and R. S. Marinova, Adaptive SIR model with vaccination: simultaneous identification of rates and functions illustrated with COVID-19, Scientific Reports 12 (2022):15688. 14. [14]. A. Nobe, Exact solutions to SIR epidemic models via integrable discretization, 2023, ⟨arXiv:2303.17198⟩. 15. [15]. M. O. Oke, O. M. Ogunmiloro, C. T. Akinwumi, S. O. Ayinde, T. O. Ogunlade and K. J. Adebayo, On the application of optimal control strategies to a generalized SVIR model, J. Phys. Conf. Ser. 1734 (2021) 012051. 16. [16]. R. Schlickeiser and M. Kröger, Analytic modeling of the temporal evolution of epidemic outbreaks accounting for vaccinations, Physics 2021, 3, 386–426. 17. [17]. R. Schlickeiser and M. Kröger, Key epidemic parameters of the SIRV model determined from past COVID-19 mutant waves, COVID 2023, 3(4), 592–600. 18. [18]. G. Shabbir, H. Khan and M. A. Sadiq, A note on exact solution of SIR and SIS epidemic models, 2010, ⟨arXiv:1012.5035⟩. 19. [19]. J. Wang, M. Guo and S. Liu, SVIR Epidemic model with age structure in susceptibility-vaccination effects and relapse, IMA J. Appl. Math. 82 (2018), 945–970. 20. [20]. J. Wang, R. Zhang and T. Kuniya, The dynamics of an SVIR epidemiological model with infection age, IMA J. Appl. Math. 81 (2016), 321–343. 21. [21]. J. Wang, R. Zhang and T. Kuniya, A reaction-diffusion Susceptible-Vaccinated-Infected-Recovered model in a spatially heterogeneous environment with Dirichlet boundary condition, Math. Comput. Simulation 190 (2021), 848–865. 22. [22]. N. Yoshida, Exact solution of the Susceptible-Infectious-Recovered-Deceased (SIRD) epidemic model, Electron. J. Qual. Theory Differ. Equ. 2022, No. 38, 1–24. 23. [23]. N. Yoshida, Exact solution of the Susceptible-Infectious-Recovered (SIR) epidemic model, 2022, ⟨arXiv:2210.00444⟩. 24. [24]. N. Yoshida, Existence of exact solution of the Susceptible-Exposed-Infectious-Recovered (SEIR) epidemic model, J. Differential Equations 355 (2023), 103–143. 25. [25]. N. Yoshida, Exact solution of the Susceptible-Exposed-Infectious-Recovered-Deceased (SEIRD) epidemic model, Electron. J. Qual. Theory Differ. Equ. 2024, No. 8, 1–37. [1]: /embed/graphic-1.gif [2]: /embed/graphic-2.gif [3]: /embed/graphic-3.gif [4]: /embed/graphic-4.gif [5]: /embed/graphic-5.gif [6]: /embed/inline-graphic-1.gif [7]: /embed/inline-graphic-2.gif [8]: /embed/inline-graphic-3.gif [9]: /embed/graphic-6.gif [10]: /embed/inline-graphic-4.gif [11]: /embed/inline-graphic-5.gif [12]: /embed/inline-graphic-6.gif [13]: /embed/graphic-7.gif [14]: /embed/inline-graphic-7.gif [15]: /embed/graphic-8.gif [16]: /embed/graphic-9.gif [17]: /embed/graphic-10.gif [18]: /embed/graphic-11.gif [19]: /embed/graphic-12.gif [20]: /embed/graphic-13.gif [21]: /embed/graphic-14.gif [22]: /embed/graphic-15.gif [23]: /embed/graphic-16.gif [24]: /embed/graphic-17.gif [25]: /embed/graphic-18.gif [26]: /embed/graphic-19.gif [27]: /embed/graphic-20.gif [28]: /embed/graphic-21.gif [29]: /embed/graphic-22.gif [30]: /embed/inline-graphic-8.gif [31]: /embed/inline-graphic-9.gif [32]: /embed/graphic-23.gif [33]: /embed/graphic-24.gif [34]: /embed/graphic-25.gif [35]: /embed/graphic-26.gif [36]: /embed/graphic-27.gif [37]: /embed/graphic-28.gif [38]: /embed/graphic-29.gif [39]: /embed/graphic-30.gif [40]: /embed/graphic-31.gif [41]: /embed/graphic-32.gif [42]: /embed/graphic-33.gif [43]: /embed/inline-graphic-10.gif [44]: /embed/inline-graphic-11.gif [45]: /embed/inline-graphic-12.gif [46]: /embed/graphic-34.gif [47]: /embed/inline-graphic-13.gif [48]: /embed/inline-graphic-14.gif [49]: /embed/graphic-35.gif [50]: /embed/inline-graphic-15.gif [51]: /embed/graphic-36.gif [52]: /embed/graphic-37.gif [53]: /embed/graphic-38.gif [54]: /embed/graphic-39.gif [55]: /embed/graphic-40.gif [56]: /embed/graphic-41.gif [57]: /embed/graphic-42.gif [58]: /embed/graphic-43.gif [59]: /embed/graphic-44.gif [60]: /embed/graphic-45.gif [61]: /embed/graphic-46.gif [62]: /embed/inline-graphic-16.gif [63]: /embed/inline-graphic-17.gif [64]: /embed/graphic-47.gif [65]: /embed/inline-graphic-18.gif [66]: /embed/inline-graphic-19.gif [67]: /embed/graphic-48.gif [68]: /embed/graphic-49.gif [69]: /embed/graphic-50.gif [70]: /embed/graphic-51.gif [71]: /embed/graphic-52.gif [72]: /embed/inline-graphic-20.gif [73]: /embed/inline-graphic-21.gif [74]: /embed/graphic-53.gif [75]: /embed/graphic-54.gif [76]: /embed/inline-graphic-22.gif [77]: /embed/graphic-55.gif [78]: /embed/graphic-56.gif [79]: /embed/graphic-57.gif [80]: /embed/inline-graphic-23.gif [81]: /embed/inline-graphic-24.gif [82]: /embed/graphic-58.gif [83]: /embed/graphic-59.gif [84]: /embed/graphic-60.gif [85]: /embed/graphic-61.gif [86]: /embed/graphic-62.gif [87]: /embed/graphic-63.gif [88]: /embed/graphic-64.gif [89]: /embed/graphic-65.gif [90]: /embed/graphic-66.gif [91]: /embed/graphic-67.gif [92]: /embed/graphic-68.gif [93]: /embed/graphic-69.gif [94]: /embed/graphic-70.gif [95]: /embed/graphic-71.gif [96]: /embed/graphic-72.gif [97]: /embed/graphic-73.gif [98]: /embed/graphic-74.gif [99]: /embed/graphic-75.gif [100]: /embed/graphic-76.gif [101]: /embed/graphic-77.gif [102]: /embed/graphic-78.gif [103]: /embed/graphic-79.gif [104]: /embed/graphic-80.gif [105]: /embed/graphic-81.gif [106]: /embed/graphic-82.gif [107]: /embed/graphic-83.gif [108]: /embed/graphic-84.gif [109]: /embed/graphic-85.gif [110]: /embed/graphic-86.gif [111]: /embed/graphic-87.gif [112]: /embed/graphic-88.gif [113]: /embed/graphic-89.gif [114]: /embed/graphic-90.gif [115]: /embed/graphic-91.gif [116]: /embed/graphic-92.gif [117]: /embed/graphic-93.gif [118]: /embed/inline-graphic-25.gif [119]: /embed/graphic-94.gif [120]: /embed/inline-graphic-26.gif [121]: /embed/inline-graphic-27.gif [122]: /embed/inline-graphic-28.gif [123]: /embed/inline-graphic-29.gif [124]: /embed/graphic-95.gif [125]: /embed/graphic-96.gif [126]: /embed/graphic-97.gif [127]: /embed/graphic-98.gif [128]: /embed/graphic-99.gif [129]: /embed/graphic-100.gif [130]: /embed/inline-graphic-30.gif [131]: /embed/inline-graphic-31.gif [132]: /embed/graphic-101.gif [133]: /embed/graphic-102.gif [134]: /embed/graphic-103.gif [135]: /embed/graphic-104.gif [136]: /embed/graphic-105.gif [137]: /embed/graphic-106.gif [138]: /embed/inline-graphic-32.gif [139]: /embed/graphic-107.gif [140]: /embed/graphic-108.gif [141]: /embed/graphic-109.gif [142]: /embed/inline-graphic-33.gif [143]: /embed/graphic-110.gif [144]: /embed/graphic-111.gif [145]: /embed/graphic-112.gif [146]: /embed/graphic-113.gif [147]: /embed/graphic-114.gif [148]: /embed/inline-graphic-34.gif [149]: /embed/inline-graphic-35.gif [150]: /embed/inline-graphic-36.gif [151]: /embed/inline-graphic-37.gif [152]: /embed/inline-graphic-38.gif [153]: /embed/graphic-115.gif [154]: /embed/graphic-116.gif [155]: /embed/graphic-117.gif [156]: /embed/inline-graphic-39.gif [157]: /embed/graphic-118.gif [158]: /embed/graphic-119.gif [159]: /embed/inline-graphic-40.gif [160]: /embed/inline-graphic-41.gif [161]: /embed/inline-graphic-42.gif [162]: /embed/inline-graphic-43.gif [163]: /embed/inline-graphic-44.gif [164]: /embed/inline-graphic-45.gif [165]: /embed/inline-graphic-46.gif [166]: F2/embed/inline-graphic-47.gif [167]: F2/embed/inline-graphic-48.gif [168]: F2/embed/inline-graphic-49.gif [169]: F2/embed/inline-graphic-50.gif [170]: /embed/inline-graphic-51.gif [171]: /embed/inline-graphic-52.gif [172]: /embed/graphic-122.gif [173]: /embed/graphic-123.gif [174]: /embed/graphic-124.gif [175]: /embed/inline-graphic-53.gif [176]: F3/embed/inline-graphic-54.gif [177]: F3/embed/inline-graphic-55.gif [178]: F3/embed/inline-graphic-56.gif [179]: /embed/inline-graphic-57.gif [180]: /embed/graphic-126.gif [181]: /embed/graphic-127.gif [182]: /embed/graphic-128.gif [183]: /embed/graphic-129.gif [184]: /embed/graphic-130.gif [185]: /embed/inline-graphic-58.gif [186]: /embed/graphic-131.gif [187]: /embed/inline-graphic-59.gif [188]: /embed/graphic-132.gif [189]: /embed/graphic-133.gif [190]: /embed/graphic-134.gif [191]: /embed/graphic-135.gif [192]: /embed/inline-graphic-60.gif [193]: /embed/graphic-136.gif [194]: /embed/graphic-137.gif [195]: /embed/inline-graphic-61.gif [196]: /embed/graphic-138.gif [197]: /embed/inline-graphic-62.gif [198]: /embed/graphic-139.gif [199]: /embed/inline-graphic-63.gif [200]: /embed/inline-graphic-64.gif [201]: /embed/graphic-140.gif [202]: /embed/graphic-141.gif [203]: /embed/graphic-142.gif [204]: /embed/inline-graphic-65.gif [205]: /embed/graphic-143.gif [206]: /embed/graphic-144.gif [207]: /embed/graphic-145.gif [208]: /embed/inline-graphic-66.gif [209]: /embed/inline-graphic-67.gif [210]: /embed/inline-graphic-68.gif [211]: /embed/inline-graphic-69.gif [212]: /embed/inline-graphic-70.gif [213]: /embed/graphic-146.gif [214]: /embed/graphic-147.gif [215]: /embed/graphic-148.gif [216]: /embed/inline-graphic-71.gif [217]: /embed/graphic-149.gif [218]: /embed/graphic-150.gif [219]: /embed/graphic-151.gif [220]: /embed/graphic-152.gif [221]: /embed/inline-graphic-72.gif [222]: /embed/graphic-153.gif [223]: /embed/graphic-154.gif [224]: /embed/inline-graphic-73.gif [225]: /embed/inline-graphic-74.gif [226]: /embed/graphic-155.gif [227]: /embed/inline-graphic-75.gif [228]: /embed/graphic-156.gif [229]: /embed/inline-graphic-76.gif [230]: /embed/inline-graphic-77.gif [231]: /embed/graphic-157.gif [232]: /embed/inline-graphic-78.gif [233]: /embed/inline-graphic-79.gif [234]: /embed/graphic-158.gif [235]: /embed/inline-graphic-80.gif [236]: /embed/graphic-159.gif [237]: /embed/graphic-160.gif [238]: /embed/graphic-161.gif [239]: /embed/inline-graphic-81.gif [240]: F4/embed/inline-graphic-82.gif [241]: F4/embed/inline-graphic-83.gif [242]: F4/embed/inline-graphic-84.gif [243]: F4/embed/inline-graphic-85.gif [244]: F4/embed/inline-graphic-86.gif [245]: /embed/inline-graphic-87.gif [246]: /embed/graphic-163.gif [247]: /embed/graphic-164.gif [248]: /embed/graphic-165.gif [249]: /embed/graphic-166.gif [250]: /embed/graphic-167.gif [251]: /embed/inline-graphic-88.gif [252]: /embed/graphic-168.gif [253]: /embed/graphic-169.gif [254]: /embed/graphic-170.gif [255]: /embed/graphic-171.gif [256]: /embed/graphic-172.gif [257]: /embed/graphic-173.gif [258]: /embed/graphic-174.gif [259]: /embed/graphic-175.gif [260]: /embed/inline-graphic-89.gif [261]: /embed/graphic-176.gif [262]: /embed/graphic-177.gif [263]: /embed/inline-graphic-90.gif [264]: /embed/inline-graphic-91.gif [265]: /embed/graphic-178.gif [266]: /embed/inline-graphic-92.gif [267]: /embed/inline-graphic-93.gif [268]: /embed/inline-graphic-94.gif [269]: /embed/inline-graphic-95.gif [270]: /embed/graphic-179.gif [271]: /embed/graphic-180.gif [272]: /embed/graphic-181.gif [273]: /embed/graphic-182.gif [274]: /embed/graphic-183.gif [275]: /embed/graphic-184.gif [276]: /embed/graphic-185.gif [277]: /embed/graphic-186.gif [278]: /embed/graphic-187.gif [279]: /embed/graphic-188.gif [280]: /embed/graphic-189.gif [281]: /embed/graphic-190.gif [282]: /embed/graphic-191.gif [283]: /embed/inline-graphic-96.gif [284]: /embed/graphic-192.gif [285]: /embed/inline-graphic-97.gif [286]: /embed/graphic-193.gif [287]: /embed/graphic-194.gif [288]: /embed/graphic-195.gif [289]: /embed/graphic-196.gif [290]: /embed/inline-graphic-98.gif [291]: /embed/graphic-197.gif [292]: /embed/graphic-198.gif [293]: /embed/graphic-199.gif [294]: /embed/graphic-200.gif [295]: /embed/graphic-201.gif [296]: /embed/graphic-202.gif [297]: /embed/graphic-203.gif [298]: /embed/graphic-204.gif [299]: /embed/graphic-205.gif [300]: /embed/graphic-206.gif [301]: /embed/graphic-207.gif [302]: /embed/graphic-208.gif [303]: /embed/graphic-209.gif [304]: /embed/inline-graphic-99.gif [305]: /embed/graphic-210.gif [306]: /embed/graphic-211.gif [307]: /embed/graphic-212.gif [308]: /embed/graphic-213.gif [309]: /embed/graphic-214.gif [310]: /embed/inline-graphic-100.gif [311]: /embed/inline-graphic-101.gif [312]: /embed/graphic-215.gif [313]: /embed/graphic-216.gif [314]: /embed/graphic-217.gif [315]: /embed/graphic-218.gif [316]: /embed/graphic-219.gif [317]: /embed/inline-graphic-102.gif [318]: /embed/graphic-220.gif [319]: /embed/graphic-221.gif [320]: /embed/inline-graphic-103.gif [321]: /embed/graphic-222.gif [322]: /embed/inline-graphic-104.gif [323]: /embed/graphic-223.gif [324]: /embed/graphic-224.gif [325]: /embed/graphic-225.gif [326]: /embed/graphic-226.gif [327]: /embed/graphic-227.gif [328]: /embed/inline-graphic-105.gif [329]: /embed/graphic-228.gif [330]: /embed/graphic-229.gif [331]: /embed/graphic-230.gif [332]: /embed/graphic-231.gif [333]: /embed/inline-graphic-106.gif [334]: /embed/graphic-232.gif [335]: /embed/graphic-233.gif [336]: /embed/graphic-234.gif [337]: /embed/graphic-235.gif [338]: /embed/graphic-236.gif [339]: /embed/inline-graphic-107.gif [340]: /embed/inline-graphic-108.gif [341]: /embed/graphic-237.gif [342]: /embed/graphic-238.gif [343]: /embed/graphic-239.gif [344]: /embed/inline-graphic-109.gif [345]: /embed/graphic-240.gif [346]: /embed/graphic-241.gif [347]: /embed/graphic-242.gif [348]: /embed/graphic-243.gif [349]: /embed/graphic-244.gif [350]: /embed/inline-graphic-110.gif [351]: /embed/graphic-245.gif [352]: /embed/graphic-246.gif [353]: /embed/graphic-247.gif [354]: /embed/graphic-248.gif [355]: /embed/graphic-249.gif [356]: F5/embed/inline-graphic-111.gif [357]: F5/embed/inline-graphic-112.gif [358]: /embed/graphic-251.gif [359]: /embed/inline-graphic-113.gif [360]: /embed/inline-graphic-114.gif [361]: /embed/inline-graphic-115.gif [362]: /embed/graphic-252.gif [363]: /embed/graphic-253.gif [364]: /embed/graphic-254.gif [365]: /embed/inline-graphic-116.gif [366]: /embed/inline-graphic-117.gif [367]: F6/embed/inline-graphic-118.gif [368]: F6/embed/inline-graphic-119.gif [369]: /embed/graphic-256.gif [370]: /embed/inline-graphic-120.gif [371]: /embed/graphic-257.gif [372]: /embed/inline-graphic-121.gif [373]: /embed/inline-graphic-122.gif [374]: /embed/inline-graphic-123.gif [375]: /embed/inline-graphic-124.gif [376]: /embed/graphic-258.gif [377]: /embed/graphic-259.gif [378]: /embed/graphic-260.gif [379]: /embed/graphic-261.gif [380]: /embed/graphic-262.gif [381]: /embed/graphic-263.gif [382]: /embed/inline-graphic-125.gif [383]: /embed/graphic-264.gif [384]: /embed/graphic-265.gif [385]: /embed/graphic-266.gif [386]: /embed/graphic-267.gif [387]: /embed/graphic-268.gif [388]: /embed/graphic-269.gif [389]: /embed/graphic-270.gif [390]: /embed/graphic-271.gif [391]: /embed/graphic-272.gif [392]: /embed/graphic-273.gif [393]: /embed/graphic-274.gif [394]: /embed/graphic-275.gif [395]: /embed/inline-graphic-126.gif [396]: /embed/graphic-276.gif [397]: /embed/inline-graphic-127.gif [398]: /embed/graphic-277.gif [399]: /embed/graphic-278.gif [400]: /embed/graphic-279.gif [401]: /embed/graphic-280.gif [402]: /embed/inline-graphic-128.gif [403]: /embed/graphic-281.gif [404]: /embed/graphic-282.gif [405]: /embed/graphic-283.gif