A series of vertical profiles of ozone from Narragansett, RI reveal some expected and some surprising features. Electo Chemical Cell ozonesondes were used and soundings were made at 1800 UTC over a 45 day period in the summer of 2004 as part of a comprehensive program. Sondes equipped with GPS receivers were used on several occasions, and the tracking data and the location of sondes reported found after their descent were used to assess a balloon trajectory calculation scheme. Wind fields and other needed data from the Rapid Update Cycle model have been used for the trajectory calculations. The profiles extend above 30 km, but the analysis here is confined to p > 100 hPa and the emphasis is on tropospheric features. At least through the first month of the campaign, relatively few instances of conditions favoring photochemical production of ozone in the polluted boundary layer occurred in southern New England, and near-surface ozone mixing ratios were below average. The variability in the ozone profile was notably high in the middle and upper troposphere, and the relationship of high and low ozone abundances to dynamical features will be discussed. An event in which titration of ozone by NO or other compounds may explain exceptionally low mixing ratios near the surface in the mid-afternoon will be discussed. In the larger program profiles were made by other groups from a number of sites, and extensive additional measurements were made from surface sites and from instrumented aircraft, and the data combined with satellite observations and chemical transport model simulations, but the emphasis here is on the data from Narragansett. Sponsorship and scientific collaboration for this project from NOAA (Climate Monitoring and Diagnostics Laboratory and the New England Air Quality Study - Intercontinental Transport and Chemical Transformation program) and NASA (Intercontinental Chemical Transport Experiment - North America) are acknowledged.