Summary An existing canine genomic bacterial artificial chromosome (BAC) library was expanded by adding 115 200 clones with insert lengths not yet represented or under‐represented. The final version of the library consists of 211 968 clones with an estimated average insert size of 110 000 base pairs. Clones were grown individually and glycerol permanents were arrayed in 2208 96‐well microtitre plates. DNA of each clone was prepared by microwave treatment and organized in a three‐dimensional DNA pooling system (92 superpools) allowing for rapid screening by polymerase chain reaction (PCR). Screening of the library for 111 microsatellite loci representing all canine chromosomes revealed a recovery rate of 93% and a fourfold genome coverage. In addition, 50 BAC clones containing microsatellites or genes related to ongoing projects in the laboratories of the authors and eight other research groups were successfully recovered. The present library is the first canine BAC library amenable to PCR screening and is an invaluable tool for cloning candidate loci and developing genetic markers in specific chromosomal regions. This library will also enhance mapping efforts in related canid species and is an important source for geneticists studying inherited diseases common to both dog and humans. Interested researchers can access the library following the instructions at http://www.dogmap.ch .

SummaryWe have constructed a canine bacterial artificial chromosome library amenable to PCR screening. The library consists of 96 768 clones and was initially screened with 112 microsatellites representing all canine chromosomes. For 87 primer sets (77%) one to seven positive superpools were identified. The library will be expanded by adding additional superpools in order to increase the genome coverage. Interested researchers can access the library following the rules published at http://www.dogmap.ch.

In the paper the problems connected with numerical modelling of the solidification and cooling processes in a system casting-mould are discussed. The task is treated as a boundary-initial problem in which the crystallization process (in micro scale) is taken into account, in other words the component describing a capacity of internal heat sources in adequate energy equation results from the analysis of crystallization process on a microscopic level. The mathematical model of crystallization process was constructed on a basis of the theory given by Kolmogorov. The number of nuclei and also their temporary dimensions (i.e. crystallization model) result from the local values of undercooling below the equilibrium temperature [1, 2, 3]. Such approach is widely known - but the numerical procedures worked out by the authors make possible to 'follow' the vicissitudes of successive grains generations and it is a new element in mathematical modelling of crystallization process. Numerical solution of the boundary-initial problem discussed has been obtained on the basis of the Boundary Element Method. It is well known that this method assures a relatively good approximation of geometrical and boundary conditions. In the case of solidification process modelling this advantage is very essential because of the great values of temperature gradients near the casting-mould contact surface.As an example the tasks concerning the aluminium plate solidifying in the typical sand moulds will be presented, at the same time it is possible (in a simple way) to widen the algorithm in the direction of 2D or 3D problems and more complex materials (e.g. binary alloys)(*)).

The boundary element method is applied to micro/macro modelling of solidification process proceeding in 2D iron casting domain. The heat transfer process is described by the Fourier equation with additional term determining the kinetics of crystallization (the so-called source function). The model bases on the laws determining the course of nucleation and nuclei growth, while the relation between linear growth rate and the liquid fraction in casting volume is given by the Johnson Mehl Avrami Kolmogorov formulas. In the paper the mathematical description of the process, the method of numerical solution and also the example of computations are presented. 1 Governing equations Solidification of pure metals and eutectic alloys A differential equation describing the course of solidification process (a micro/macro model) is the parabolic one with additional term qy(X, t) called a source function: cp . 0 = div[Agrad7(X, t)] + q^(X, t) (1) dt where c, p, X are the specific heat, mass density and thermal conductivity, 7, X, t are the temperature, spatial co-ordinate and time. Because the solidification process proceeds in a rather small interval of temperature one can assume the constant values of thermophysical parameters of material considered and finally the following energy equation should be taken into account 0 , , dt cp Transactions on Engineering Sciences vol 12, © 1996 WIT Press, www.witpress.com, ISSN 1743-3533