" This interesting optical illusionwhich suggests the wonderful island in the Atlantic, seen from the isles of Aran near Galway, alluded to in the 'Chorographical description of West, or H-Ier-Connaught', of R. O'Flahertywas caused by the peculiar angle of the light from the setting sun, the reflection of the water of the Solway, and the refraction of the vapour and clouds above the Solway.

From the summit of this hill he had a curious example of the effects of refraction in this region in a similar illusion to that which suggested Poole's inland sea.

The effect of refraction on this level country, when heated by the midday sun, was so great as to cause many of the low sandy ridges to appear like large lakes and inlets of the sea, as in some instances the more distant hills were obscured by its effects.

By preparing prisms of different varieties of tabasheer, Brewster proceeded to determine its refractive index, arriving at the remarkable result that tabasheer "has a lower index of refraction than any other known solid or liquid, and that it actually holds an intermediate place between water and gaseous bodies!"

But all these suppositions were destined to undergo refractions through the medium of a new fact.

Whether it was due to the slanting sunlight, or the refraction from the wonderfully illumined water, I cannot say, but, whatever the cause, I found it difficult to focus my sight properly upon the flying apparition.

The computations for the observations have been exceptionally heavy, from the circumstance that the Sun was very low (86° 14' Z.D. at the last observation) and that it has therefore been necessary to compute the refraction with great accuracy, involving the calculation of the zenith distance for every observation.

The distant blue belt of timber along the Gizhiga River wavered and trembled in its outlines as if seen through currents of heated air, and the white ghost-like mountains thirty miles away to the southward were thrown up and distorted by refraction into a thousand airy, fantastic shapes which melted imperceptibly one into another, like a series of dissolving views.

Experiment proved that this gave too high values for refraction near the vertical compared with those near the horizon, so Kepler "went off at a tangent" and tried a totally new set of ideas, which all reduced to the absurdity of a refraction which vanished at the horizon.

'Relying on this, we have now computed our mean refractions by diminishing those of Bessel's Fundamenta in the proportion of 1 to 0.99797.'The