{"id":190,"date":"2010-06-22T07:45:40","date_gmt":"2010-06-22T11:45:40","guid":{"rendered":"http:\/\/blogs.williams.edu\/Morgan\/?p=190"},"modified":"2019-05-09T05:04:09","modified_gmt":"2019-05-09T10:04:09","slug":"variation-formulae-for-perimeter-and-volume-densities","status":"publish","type":"post","link":"https:\/\/sites.williams.edu\/Morgan\/2010\/06\/22\/variation-formulae-for-perimeter-and-volume-densities\/","title":{"rendered":"Variation Formulae for Perimeter and Volume Densities"},"content":{"rendered":"<p>For <strong>R<\/strong><sup>n+1<\/sup> with volume density <em>f<\/em> and perimeter density <em>g<\/em>, for a normal variation <em>u<\/em> of a surface with classical mean curvature <em>H<\/em>, the first variation of volume and perimeter are given by:<\/p>\n<p><img src='https:\/\/s0.wp.com\/latex.php?latex=%5Cdelta+%5E1V%3D-%5Cint+uf&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='\\delta ^1V=-\\int uf' title='\\delta ^1V=-\\int uf' class='latex' \/>,<\/p>\n<p><img src='https:\/\/s0.wp.com\/latex.php?latex=%5Cdelta+%5E1P%3D-%5Cint+%5B%28g%2Ff%29nH+-+%281%2Ff%29%28%5Cpartial+g%2F%5Cpartial+n%29%5D+uf&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='\\delta ^1P=-\\int [(g\/f)nH - (1\/f)(\\partial g\/\\partial n)] uf' title='\\delta ^1P=-\\int [(g\/f)nH - (1\/f)(\\partial g\/\\partial n)] uf' class='latex' \/>.<\/p>\n<p>For a volume-preserving variation, the second variation of perimeter is given by:<br \/>\n<img src='https:\/\/s0.wp.com\/latex.php?latex=%5Cdelta+%5E2P%3D%5Cint+g%7C%5Cnabla+u%7C%5E2-g%7C%5Csigma%7C%5E2u%5E2-f%5Cfrac%7B%5Cpartial+%28g%2Ff%29%7D%7B%5Cpartial+n%7Du%5E2nH%2Bu%5E2%5Cfrac%7B%5Cpartial+%5E2g%7D%7B%5Cpartial+n%5E2%7D-%5Cfrac+1fu%5E2%5Cfrac%7B%5Cpartial+f%7D%7B%5Cpartial+n%7D%5Cfrac%7B%5Cpartial+g%7D%7B%5Cpartial+n%7D&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='\\delta ^2P=\\int g|\\nabla u|^2-g|\\sigma|^2u^2-f\\frac{\\partial (g\/f)}{\\partial n}u^2nH+u^2\\frac{\\partial ^2g}{\\partial n^2}-\\frac 1fu^2\\frac{\\partial f}{\\partial n}\\frac{\\partial g}{\\partial n}' title='\\delta ^2P=\\int g|\\nabla u|^2-g|\\sigma|^2u^2-f\\frac{\\partial (g\/f)}{\\partial n}u^2nH+u^2\\frac{\\partial ^2g}{\\partial n^2}-\\frac 1fu^2\\frac{\\partial f}{\\partial n}\\frac{\\partial g}{\\partial n}' class='latex' \/>,<\/p>\n<p>where <img src='https:\/\/s0.wp.com\/latex.php?latex=%5Csigma&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='\\sigma' title='\\sigma' class='latex' \/> is the second fundamental form, so that <img src='https:\/\/s0.wp.com\/latex.php?latex=%5Csigma%5E2&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='\\sigma^2' title='\\sigma^2' class='latex' \/> is the sum of the squares of the principal curvatures.<!--more--><\/p>\n<p>The proof is the same as for the case of simple density, <img src='https:\/\/s0.wp.com\/latex.php?latex=f%3Dg&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='f=g' title='f=g' class='latex' \/>; see\u00a0<a href=\"http:\/\/arxiv.org\/abs\/math\/0602135\">Rosales\u00a0et al. [R]<\/a>, Section 3.