818 . REPORT— 1900. 



subject liad not been lost sight of by him since then, and he now placed the results 

 befofd this meeting. He did so with some diffidence, and over-cautious meteoro- 

 logists might condemn his action, but they must remember that when Berghaus, in 

 1833, acting upon suggestions made by Zimmermann and Humboldt, published the 

 lir,-t isothermal chart the observations on temperature were even less numerous 

 than those on humidity were at present. His charts, of course, must be looked 

 upon as sketches, but he felt confiient that they brought out the broad features of 

 the subject, and to reduce the sources of error he had limited himself to indicating 

 four o-rades of mean annual humidity, the upper limits of which were respectively 

 50 per cent, (very dry), 65 per cent., 80 per cent., and 100 per cent, (very 

 damp). The relative humidity over the oceans might exceed 8(5 per cent., but 

 in certain regions (' horse latitudes') it was certainly much less, and in a portion of 

 the Southern Pacific it seemed not to exceed 65 per cent., a feature seemingly con- 

 lirmed by the salinity of that portion of the ocean, which exceeded 3'6 per cent. 



Ills second chart exhibited the Annual Range of Humidity, viz. the difference 

 between the driest and the dampest months of the year. In Britain, as in many 

 other parts of the world, where the moderating influence of the ocean was allowed 

 free scope, this difference did not exceed IG per cent., but in the interior of 

 the continents it occasionally exceeded 45 per cent., spring or summer being 

 exceedin"-ly dry, whilst the winter was excessively damp, as at Yarkand, where a 

 humidity of 30 per cent, in May contrasted strikingly with a humidity of 84 

 per cent. In December. 



This great range directed attention to the influence of temperature (and of alti- 

 tude) upon the amount of relative humidity, for during temperate weather we were 

 able to bear a great humidity with equanimity, whilst the same degree of humidity, 

 accompanied by great heat, such as is occasionally experienced during the ' heat 

 terms' of New York and recently in London, may prove disastrous to men 

 and beasts. Hence, combining humidity and temperature, the author suggested 

 mapping out the earth accordiog to sixteen hijgrothermal types, as follows : — 



1. Hot (temperature 73° and over) and very damp (humidity 81 per cent, or 

 more) : Batavia, Camaroons, Mombasa. 



2. Hot and moderately damp (66-80 per cent.) : Havana, Calcutta. 



3. Hot and dry (51-65 per cent.) : Bagdad, Lahore, Khartum. 



4. Hot and very dry (50 per cent, or less) : Disa, VVadi Haifa, Kuka. 



5. Warm (temperature 58° to 72°) and very damp : Walvisch Bay, Arica. 



6. Warm and moderately damp : Lisbon, Rome, Damascus, Tokio, New Orleans. 



7. Warm and dry : Cairo, Algiers, Kimberley. 



8. Warm and very dry : Mexico, Teheran. 



9. Cool (temperature 33° to 57°) and very damp : Greenwich, Cochabambo. 



10. Cool and moderately damp : Vienna, Melbourne, Toronto, Chicago. 



11. Cool and dry: Tashkent, Simla, Cheyenne. 



12. Cool and very dry : Yarkand, Denver. 



13. Cold (temperature 32° or less) and very damp : Ben Nevis, Sagastyr, Godt- 

 haab. 



14. Cold and moderately damp : Tomsk, Pike's Peak, Polaris House. 



15. Cold and dry: 



16. Cold and very dry : Pamir. 



The actual mean temperature of the earth amounted, according to his computa- 

 tion, to 57° F., and this isotherm, which separated types 8 and 9, also divided De 

 CandoUe's ' Mikrothermes' from the plants requiring a greater amount of warmth. 



The author further illustrated his paper by a number of diagrams giving the 

 curves of the temperature, rainfall, and humidity, and also by a chart of the world 

 exhibiting the number of rainy days. 



5. The Origin of Moeh, and their Subsequent Dissection 

 By J. E. Makr, F.R.S. 



In this paper, the influence of vegetation in modifying hill-outlines is first 

 considered, and it is shown that the concave curve of water-erosion is partly 



