Compilations of surface winds from ship reports since 1854 show a number of long period variations, including a trend toward strengthening winds over the past three decades. Some investigators indicate that these variations are real changes in the climate system, while others suggest that they are artifacts of the evolution of measurement techniques. In an attempt to resolve this issue, we have examined individual ship reports from three regions with high data densities: South China Sea, North Pacific, and North Atlantic shipping lanes. We find that the apparent surface wind strengthening from the 1950s to the present is a consequence of the increasing use of anemometers in place of sea-state estimates. The specific causes are the operational use of an incorrect conversion from Beaufort force to wind speed, and the widespread assumption that the height of shipborne anemometers is 10 m, whereas the actual mean height is 19.3 m. Correcting the conversion scale and setting the height to be 20 m largely eliminates the trend. The adjustment of anemometer winds for stratification effects further increases the consistency between the measured and estimated winds. A formula for correcting the wind speeds available in standard averaged products is presented; in addition to wind speed, the ratio of estimated to measured observations and the average air-sea temperature difference are required. Even with the correction, the pre-1950 winds appear to be weaker than the post-1950 winds. The most likely explanation is the absence of universal standards for sea state and Beaufort force before 1946. Possible remedies are discussed, but the task is daunting. Though this cautionary tale does not rule out the existence of real trends in surface winds, it does impugn their delectability.