<\/p>\n<p>It follows that balls about the origin are stable for a radial density if and only if:<\/p>\n<p>a simple density (<em>f(r)<\/em>=<em>g(r)<\/em>) is log convex;<\/p>\n<p>a perimeter density <em>g<\/em>(<em>r<\/em>)\u00a0(<em>f<\/em>=1) satisfies (<em>n<\/em>\/<em>r<\/em>)<em>g<\/em>&#8216; + <em>g<\/em>&#8221; \u2265 0;<\/p>\n<p>a volume density <em>f(r)<\/em> (<em>g<\/em>=1) is nonincreasing;<\/p>\n<p>a conformal change of metric <em>f(r)<\/em> = \u03bb<sup>n+1<\/sup>, <em>g(r)<\/em> = \u03bb<sup>n<\/sup> satisfies \u2212\u03bb\u03bb&#8217;\/r + \u03bb\u03bb&#8221; \u2212 2\u03bb&#8217;<sup>2<\/sup> \u2265 0.<\/p>\n<p>The conformal case can be checked by using the standard second variation formula for a Riemannian metric:<\/p>\n<p><img src='https:\/\/s0.wp.com\/latex.php?latex=%5Cdelta+%5E2P%3D%5Cint+%7C%5Cnabla+u%7C%5E2-%7C%5Csigma%7C%5E2u%5E2-Ric%28u%2Cu%29&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='\\delta ^2P=\\int |\\nabla u|^2-|\\sigma|^2u^2-Ric(u,u)' title='\\delta ^2P=\\int |\\nabla u|^2-|\\sigma|^2u^2-Ric(u,u)' class='latex' \/>.<\/p>\n<p>In the planar case (n=1), the curvature \u03ba is given by<\/p>\n<p><img src='https:\/\/s0.wp.com\/latex.php?latex=%5Ckappa+%3D+-%5Cfrac%7BdP%7D%7BdA%7D+%3D+-%5Cfrac%7B%5Clambda+%2B+r%5Clambda%5E%5Cprime%7D%7B%5Clambda+%5E2r%7D&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='\\kappa = -\\frac{dP}{dA} = -\\frac{\\lambda + r\\lambda^\\prime}{\\lambda ^2r}' title='\\kappa = -\\frac{dP}{dA} = -\\frac{\\lambda + r\\lambda^\\prime}{\\lambda ^2r}' class='latex' \/><\/p>\n<p>and the Gauss curvature is given by the <a href=\"http:\/\/en.wikipedia.org\/wiki\/Ricci_curvature\">formula<\/a><\/p>\n<p><img src='https:\/\/s0.wp.com\/latex.php?latex=G%3D-%5Clambda%5E%7B-2%7D%5CDelta+log%5Clambda&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='G=-\\lambda^{-2}\\Delta log\\lambda' title='G=-\\lambda^{-2}\\Delta log\\lambda' class='latex' \/><br \/>\n= \u03bb<sup>-4<\/sup>(\u03bb\u03bb&#8217;\/r + \u03bb\u03bb&#8221; \u22122\u03bb&#8217;<sup>2<\/sup>).<\/p>\n<p>Therefore |\u03c3|<sup>2<\/sup> = <em>n<\/em>\u03ba<sup>2<\/sup> and Ric(<em>u,u<\/em>) = <em>nGu<\/em><sup>2<\/sup>. Also, the gradient of <em>u<\/em> is larger by a factor of \u03bb in the background metric. Now the second variation formula yields the same condition for stability of spheres. (For the lowest eigenvalue, the gradient term and the 1\/<em>r<\/em><sup>2<\/sup> term cancel out.)<\/p>\n<p><em>Remark<\/em> February 14, 2011. In the plane with density <em>r<\/em><sup><em>p<\/em><\/sup> (<em>p<\/em>&gt;0), isoperimetric regions are balls with the origin on the boundary [<a href=\"http:\/\/nyjm.albany.edu\/j\/2010\/16-4v.pdf\">Dahlberg et al.<\/a>]. For unequal radial volume and perimeter densities, such balls are not even in equilibrium.<\/p>\n<p><em>Corollary\u00a0<\/em>June 25, 2016. In a cone over a round <em>n<\/em>-sphere of mean curvature 1\/<em>a<\/em>, volume density\u00a0<em>f(r)<\/em>, and perimeter density\u00a0<em>g(r)<\/em>,\u00a0\u00a0a geodesic sphere about the origin of radius\u00a0<em>r <\/em>has nonnegative second variation\u00a0if and only if<\/p>\n<p>(1) \u00a0<img src='https:\/\/s0.wp.com\/latex.php?latex=n%28a%5E2+-1%29+%2B+nr%28f%2Fg%29%28g%2Ff%29%27+%2B+r%5E2+g%22%2Fg+-+r%5E2+%281%2Ffg%29%28f%27g%27%29+%5Cge+0.