Laser-induced breakdown spectroscopy (LIBS) is an elemental analysis technique that is based on the measurement of atomic emissions generated on a sample surface by a laser-induced microplasma. Although often recognized in the literature as a well-established analytical technique, LIBS remains untested relative to the quantitative analysis of elements in chemically complex matrices, such as soils. The objective of this study was to evaluate the capabilities of LIBS relative to the elemental characterization of surface soils. Approximately 65 surface soil samples from the Pond Creek watershed in east Tennessee were collected and subjected to total dissolution and elemental analysis by inductively coupled argon plasma-optical emission spectroscopy (ICP–OES). The samples were analyzed by LIBS using a Nd:YAG laser at 532 nm, with a beam energy of 25 mJ per pulse, a pulse width of 5 ns, and a repetition rate of 10 Hz. The wavelength range for the LIBS spectra collection was 200 to 600 nm, with a resolution of 0.03 nm. Elemental spectral lines were identified through the analysis of analytical reagent-grade chemicals and the NIST and Kurucz spectral databases. The elements that dominated the LIBS spectra were Al, Ca, Fe, and Mg. In addition, emission lines for Ti, Ba, Na, Cu, and Mn were isolated. The emission lines of Cr, Ni, and Zn, which were >100 mg kg−1 in numerous soil samples, were not detected. Further, spectral emission lines for P and K are >600 nm, eliminating them from LIBS analysis. The integrated peak areas of interference-free elemental emission lines were determined, then normalized to the area of the 288.16 nm Si(I) emission (internal standard) to reduce the variability between replicate analyses. The normalized spectral areas, coupled with linear regression (standard curves for single wavelength response) and multivariate techniques (chemometrics and multiple wavelengths), were used to predict ICP–OES elemental data. In general, the quantitative capabilities of LIBS proved disappointing. Detection and quantitation were generally restricted to those elements with concentrations > 0.5 g kg−1 The correlation between LIBS response and elemental content was poor (r < 0.98). Further, the relative errors of prediction for the LIBS-detected elements were less than acceptable for an analytical technique (<20%), ranging from ∼20 to ∼40% using linear regression analysis, and from 18 to 48% using partial least squares analysis. Based on these findings, the analytical capability of the LIBS method for soil metals analysis should be considered questionable.