Origin and distribution of the polyatomic molecules in the atmosphere

<jats:title>Abstract</jats:title> <jats:p>Of those gases which occur in the upper atmosphere and have strong absorption bands in the infra-red part of the spectrum and which must, therefore, be con­sidered when calculating the absorption and radiation of heat in the atmosphere, only carbon dioxide is uniformly mixed with the air at all heights which we are likely to be dealing with; it will not be considered further here. The vertical distributions of water vapour and ozone are of great interest, particularly when considered together. Water vapour, originating at ground level, usually decreases rather rapidly with increasing height, particularly in the lower stratosphere. This leads to extremely low concentrations at a height of about 15 km. On the other hand, ozone, being formed by the action of solar ultra-violet radiation at a height of 30 km or more, decreases in concentration downwards. We find, therefore, ozone diffusing downwards and water vapour diffusing upwards through the same region of the atmosphere, but, as we shall see, with very different lapse rates. Water vapour The standard hygrometers which are used to measure the humidity from free balloons are only satisfactory at temperatures above about 235°K, and our knowledge of the humidity at high levels in the atmosphere is almost entirely dependent on measurements made with frost-point hygrometers carried on air­craft. The work of the Meteorological Research Flight of the British Meteorological Office is notable for the very large number of measurements made from Mosquito aircraft to a height of about 12 km and more recently from Canberra aircraft to 15 km. Most unfortunately, hardly any measurements having similar accuracy have been made in other parts of the world. However, at the present time Dr A. W. Brewer is in north Norway making such measurements with the kind co-operation of the Norwegian Air Force and I had hoped that some results might have been available in time to report them at this Discussion (see note at end of paper).</jats:p>