+&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='n(a^2 -1) + nr(f\/g)(g\/f)&#039; + r^2 g&quot;\/g - r^2 (1\/fg)(f&#039;g&#039;) \\ge 0. ' title='n(a^2 -1) + nr(f\/g)(g\/f)&#039; + r^2 g&quot;\/g - r^2 (1\/fg)(f&#039;g&#039;) \\ge 0. ' class='latex' \/><\/p>\n<p>For a single density <img src='https:\/\/s0.wp.com\/latex.php?latex=g+%3D+f&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='g = f' title='g = f' class='latex' \/>, (1) becomes<\/p>\n<p>(2)\u00a0<img src='https:\/\/s0.wp.com\/latex.php?latex=n%28a%5E2+-+1%29%2Br%5E2+%28log+f%29%22+%5Cge+0.+&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='n(a^2 - 1)+r^2 (log f)&quot; \\ge 0. ' title='n(a^2 - 1)+r^2 (log f)&quot; \\ge 0. ' class='latex' \/><\/p>\n<p>which for <img src='https:\/\/s0.wp.com\/latex.php?latex=a%3D1&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='a=1' title='a=1' class='latex' \/>\u00a0reduces to the standard log convexity condition.\u00a0For volume density <img src='https:\/\/s0.wp.com\/latex.php?latex=f+%3D+r%5Em&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='f = r^m' title='f = r^m' class='latex' \/> and perimeter density <img src='https:\/\/s0.wp.com\/latex.php?latex=g+%3D+r%5Ek&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='g = r^k' title='g = r^k' class='latex' \/>, (1) and (2) become<\/p>\n<p>(1a) <img src='https:\/\/s0.wp.com\/latex.php?latex=+m+%5Cle+k-1%2Bna%5E2+%2F%28k%2Bn%29%2C&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt=' m \\le k-1+na^2 \/(k+n),' title=' m \\le k-1+na^2 \/(k+n),' class='latex' \/><\/p>\n<p>in agreement with Alvino et al. [A, (5.53)] when <img src='https:\/\/s0.wp.com\/latex.php?latex=a%3D1&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='a=1' title='a=1' class='latex' \/>;<\/p>\n<p>(2a) <img src='https:\/\/s0.wp.com\/latex.php?latex=n%28a%5E2-1%29+%5Cge+m.&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='n(a^2-1) \\ge m.' title='n(a^2-1) \\ge m.' class='latex' \/><\/p>\n<p>In the conformal case <img src='https:\/\/s0.wp.com\/latex.php?latex=m%3Dk%28n%2B1%29%2Fn&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='m=k(n+1)\/n' title='m=k(n+1)\/n' class='latex' \/>, (1a) becomes <img src='https:\/\/s0.wp.com\/latex.php?latex=k+%5Cle+n%28a-1%29&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='k \\le n(a-1)' title='k \\le n(a-1)' class='latex' \/>. For <img src='https:\/\/s0.wp.com\/latex.php?latex=n%3D1&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='n=1' title='n=1' class='latex' \/>, the cone is equivalent to the <img src='https:\/\/s0.wp.com\/latex.php?latex=%5Ctheta+%3D+%5Cpi+%2Fa&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='\\theta = \\pi \/a' title='\\theta = \\pi \/a' class='latex' \/> sector, and (2a) becomes\u00a0<img src='https:\/\/s0.wp.com\/latex.php?latex=%5Ctheta+%5Cle+%5Cpi+%2F%5Csqrt%7Bm%2B1%7D&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='\\theta \\le \\pi \/\\sqrt{m+1}' title='\\theta \\le \\pi \/\\sqrt{m+1}' class='latex' \/>\u00a0of Diaz et al. [D, Prop. 4.16].<\/p>\n<p><em>Proof.