The ratio of the number of observed taxa to that expected to occur in the absence of human‐caused stress (O/E) is an intuitive and ecologically meaningful measure of biological integrity. We examined how O/E ratios derived from stream invertebrate data varied among 234 unimpaired reference sites and 254 test sites potentially impaired by past logging. Data were collected from streams in three montane ecoregions in California. Two sets of River Invertebrate Prediction and Classification System (RIVPACS) predictive models were built: one set of models was based on near‐species taxonomic resolution; the other was based on family identifications. Two models were built for each level of taxonomic resolution: one calculated O and E based on all taxa with probabilities of capture (Pc) > 0; the other calculated O and E based on only those taxa with Pc ≥ 0.5. Evaluations of the performance of each model were based on three criteria: (1) how well models predicted the taxa found at unimpaired sites, (2) the degree to which O/E values differed among unimpaired reference sites and potentially impaired test sites, and (3) the degree to which test site O/E values were correlated with independent measures of watershed alteration. Predictions of species models were more accurate than those of family models, and predictions of the Pc ≥ 0.5 species model were more robust than predictions of the Pc ≥ 0 model. O/E values derived from both species models were related to land use variables, but only assessments based on the Pc ≥ 0.5 model were insensitive to naturally occurring differences among streams, ecoregions, and years.

A multivariate search for pollination syndromes among penstemons. -Oikos 104: 345-361.The seeming ubiquity of spatio-temporal variation in pollination regime suggests that flowers ought to be adapted to a wide range of pollinators, yet many comparative biologists perceive that in groups with complex flowers there is considerable specialization onto pollination syndromes. Statistical documentation of such syndromes has been presented for very few groups of flowers. Accordingly, we measured, for 49 species of Penstemon and close relatives, both the morphology of the flowers and visitation by pollinators. We describe the mechanics of pollination for representative species. Ordinations show a distinct difference between hummingbird-pollinated species and hymenopteran-pollinated species. Flower color is particularly good at separating hummingbird-from hymenopteran-flowers. Other characters are also correlated with this dichotomy. Within the hymenopteran-pollinated species, there are additional relationships between floral morphology and the size of the principal pollinators. Flowers frequented by large bees, such as Xylocopa, have large open vestibules and relatively short floral tubes. Flowers frequented by smaller bees, such as Osmia, have long narrow floral tubes. Unlike nectar-collecting bees, pollencollecting bees tend to be attracted to flowers of the hummingbird syndrome. The overarching pattern was that syndrome characterizations were successful at predicting pollination by hummingbirds versus Hymenoptera, two types of animals that are profoundly different, but less successful at predicting visitation by one kind of bee versus another.

No abstract

The ratio of the number of observed taxa to that expected to occur in the absence of human-caused stress (O/E) is an intuitive and ecologically meaningful measure of biological integrity. We examined how O/E ratios derived from stream invertebrate data varied among 234 unimpaired reference sites and 254 test sites potentially impaired by past logging. Data were collected from streams in three montane ecoregions in California. Two sets of River Invertebrate Prediction and Classification System (RIVPACS) predictive models were built: one set of models was based on near-species taxonomic resolution; the other was based on family identifications. Two models were built for each level of taxonomic resolution: one calculated O and E based on all taxa with probabilities of capture (P c) Ͼ 0; the other calculated O and E based on only those taxa with P c Ն 0.5. Evaluations of the performance of each model were based on three criteria: (1) how well models predicted the taxa found at unimpaired sites, (2) the degree to which O/E values differed among unimpaired reference sites and potentially impaired test sites, and (3) the degree to which test site O/E values were correlated with independent measures of watershed alteration. Predictions of species models were more accurate than those of family models, and predictions of the P c Ն 0.5 species model were more robust than predictions of the P c Ն 0 model. O/E values derived from both species models were related to land use variables, but only assessments based on the P c Ն 0.5 model were insensitive to naturally occurring differences among streams, ecoregions, and years.