The first theoretical model of magnetic reconnection was that of Sweet-Parker. In this theory the breaking of the frozen-in condition that facilitates the topological change in the magnetic field required for reconnection occurs in a macroscopic current layer called the dissipation region. The high elongation of this current layer (the width being controlled by the small plasma resistivity) combined with inflow-outflow continuity yields rates of flux annihilation that scale like sqrt(eta). For classical values of the resistivity this rate of reconnection is far smaller than that inferred from observations in both astrophysical and laboratory systems. The Petschek theory of reconnection, in which the dissipation region is spatially localized, yields high rates of reconnection. Computer simulations demonstrated that the Sweet-Parker structure of the dissipation region predominates in systems with uniform resistivity. The predominance of such elongated layers was shown to be a consequence of the singular behavior of post-reconnection equilibria.

Two models for boosting the Sweet-Parker rates of reconnection have found support from computation and observational data: the kinetic reconnection model and the anomalous resistivity model. In the kinetic reconnection model fast reconnection is facilitated by the coupling to non-MHD, dispersive waves at the small spatial scales that characterize the dissipation region. These waves shorten the elongated Sweet-Parker current layer into a Petschek-like open X-line configuration with its characteristic fast rate of reconnection. Both dedicated laboratory reconnection experiments as well as satellite observations have provided evidence for the coupling to these dispersive waves.

The anomalous resistivity hypothesis dates from the early 1960's. The suggestion was that the large relative streaming between electron and ions, a consequence of the strong current present in the dissipation region, would drive turbulence that could cause scattering and thereby enhance the drag between electrons and ions, e.g., an enhanced and therefore ``anomalous'' resistivity. Anomalous resistivity has been widely invoked over the past four decades to justify the resistive MHD description of magnetic reconnection in nature. However, the theoretical underpining for the idea has remained weak. There is observational evidence for the presence of turbulence in laboratory reconnection experiments, measurements of sawteeth in tokamaks and in the Earth's magnetosphere. The strongest evidence in favor of the anomalous resistivity hypothesis has come from recent measurement of magnetic fluctuations on the MRX experiment and the widespread identification of ``electron holes'' in boundary layers of the Earth's magnetosphere where reconnection is expected to be active. 3-D particle simulations of magnetic reconnection are now for the first time able to address the anomalous resistivity problem and have demonstrated that electron-hole formation and associated anomalous resistivity are a natural consequence of reconnection in high temperature plasma.

While evidence is now substantial that anomalous resistivity may be active during reconnection, we are far from having a comprehensive understanding of the detailed mechanisms sufficient to model the effect in nearly collisionless systems such as tokamaks where reconnection electric fields can exceed the Dreicer runaway field during sawteeth and disruptions. To address this question we propose a program with three linked elements. The first is a comprehensive set of laboratory experiments to be carried out on the LAPD device at UCLA to explore the development of electron-ion streaming instabilities and associated heating and anomalous resistivity. The second element is turbulence measurements during reconnection on VTF at MIT. In the third element, parallel 3-D particle simulations with the p3d code will be carried out and the results compared with the experimental measurements. The relative roles of the ion acoustic instability, which can be driven unstable when T_e >> T_i, the Buneman instability and off-angle lower-hybrid modes will be explored. Questions such as the following will be addressed. Can the ion acoustic instability halt the runaway of electrons or does the system evolve into the stronger Buneman unstable configuration and develop electron holes as a nonlinear state? Since the instabilities causing the anomalous resistivity have rapid timescales compared to the reconnection time a multiscale simplification is desirable. For example, theoretical scaling laws for the turbulence and associated resistivity will be developed for implementation in large scale reconnection models. A more ambitious approach involving coupling fast time-scale codes calculating the average effect of the turbulence to slow time-scale codes calculating the reconnection itself is also being developed.