\u00a0<\/em>The lowest mode on <img src='https:\/\/s0.wp.com\/latex.php?latex=S%5En&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='S^n' title='S^n' class='latex' \/> comes from translation in <img src='https:\/\/s0.wp.com\/latex.php?latex=R%5E%7Bn%2B1%7D&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='R^{n+1}' title='R^{n+1}' class='latex' \/>, i.e. <img src='https:\/\/s0.wp.com\/latex.php?latex=u+%3D+x_1&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='u = x_1' title='u = x_1' class='latex' \/>, with <img src='https:\/\/s0.wp.com\/latex.php?latex=%7C%5Cnabla+u%7C%5E2&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='|\\nabla u|^2' title='|\\nabla u|^2' class='latex' \/> on the average <em>n<\/em>\u00a0times as large as <img src='https:\/\/s0.wp.com\/latex.php?latex=u%5E2&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='u^2' title='u^2' class='latex' \/> when <img src='https:\/\/s0.wp.com\/latex.php?latex=a%3D1&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='a=1' title='a=1' class='latex' \/>\u00a0and <img src='https:\/\/s0.wp.com\/latex.php?latex=n+a%5E2&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='n a^2' title='n a^2' class='latex' \/>\u00a0times as large in general. <img src='https:\/\/s0.wp.com\/latex.php?latex=%7C%5Csigma%7C%5E2&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='|\\sigma|^2' title='|\\sigma|^2' class='latex' \/> is <img src='https:\/\/s0.wp.com\/latex.php?latex=n%2Fr%5E2&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='n\/r^2' title='n\/r^2' class='latex' \/>.<\/p>\n<p><strong>Theorem<\/strong> [E, 6.3].\u00a0Let <em>S<\/em> be a smooth Riemannian disk, sphere, or annulus of revolution with metric <img src='https:\/\/s0.wp.com\/latex.php?latex=ds%5E2+%3D+dr%5E2+%2B+%5Cphi%28r%29%5E2+d%5Ctheta%5E2&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='ds^2 = dr^2 + \\phi(r)^2 d\\theta^2' title='ds^2 = dr^2 + \\phi(r)^2 d\\theta^2' class='latex' \/> and density <em>f(r)<\/em>. Then the circle of revolution at distance <em>r<\/em> has nonnegative second variation\u00a0if and only if<\/p>\n<p><img src='https:\/\/s0.wp.com\/latex.php?latex=Q%28r%29+%3D+%5Cphi%27%28r%29%5E2+-+%5Cphi%28r%29%5Cphi%22%28r%29+-+%5Cphi%28r%29%5E2+%28log+f%29%22+%5Cle+1&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='Q(r) = \\phi&#039;(r)^2 - \\phi(r)\\phi&quot;(r) - \\phi(r)^2 (log f)&quot; \\le 1' title='Q(r) = \\phi&#039;(r)^2 - \\phi(r)\\phi&quot;(r) - \\phi(r)^2 (log f)&quot; \\le 1' class='latex' \/>.<\/p>\n<p><strong>Proposition<\/strong> (May 7, 2019). <em>Similarly, in an (n+1)D ambient of revolution, the sphere at distance r has nonnegative second variation if and only if<\/em><\/p>\n<p><img src='https:\/\/s0.wp.com\/latex.php?latex=Q%28r%29+%3D+%5Cphi%27%28r%29%5E2+-+%5Cphi%28r%29%5Cphi%22%28r%29+-+%5Cphi%28r%29%5E2+%28log+f%29%22%2Fn+%5Cle+1&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='Q(r) = \\phi&#039;(r)^2 - \\phi(r)\\phi&quot;(r) - \\phi(r)^2 (log f)&quot;\/n \\le 1' title='Q(r) = \\phi&#039;(r)^2 - \\phi(r)\\phi&quot;(r) - \\phi(r)^2 (log f)&quot;\/n \\le 1' class='latex' \/>.<\/p>\n<p><em>Proof<\/em>. The Gauss curvature of a radial section is <img src='https:\/\/s0.wp.com\/latex.php?latex=-%5Cphi%22%2F%5Cphi&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='-\\phi&quot;\/\\phi' title='-\\phi&quot;\/\\phi' class='latex' \/> [E, 6.3], so the outward Ricci curvature is <img src='https:\/\/s0.wp.com\/latex.php?latex=-n%5Cphi%22%2Fh&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='-n\\phi&quot;\/h' title='-n\\phi&quot;\/h' class='latex' \/>. Each principal