In 2001, the U.S. Geological Survey National Water Quality Assessment Program began a series of studies in the contiguous United States to examine the effects of urbanization on the chemical, physical, and biological characteristics of streams. Small streams in the Texas Blackland Prairie level III ecoregion in and near the Dallas-Fort Worth metropolitan area were the focus of one of the studies. The principal objectives of the study, based on data collected in 2003-04 from 28 subbasins of the Trinity River Basin, were to (1) define a gradient of urbanization for small Blackland Prairie streams in the Trinity River Basin on the basis of a range of urban intensity indexes (UIIs) calculated using land-use/land-cover, infrastructure, and socioeconomic characteristics; (2) assess the relation between this gradient of urbanization and the chemical, physical, and biological characteristics of these streams; and (3) evaluate the type of relation (that is, linear or nonlinear, and whether there was a threshold response) of the chemical, physical, and biological characteristics of these streams to the gradient of urbanization. Of 94 water-chemistry variables and one measure of potential toxicity from a bioassay, the concentrations of two pesticides (diazinon and sima-zine) and one measure of potential toxicity (P450RGS assay) from compounds sequestered in semipermeable membrane devices were significantly positively correlated with the UII. No threshold responses to the UII for diazinon and simazine concentrations were observed over the entire range of the UII scores. The linear correlation for diazinon with the UII was significant, but the linear correlation for simazine with the UII was not. No statistically significant relations between the UII and concentrations of suspended sediment, total nitrogen, total phosphorous, or any major ions were indicated. Eleven of 59 physical variables from streamflow were significantly correlated with the UII. Temperature was not significantly correlated with the UII, and none of the physical habitat measurements were significantly correlated with the UII. Seven physical variables categorized as streamflow flashiness metrics were significantly positively correlated with the UII, two of which showed a linear but not a threshold response to the UII. Four flow-duration metrics were significantly negatively correlated with the UII, of which two showed a linear response to the UII, one showed a threshold response, and one showed neither. None of the fish metrics were significantly correlated with the UII in the Blackland Prairie streams. Two qualitative multi-habitat benthic macroinvertebrate metrics, predator richness and percentage filterer-collector richness, were significantly correlated with the UII; predator richness was negatively correlated with the UII, and percentage filterer-collector richness was positively correlated with the UII. No threshold response to the UII was observed for either metric, but both showed a significant linear response to the UII. Three richest targeted habitat (RTH) benthic macroinvertebrate metrics, Margalef's richness, predator richness, and omnivore richness were significantly negatively correlated with the UII. Margalef's richness was the only RTH metric that indicated a threshold response to the UII. The majority of unique taxa collected in the periphytic algae samples were diatoms. Six RTH periphytic algae metrics were correlated with the UII and five of the six showed no notable threshold response to the UII; but all five showed significant linear responses to the UII. Only the metric OT_VL_DP, which indicates the presence of algae that are tolerant of low dissolved oxygen conditions, showed a threshold response to the UII. Six depositional target habitat periphytic algae metrics were correlated with the UII, five of which showed no threshold response to the UII; three of the five showed significant linear responses to the UII, one showed a borderline significant