On the basis of hydrographic data collected by the Texas-Louisiana Shelf Circulation and Transport Processes Study (LATEX A) and on earlier cruises, we examined the energetic scales of spatial variability over the Texas-Louisiana continental shelf. Shelf-scale spatial reference fields were sought to represent the general distributions of circulation and water properties over the shelf at the time of the observations. Various methods were explored for determining such reference fields of potential temperature, salinity, and geopotential anomaly at the sea surface relative to 70 dB. Spatial reference fields obtained from mean May fields and from polynomials fitted to individual May cruise data were compared. On the basis of those comparisons, quadratics were selected to fit property distributions from individual cruises and so to yield reference fields. Smaller-scale anomaly fields were obtained by removing the reference fields from the observed distributions. Calculation of correlation versus separation distance based on these anomaly fields then allowed estimation of spatial scales of anomaly fields for cross-shelf and along-shelf transects. The zero-crossing scale and the Gaussian decay scale are shown to be essentially the same, and the zero-crossing scale is used. The principal results for the anomaly scales are (1) cross-shelf scales over the western shelf are shorter (order 15 km) than those in the eastern and central regions (order 20 km), (2) along-shelf spatial scales are of the order of 35 km, (3) there is no significant difference in cross-shelf scales at the surface, middepth, and bottom, and (4) along-shelf scales are essentially the same over the western and eastern regions of the shelf, over the midshelf (50-m isobath) and along the shelf break (200-m isobath), and at different depths along the 200-m isobath. The same spatial scales are found when using data with spatial resolution of 1--10 km cross shelf and 10--20 along shelf to obtain the anomaly fields, so the data resolution used is adequate to represent the scales. The variances of the observed (shelf-wide) salinity, temperature, and geopotential anomaly are greater cross shelf than along shelf. The variance of the cross-shelf anomaly fields is around 10% of the shelf-wide fields; that of the along-shelf anomaly fields is about 35% of that in the shelf-wide fields. The analysis of scales when grouped by season did not show persuasive evidence of seasonal variation. ¿ American Geophysical Union 1996