It is not easy to make a bag of canvas which will hold water, but by dint of fine stitching and a special kind of beeswax, our naval leader succeeded in producing three chargals which did yeoman service. The map on which we were to rely was a French one, forty years old, and on a scale of about twenty-four miles to the inch. An officer had bought it for five pounds from a Greek dentist at Kastamoni. As it happened it was not bought primarily for escape purposes, but we persuaded him to sell it to us on his leaving Changri for Geddos. In this the hill features were very indistinctly shown by vague hachuring, and even a big river such as the Kizil Irmak was in several places shown dotted, signifying not that this dried up during parts of the year, but that no one had surveyed it. An up-to-date but very small map had been received from home by means of a series of six "bananas," each containing a tiny section; but, owing to our change of plan, this showed little of our proposed route. The "sun compass" needs some explanation. This was an invention of Captain A. B. Matthews, D.S.O., R.E., who had been a prisoner of war at Yozgad since the fall of Kut-el-Amara. Wishing to make a rough survey of the immediately surrounding country for the use of the Hunt Club, and finding that local magnetic attraction made a compass altogether unreliable, he bethought him of a simple means of utilising the sun, which in the wonderful climate of Asia Minor is rarely obscured throughout the spring, summer, or autumn. The "sun compass" consists merely of a thin wooden disc of say 5 inches diameter, with the outer edge divided into 360 degrees, and with a hole at the centre through which can be inserted a piece of stiff straight wire. A table of the sun's bearing at any hour on any day completes the instrument. In actual use the disc is held horizontally, with the graduations upwards, and the wire kept vertical and protruding above the disc. Then, by turning the latter till the shadow of the wire falls on the sun's bearing plus 180 degrees, you have the disc set to read off true bearings in any direction. Captain Matthews was also responsible for the star charts. By means of two maps of the heavens obtained from a book on travel, published by the Royal Geographical Society, he devised from first principles a "bus" consisting of three concentric cardboard discs. By means of these it was possible, almost mechanically, to read off the bearings of the brighter stars in the main constellations for any hour and any night of the year. It was thus possible to obtain a series of charts showing on which star one should march for any required bearing, and at any particular time. We prepared them for all hours of the nights from the 1st August to the 15th September 1918. This chart-book was of value as a check on a magnetic compass by night, but assumed an elementary knowledge of at least those constellations which would be of use for the particular purpose in view.