SIR,—In his letter about Passchendaele (Spectator , December 27) Brigadier Desmond
Young states : fact the monsoon conditions of August, 1917, were no worse than in the three previous years.' This statement needs correction; it is quite contrary to the evidence of the actual records which show that the weather in August, 1917, in and behind the battle area, was exceptionally bad. The rainfall directly affecting the first month of the offensive was more than double the average; it was over five times the amount for the same period in 1915 and in 1916. The period is July 29 to August 28. The rainfall at Vlamertinghe was 157 mm. (6.18 inches) in 1917 and 29 mm. in 1916; at St. Omer it was 30 mm. in 1915 (there was then no rain-gauge at Vlamertinghe)• The quite exceptional heavy rain from July 29 to August 4, 1917, was followed by muggy, stagnant weather which prevented the drying by evaporation, normal in the intervals of fair weather at that time of the year.
Brigadier Young may have been misled by an unconsidered statement of Charteris that in Flan- ders the weather broke early in August with the regularity of the Indian monsoon. This statement is so contrary to recorded facts that, to a meteorologist, it seems too ridiculous to need formal refutation. In 1915 there were twenty-two days in August with- out any rain and in 1916 there was no rain from the middle of July to the middle of August and then only small amounts until the end of the month. The official record for Lille, 1878 to 1913, showed that the first ten days of August had had 40 per cent. less rain than the last ten days of July. —Yours faithfully, G. GOLD Formerly Commandant, Meteorological Section R.E., France, 1915.19 8 Hurst Close, NW11