164 



SCIENCE. 



[Vol. XV. No. 370. 



The formation of this class of clouds has, however, had a very 

 disturbing effect upon the conclusions arrived at as to the cause 

 of rain, particularly as they are not absolutely rainless, but oc- 

 casionally give a shower of rain. A shower, however, is a distinct 

 case, which has nothing in common with the great rain, day's 

 rain, or cyclone-rain, capable of yielding eight to ten inches of 

 rain per diem. 



To continue our sketch from nature, at a considerable distance 

 overhead we will generally on such a fine day, notice some ex- 

 tremely thin and airy clouds, — the so-called mare's-tails, cirrus, 

 or cirro-stratus. The sky is often suddenly changed from a per- 

 fectly cloudless one to one completely covered by a thin layer 

 ("pallium," or cloak, as it has been called) of clouds; and rec- 

 ords show that not only has it been the case with the small part 

 of the sky we can obsei've from one place, but that the sky has 

 been suddenly covered by these clouds far and wide for thousands 

 and thousands of square miles. 



As to the height at which these clouds are found, I have par- 

 ticularly asked Mr. Glaisber, who is famous for his wonderful 

 balloon-ascents. He told me that he had gone up five and six 

 miles, and passed through other clouds, but he never seemed to 

 get any nearer to the cirms-clouds. ' He even went up seven 

 miles ; but then he became senseless, and unable to observe any 

 thing. To estimate their height at thirty or forty miles seems, 

 therefore, hardly to be an exaggeration. 



What is the cause of these clouds, and where do they get their 

 supply of vapor to keep them permanent . often during the whole 

 long day in the face of the shining sun? As they are strata-like, 

 and entirely different in their shape from cumulus-clouds, we 

 may feel certain that they are not, like the cumulus-clouds, 

 caused by the ascent of damp air. But if their supply of vapor 



a there is always shelter during such a gale, as shown by the- 

 sheep and cattle which gather there. In other words, the 

 current rises to its maximum height at b above the inland slope 

 of the mountain. Consequently that is where, according to the 

 theory I am opposing, we should expect the greatest downpour; 

 but there is generally next to none, while the rain nearly all 

 falls on the front side of the mountain. At c the current is 

 forced out of its horizontal direction, but a force can only be 

 communicated to an elastic body like air by compressing it. The 

 rain has therefore been caused by compression of saturated air. 



Fig. 5 represents a section of the atmosphere of the southern 

 hemisphere for the month of July, the section being made through 

 the Tropic of Capricorn (see map in my paper, ' ' On the Cause 

 of Trade- Winds"). The intersected parts of the three southern 

 continents are at that time of the year in a dried-up state, and 

 the air-cushions which consequently develop over their surfaces 

 are thrown westwards over the oceans. The height of the vapor 

 atmosphere over the surface of the earth is varying and at its 

 maximum in the anti-cyclones. If we should imagine for a 

 moment that there is no sui'face-evaporation, or that the earth is 



^^ 



a, moist air ; fo, rotating body of surface-air. 



is not to be found below them, it must be on the other side of 

 the strata, or above them. The occurrence of cirrus-clouds is 

 therefore an unquestionable proof of the existence of a uniform 

 layer of saturated air at an exceedingly high level. 



Our table has shown us clearly where the invisible vapors must 

 go to, and the cirrus-clouds now show us where they are stored 

 up at a great distance from the earth. Our only difficulty now 

 is to explain how the vapors are brought down from this high 

 level, or how they become condensed into clouds and rain. For 

 all we know, a cyclone is a body of surface-air brought into 

 rotary motion, and the effect of this rotation is that rain occurs, 

 if anywhere, at or towards the centre of the rotating body of sur- 

 face-air; and this takes place whether the cyclone passes over 

 the sea, over moist ground, or over dry land. The centrifugal 

 force sweeps the surface-air from the centre of the cyclone. The 

 partial vacuum which is thereby produced can only be filled up 

 by the descent of the air or vapor above the rotating body of sur- 

 face-air. This is thereby brought under greater pressure ; and, 

 as the experiment referred to above shows that condensation or 

 rain can be produced by compression, we have hereby arrived at 

 a possible explanation of cyclone-rain. This theory agrees with 

 such a general observation as that the clouds are at their lowest 

 level when rain comes from them. 



In agreement with this compression theory, we may explain 

 the prevalence of rain on the rising slope of coast mountains, or 

 mountain-ridges in general. While rain falls at the centre of a 

 cyclone, the sky at some distance from where the rain falls is in 

 a condition not far from giving off rain, and so the extra press- 

 ure brought to bear upon this saturated air by meeting an ob- 

 stacle, such as a mountain-ridge, causes rain to set in. 



The sketch may represent in section a mountainous coast, 

 against which the moisture-laden clouds are driving from out at 

 sea. The current of air, by meeting this obstacle, is caused to 

 rise, following something like the course shown in Fig. 4. At 



FIG 5. — IDEAL SECTION OF ATMOSPHEEE AT THE TROPIC OF 



CAPEICORN, 



(The circles indicate the outer atmosphere of invisible vapor.) 



perfectly dried up, the dry air would arrange itself in a uniform 

 layer between the earth and the vapor atmosphere as a continuous 

 air-cushion of uniform thickness, and there could be no possi- 

 bility of rainfall. The tendency of the vapor atmosphere is to- 

 wards such a regular shape, but this tendency is counteracted by 

 the varying degree of evaporation at different parts of the earth's 

 surface. A strong surface-evaporation has the effect of decreas- 

 ing the height of the vapor atmosphere over the surface of the 

 earth, while little or no surface-evaporation has the opposite ef- 

 fect. In the space between the anti -cyclones the height of the 

 vapor atmospHere over the ground is comparatively small, and it 

 reaches a minimum when (say, for instance, in the V-depression 

 between two anti-cyclones) the surface-air is, by the currents, 

 of opposite directions along the borders of the anti-cyclones, 



