During three years 231 homing pigeons were transported from northwest Ohio, U. S. A., to southern Germany, and 216 pigeons were transported from southern Germany to Ohio, a distance of 7000 to 8000 km. They were released in the foreign continent, their initial departure bearings were recorded, and recoveries were collected. For comparison, control releases were conducted within the home continent at distances of 200-300 km in the same as well as in the opposite direction. Both experienced and inexperienced birds zvere used. Normally, the circadian clocks were synchronized with the local time of the release area. The main results were as follows: . (1) At departure, the homeward orientation appeared to be as good (or poor) for the transatlantic releases as it was for the cisatlantic releases. (2) In America there was a general tendency for headings to deviate to the north of the straight line course towards home, while in Germany the birds deviated generally to the south. This was true for both American and German pigeons, indicating an influence of the release area on the deviations. (3) There were, mainly in inexperienced birds, considerable differences in initial orientation in different years. In most cases deflections of varying degrees occurred rather parallel in cisatlantic and transatlantic releases, though with opposite signs (i.e. left or right). (4) On the whole, and at either side of the ocean, the experienced birds were better oriented towards home than the first flight pigeons. (5) The deviations were, on the average, smaller from the rhumb line course towards the loft than they were from the great circle route. (6) There was only one criterion indicating an impairment in the foreign continent : The percentage of birds landing in the vicinity of the release point instead of leaving immediately was considerably higher in the foreign continent than in the home continent. (7) In most cases the distributions of the recoveries were in general accordance with those of the bearings at departure (with the exception of inexperienced American birds in Germany). The recoveries for both groups of experienced pigeons suggest fairly good homeward orientation in the foreign continent. (8) In Germany, inexperienced American pigeons were released with their circadian clocks set by three different light/dark schedules. Their bearings at departure deviated from each other in accordance with the phase-angle differences of their clocks as it would be expected on the basis of sun-compass orientation. (These results could also be explained in terms of a sun navigation hypothesis. The results with the experienced birds, however, do not support such a hypothesis.)