curvature equals the mean curvature<\/p>\n<p><img src='https:\/\/s0.wp.com\/latex.php?latex=dA%2FndV+%3D+%28log+A%29%27%2Fn+%3D+log%28%5Cphi%5En%29%27%2Fn+%3D+%5Cphi%27%2F%5Cphi%2C&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='dA\/ndV = (log A)&#039;\/n = log(\\phi^n)&#039;\/n = \\phi&#039;\/\\phi,' title='dA\/ndV = (log A)&#039;\/n = log(\\phi^n)&#039;\/n = \\phi&#039;\/\\phi,' class='latex' \/><\/p>\n<p>so the second fundamental form <img src='https:\/\/s0.wp.com\/latex.php?latex=II&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='II' title='II' class='latex' \/> satisfies <img src='https:\/\/s0.wp.com\/latex.php?latex=II%5E2+%3D+nh%27%5E2%2Fh%5E2&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='II^2 = nh&#039;^2\/h^2' title='II^2 = nh&#039;^2\/h^2' class='latex' \/>. By Rosales et al. [R, Rmk. 3.7], the second variation of a sphere S about the origin for a normal (i.e. radial) variation <em>u<\/em> that preserves volume to first order (\u222b<em>u<\/em> = 0) satisfies<\/p>\n<p><img src='https:\/\/s0.wp.com\/latex.php?latex=%5Cdelta%5E2%28u%29+%3D+%5Cint+%7C%5Cnabla+u%7C%5E2+-+u%5E2%28Ric%28N%2CN%29+-+%28log+f%29%22+%2B+II%5E2%29&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='\\delta^2(u) = \\int |\\nabla u|^2 - u^2(Ric(N,N) - (log f)&quot; + II^2)' title='\\delta^2(u) = \\int |\\nabla u|^2 - u^2(Ric(N,N) - (log f)&quot; + II^2)' class='latex' \/>.<\/p>\n<p>By the generalized Wirtinger Inequality [O, (3.1)] (Poincar\u00e9 Inequality), the second variation is nonnegative if and only if<\/p>\n<p><img src='https:\/\/s0.wp.com\/latex.php?latex=n%2Fh%5E2+%5Cle+Ric%28N%2CN%29+-+%28log+f%29%22+%2B+II%5E2+%3D+-n%5Cphi%22%2F%5Cphi+-+%28log+f%29%22+%2B+n%5Cphi%27%5E2%2F%5Cphi%5E2&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='n\/h^2 \\le Ric(N,N) - (log f)&quot; + II^2 = -n\\phi&quot;\/\\phi - (log f)&quot; + n\\phi&#039;^2\/\\phi^2' title='n\/h^2 \\le Ric(N,N) - (log f)&quot; + II^2 = -n\\phi&quot;\/\\phi - (log f)&quot; + n\\phi&#039;^2\/\\phi^2' class='latex' \/>,<\/p>\n<p>i.e.,<\/p>\n<p><img src='https:\/\/s0.wp.com\/latex.php?latex=1+%5Cge+%5Cphi%27%5E2+-+%5Cphi%5Cphi%22+-+%5Cphi%5E2%28log+f%29%22%2Fn&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='1 \\ge \\phi&#039;^2 - \\phi\\phi&quot; - \\phi^2(log f)&quot;\/n' title='1 \\ge \\phi&#039;^2 - \\phi\\phi&quot; - \\phi^2(log f)&quot;\/n' class='latex' \/>. QED.<\/p>\n<p><strong>Corollary.<\/strong> <em>In an (n+1)D ambient of revolution, for decreasing ambient curvature and convex density, spheres about the origin are stable for fixed volume, while for increasing curvature and concave density, they are unstable. <\/em><\/p>\n<p><em>Proof.<\/em>\u00a0For infinitesimal spheres, the ambient is Euclidean and <em>Q<\/em> = 0.\u00a0For constant density <em>f<\/em>, <em>Q<\/em> and the Gauss curvature <em>K<\/em> of a radial section satisfy <img src='https:\/\/s0.wp.com\/latex.php?latex=Q%27+%3D+%5Cphi%5E2+K%27&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='Q&#039; = \\phi^2 K&#039;' title='Q&#039; = \\phi^2 K&#039;' class='latex' \/> (trivial one-line computation as in [R, Lemma 1.6]), [E, Sect. 6], [M]). In particular, <em>Q<\/em> and <em>K<\/em> increase or decrease together. The convexity or concavity of the density only causes <em>Q<\/em> to be still smaller or larger.