177 
1911-12.] The Fata Morgana 
scattered huts. Such are the palaces which the fairy Morgana creates 
before our wondering eyes. 
The position of its first appearing varies with each occasion. It comes 
into view now here, now there. It does not remain steady, hut moves more 
or less slowly in one or other direction, never lingering in the same neighbour- 
hood for more than ten or twenty minutes. As the mirage passes on 
towards another part of the horizon, the coast-line recovers its ordinary 
aspect. 
The phenomenon as a whole does not last long. Barely an hour will be 
consumed as it moves from one end to the other of the half circle of the 
opposite coast. I have never seen it before noon, and never after six o’clock 
in the evening. 
The Fata Morgana is visible only to an observer whose eye is a few 
metres above the level of the lake. The best height would seem to be from 
two to four metres (six to twelve feet). A shift of a foot or two above or 
below the best position in any given case is sufficient to make the phenomenon 
disappear. This limited range in the position of the eye which ensures the 
visibility of the mirage at once explains the astonishing scarcity of good 
observations. 
Following Dufour, we are in the habit of speaking of the Fata Morgana 
as being due to abnormal refraction, although we know perfectly well that 
there is nothing abnormal in natural phenomena. Given the conditions, the 
consequences necessarily follow. 
“Abnormal” though we call it, the Fata Morgana comes between two 
phenomena which we shall call normal, because they are frequent and easily 
observed. It succeeds the one and precedes the other. 
Under the category of optical refractions in air over the surface of a 
lake, I regard as “ normal ” phenomena those which accompany refraction in 
air over water whose temperature is either higher or lower than that of the 
superposed air. When the air is cooled by contact with cold water, the 
successive layers of air rise in temperature from below upwards. This 
might be called the direct thermal gradient. When, on the other hand, the 
lower layers are warmed by contact with warmer water, the thermal 
gradient is inverse, the temperature in the air falls as the height increases 
through a limited stratum of air. 
The mirage phenomena which accompany refraction over warm water 
are the most frequent. This condition holds throughout the whole day 
during autumn and winter, and during the morning hours in springtime 
and part of summer. The air is then colder than the surface on which 
it rests; the thermal gradient is of the inverse type, and the curve of 
VOL. xxxn. 12 
