DescriptionUsing an array of multi-wavelength data, we examine a variety of astrophysical problems with gravitational lensing. First, we seek to understand the mass distribution of an early-type galaxy with an analysis of the lens Q0957+561. We dissect the lens galaxy into luminous and dark components, and model the environment using results from weak lensing. Combining constraints from newly-discovered lensed images and stellar population models we find the lens has a density profile which is shallower than isothermal, unlike those of typical early-type galaxies. Finally, using the measured time delay between the quasar images we find the Hubble constant to be H_0=79.3^{+6.7}_{-8.5} km s^{-1} Mpc^{-1}. One intriguing application of lensing is to exploit the lens magnification boost to study high-redshift objects in greater detail than otherwise possible. Here, we analyze the mid-infrared properties of two lensed z~2 star-forming galaxies, SDSS J120602.09+514229.5 and SDSS J090122.37+181432.3, using Spitzer/IRS spectra to study their rest-frame ~ 5-12 micron emission. Both systems exhibit strong polycyclic aromatic hydrocarbon (PAH) features in the spectra, indicating strong star formation and the absence of significant AGN activity. For SDSS J090122.37+181432.3, this detection belies that inferred from optical measurements, indicating mid-IR spectroscopy provides key information needed to understand the properties of high-redshift star-forming galaxies. While lensing provides measurements of the macroscopic properties of lens systems, it can also shed light on small-scale structure of galaxies. To identify and understand lens substructure, we examine the multi-wavelength properties of flux ratios for six lenses. Variations of the flux ratios with wavelength can be used to study the lensed quasars and the small-scale mass distribution of lens galaxies. We detect strong multi-wavelength variations in the lenses HE 0435-1223 and SDSS 0806+2006. For HE 0435-1223, we study its substructure with a series of lens models which add clumps of mass near the lensed images. We detect the presence of a clump near image A, with a mass of log(M_A( 0.00092.