<\/p>\n<p>[A] A. Alvino, F. Brock, F. Chiacchio, A. Mercaldo, M.R. Posteraro, Some isoperimetric inequalities on <img src='https:\/\/s0.wp.com\/latex.php?latex=R%5EN&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='R^N' title='R^N' class='latex' \/> with respect to weights <img src='https:\/\/s0.wp.com\/latex.php?latex=%7Cx%7C%5E%5Calpha&#038;bg=ffffff&#038;fg=000000&#038;s=0' alt='|x|^\\alpha' title='|x|^\\alpha' class='latex' \/>, arXiv.org (2016). .<\/p>\n<p>[D] Alexander D\u00edaz, Nate Harman, Sean Howe, David Thompson. Isoperimetric problems in sectors with density, Adv. Geom. 12 (2012), 589\u2013619; arXiv.org (2010).<\/p>\n<p>[E]\u00a0Max Engelstein, Anthony Marcuccio, Quinn Maurmann, and Taryn Pritchard, Isoperimetric problems on the sphere and on surfaces with density, New York J. Math. 15 (2009), 97\u2013123.<\/p>\n<p>[M] Frank Morgan,\u00a0 Isoperimetric symmetry breaking: a counterexample to a generalized form of the log-convex density conjecture, Anal. Geom. Metr. Spaces 4 (2016), 314-316.<\/p>\n<p>[O] Robert Osserman, <a href=\"https:\/\/projecteuclid.org\/download\/pdf_1\/euclid.bams\/1183541466\">The Isoperimetric Inequality<\/a>, Bull. Amer. Math. Soc. 84 (1978), 1182-1238.<\/p>\n<p>[R]\u00a0C\u00e9sar Rosales, Antonio Ca\u00f1ete, Vincent Bayle, and Frank Morgan. On the isoperimetric problem in Euclidean space with density. Calc. Var. PDE 31 (2008), 27-46; <a href=\"https:\/\/arxiv.org\/abs\/math\/0602135\">arXiv.org<\/a> (2006).<\/p>\n","protected":false},"excerpt":{"rendered":"<p>For Rn+1 with volume density f and perimeter density g, for a normal variation u of a surface with classical mean curvature H, the first variation of volume and perimeter are given by: , . For a volume-preserving variation, the second variation of perimeter is given by: , where is the second fundamental form, so [&hellip;]<\/p>\n","protected":false},"author":269,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[14043],"tags":[],"class_list":["post-190","post","type-post","status-publish","format-standard","hentry","category-math"],"acf":[],"_links":{"self":[{"href":"https:\/\/sites.williams.edu\/Morgan\/wp-json\/wp\/v2\/posts\/190","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/sites.williams.edu\/Morgan\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/sites.williams.edu\/Morgan\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/sites.williams.edu\/Morgan\/wp-json\/wp\/v2\/users\/269"}],"replies":[{"embeddable":true,"href":"https:\/\/sites.williams.edu\/Morgan\/wp-json\/wp\/v2\/comments?post=190"}],"version-history":[{"count":121,"href":"https:\/\/sites.williams.edu\/Morgan\/wp-json\/wp\/v2\/posts\/190\/revisions"}],"predecessor-version":[{"id":2887,"href":"https:\/\/sites.williams.edu\/Morgan\/wp-json\/wp\/v2\/posts\/190\/revisions\/2887"}],"wp:attachment":[{"href":"https:\/\/sites.williams.edu\/Morgan\/wp-json\/wp\/v2\/media?parent=190"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/sites.williams.edu\/Morgan\/wp-json\/wp\/v2\/categories?post=190"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/sites.williams.edu\/Morgan\/wp-json\/wp\/v2\/tags?post=190"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}