As I was not able to visit Stanton Drew when the significance of the northerly alignments struck me, I made an appeal to Prof. Lloyd Morgan, of whose pamphlet I have so largely made use, to obtain some theodolite observations. As a result such observations have been made by himself and Mr. Morrow, from whom I have recently received a report with full permission to make use of it in this place. The monuments are not easy to measure, as the centres of the circles are not readily determined, as so many of the stones are either absent, recumbent or buried. In my rough reading of the Ordnance map given in Fig. 47, I thought I might be guided by taking centres, such that the avenues would be aligned on them as at Stonehenge. I had not then seen the Dartmoor avenues, which in some cases are not aligned on the centres. In this it is possible that I was wrong, as both Mr. Dymond’s and Mr. Morrow’s observations suggest that the avenues are really of the Dartmoor pattern. Mr. Morrow writes: “The centres of the circles are (to a certain small extent) a matter of choice, a difference of a few minutes may easily occur. In dealing with the avenues a larger discrepancy may occur. I have taken what, in my opinion, was the best centre line of each avenue and thus determined its azimuth. But I believe that originally the southern line of stones forming each avenue was directed towards the centre of the corresponding circle, and that the avenue was then completed by the erection of a parallel line of stones. A difference of a few degrees may thus be accounted for in the azimuth supposed to have been originally marked out.” About Mr. Morrow’s azimuths there can be no question. He writes: “The instruments used were, first, a 6″ theodolite, and second, a 6″ transit theodolite. The final results were obtained with the latter. It cannot be reversed when measuring elevations. I tested it very carefully for the adjustments of (_a_) line of collimation at right angles to the horizontal axis, (_b_) horizontal axis perpendicular to vertical axis, and (_c_) line of collimation and spirit level parallel to each other. The instrument was in first-rate order, the error in elevation, for example, being within that corresponding to a slope of 1 in 40,000; that is well within the limit of 20″ to which vertical angles can be read. “The meridian was obtained by two different methods applied several times, the results agreeing very closely. Readings of azimuths and altitude of sun were taken between three and four hours after noon, corrected for semi-diameter, &c., and the true bearing obtained with the aid of the latitude and the declination given in Nautical Almanac (corrected for time).