Bose-Einstein condensates made by laser cooling and evaporative cooling of atoms are the coldest known substance and are beginning to find a wide range of applications from understanding fluids to precision measurements. Most BoseEinstein condensate to date are based 'one-electron atoms', i.e. the atomic structure is determined by a single electron outside a charged core. This generally leads to an atomic structure ideal for laser cooling but without any particularly narrow spectral lines that would be ideal for precision measurement or optical clocks. On the other hand, the two-electron atoms (such a calcium) offer narrow lines associated with transitions, where an electron spin flips, and are still relatively easy to laser cool. The narrow transition in atomic calcium is key for the present proposal. It will enable us to laser cool the atoms all the way to Bose-Einstein condensation - something that has not been possible to do with other atoms due to the re-absorption of the light scattered on broader lines. An essential part of this is to trap the atoms in the strong light field of a CO2 laser in order to prevent them from falling under gravity while they are slowly cooled on the narrow line. The direct laser cooling to condensation is radically different from the traditional approach, which relies on atomic collisions. It will therefore provide new insight into the formation of condensates. The CO2 laser offers a wide choice of geometry for the condensate. We can generate condensates in 1, 2 and 3 dimensional lattices and study the interaction of many independently created condensates when they are allowed to 'see' each other due to quantum mechanical tunnelling through the separating barriers. The ultimate vision for this work is to use the narrow atomic transition for precision measurements (e.g. an optical clock) in what is known as the Heisenberg limit. That requires the preparation through Bose-Einstein condensation of highly entangled multi-particle states, e.g. N atoms in a superposition of ground and excited states such that if one is found in the ground state then they are all in the ground state or vice versa. With states like this it will be possible to obtain a precision on a measurement, that scales as 1/N (the Heisenberg limit) rather than the 1/sqrt(N) associated with Poissonian statistics. This project will explore and develop a range of technologies for the future realisation of such measurements.