154 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1935 



the sun, her crescent narrows, and the horns begin to project beyond 

 their normal position, so that she has been seen as three quarters 

 of a circle, and even as a thin bright ring, Avith a dark interior. This 

 remarkable phenomenon can be seen only when Venus is within about 

 a degree of the sun, and no chance to observe it again will occur till 

 near the end of the present century ; but it has been recorded in the 

 past by several competent observers. Such an extension of the 

 horns — and, above all, the ring-phase — can be explained only as 

 effects of twilight, the illuminated atmosphere of the planet being 

 visible across the narrow dark strip of its surface on the side farther 

 from the sun. 



For the three brightest planets, then, the presence of an atmos- 

 phere is proved by observation, in three quite different, but equally 

 conclusive ways, all of which were well known to astronomers before 

 the end of the eighteenth century. 



Later observations have added evidence of the same type — a few 

 white spots on Saturn, appearing at irregular intervals of some 

 decades, which change shape, shift, and disappear as clouds would 

 do — occasional though fugitive clouds, and a measurable effect of 

 twilight, upon Mars ; and elusive markings on Venus, which can be 

 photographed only with ultraviolet light, and change greatly 

 between one evening's observations and the next. The extent of 

 atmosphere can also be roughly estimated from the results of direct 

 telescopic observation. The surface details of Jupiter (and of 

 Saturn when any appear) may be seen, and photographed, close up 

 to the limb, despite the very oblique angle of view. It is therefore 

 evident that there can be no such extensive gaseous mantle as veils 

 the earth. At least, there is none above the visible cloud surfaces 

 of these great planets — hoAv much there may be below is another 

 matter. The rarefied layer which exists, however, suffices to cut 

 down the apparent brightness of the edge of the planets' disks. The 

 effect of contrast against a dark sky conceals this in an ordinary tele- 

 scopic view ; but the first look at one of these planets in strong twi- 

 light shows that it is actually of surprising magnitude. 



There is more " limb-light " on Mars, and there may be more 

 atmosphere above the visible surface — the real surface, this time; 

 but an atmosphere as thick as the earth's, even if free from clouds 

 or haze, would produce a much greater effect. 



For Venus the layer which produces the elongation of the crescent 

 is remarkably thin, rising only about 4,000 feet above the visible 

 surface. But this represents only the part of her atmosphere which 

 is hazy enough to be seen through the glare of our own sky close to 

 the sun. The top of the atmosphere must be much higher; and the 

 bottom, if the visible surface is composed of clouds, much lower, so 

 that its whole amount may be great. 



