The first edition of "e;The Stability of Matter: From Atoms to Stars"e; was sold out after a time unusually short for a selecta collection and we thought it ap- propriate not just to make a reprinting but to include eight new contributionso They demonstrate that this field is still lively and keeps revealing unexpected featureso Of course, we restricted ourselves to developments in which Elliott Lieb participated and thus the heroic struggle in Thomas-Fermi theory where 7 3 5 3 the accuracy has been pushed from Z 1 to Z 1 is not includedo A rich landscape opened up after Jakob Yngvason's observation that atoms in magnetic fields also are described in suitable limits by a Thomas-Fermi-type theoryo Together with Elliott Lieb and Jan Philip Solovej it was eventually worked out that one has to distinguish 5 regionso If one takes as a dimensionless measure of the magnetic field strength B the ratio Larmor radius/Bohr radius one can compare it with N "e;' Z and for each of the domains 4 3 (i) B N 1 , 4 3 (ii) B "e;' N 1 , 4 3 3 (iii) N 1 B N , 3 (iv) B "e;' N , 3 (v) B N a different version ofmagnetic Thomas-Fermi theory becomes exact in the limit N --+ ooo In two dimensions and a confining potential ("e;quantum dots"e;) the situation is somewhat simpler, one has to distinguish only (i) B N, (ii) B "e;'N

The first edition of "e;The Stability of Matter: From Atoms to Stars"e; was sold out after a time unusually short for a selecta collection and we thought it ap- propriate not just to make a reprinting but to include eight new contributionso They demonstrate that this field is still lively and keeps revealing unexpected featureso Of course, we restricted ourselves to developments in which Elliott Lieb participated and thus the heroic struggle in Thomas-Fermi theory where 7 3 5 3 the accuracy has been pushed from Z 1 to Z 1 is not includedo A rich landscape opened up after Jakob Yngvason's observation that atoms in magnetic fields also are described in suitable limits by a Thomas-Fermi-type theoryo Together with Elliott Lieb and Jan Philip Solovej it was eventually worked out that one has to distinguish 5 regionso If one takes as a dimensionless measure of the magnetic field strength B the ratio Larmor radius/Bohr radius one can compare it with N "e;' Z and for each of the domains 4 3 (i) B N 1 , 4 3 (ii) B "e;' N 1 , 4 3 3 (iii) N 1 B N , 3 (iv) B "e;' N , 3 (v) B N a different version ofmagnetic Thomas-Fermi theory becomes exact in the limit N --+ ooo In two dimensions and a confining potential ("e;quantum dots"e;) the situation is somewhat simpler, one has to distinguish only (i) B N, (ii) B "e;'N,