July 14, 192 1] 



NATURE 



627 



fan, has been arranged to pick up vent-pipes and gas- 

 flues from all digestors, as well as the exit pipes of 

 the counterpoised draught-hoods which are pulled 

 down over the evaporating pans when evaporations 

 are in progress. 



The surroundings of the laboratory are shown in 

 some of the photographs. In Fig. 2 appears the 

 adjoining research laboratory, whilst Fig. 4 shows a 

 corner of the fitting-shop and engineering store. This 

 invaluable adjunct contains a stock of pipes, fittings, 

 and tools, some small power-driven machines, includ- 

 ing a screw-cutting lathe, and working places for 



carpentering, fitting, and soldering. The chemical 

 store, which is arranged to contain casks, drums, 

 and carboys, as well as Winchesters, does not appear 

 in the photographs. 



With regard to the question of slinging and heavy 

 work generally, the numerous overhead principals 

 provide so many points from which a lifting block 

 may be hung that it was not considered necessary to 

 install a travelling crane. Two rubber-tyred bogeys, 

 one of which has been specially designed, suffice for 

 the carriage of all the heavier objects which we are 

 likely to have to handle. J. F. T. 



Great British Droughts. 

 Bv Chas. H.arding. 



T T is fortunately seldom that such persistent dry 

 •*■ weather has to be chronicled as that which 

 has now continued for several months. A more 

 complete history of the drought will doubtless be 

 written when all possible facts have been collected. 



At Greenwich Observatory the records show 

 that the rainfall has been less than the normal for 

 nine consecutive months, from October, 1920, to 

 June, 192 1. The total measurement for the whole 

 period is 978 in., which is 7-74 in. below the 

 average for the 100 years ending 1915, and only 

 56 per cent, of the normal. This is the driest 

 period from October to June in the last 105 

 years ; the next driest corresponding period 

 occurred in 1879-80, when the measurement was 

 10-50 in. There is only one longer period at 

 Greenwich — November, 1846, to January, 1848, 

 a period of fifteen consecutive months — with the 

 rainfall below the normal. The controlling factors 

 of the weather have commonly been a low baro- 

 meter in the north of the British Isles, and a 

 relatively higher barometer with anticyclonic con- 

 ditions in the South of England. 



In addition to the Greenwich observations, 

 those at Eastbourne have been chosen to 

 represent the more southern portion of the king- 

 dom. The drought at Eastbourne is scarcely so 

 severe, since the rainfall for each of the months 

 December, 1920, and January, 192 1, was in 

 excess of the average for the period of 

 thirty-five years ending 191 5, chosen as the 

 normal by the Meteorological Office. The 

 total rainfall for the nine months from 

 October, 1920, to June, 192 1, inclusive, is 

 15-62 in., which is 7-95 in. in defect, and 66 per 

 cent, of the average fall. This is 10 per cent, of 

 the average more than at Greenwich. 



Attempts have been made from time to time to 

 detect a weather cycle, but so far these have not 

 been very successful. The favourite cycle with 

 meteorologists is that corresponding with the 

 periodicity of solar activity ; but, so far as the 

 general weather is concerned, it does not yield 

 satisfactory results. Prof. Bruckner, of Berne, has 

 discussed the subject of periodic variations and 

 changes of climate in detail, and his discussion 

 is conducted on lines which perhaps might well 

 be followed by others. For the fluctuations of 

 NO. 2698, VOL. 107I 



rainfall he has made use of observations at 321 

 points on the earth's surface, and of these no 

 fewer than 198 are in Europe. Prof. Bruckner deals 

 with averages for five years, and the period found 

 for the cycle is thirty to thirty-five years. Con- 

 tinuing the cycle to the present time, a 

 period of deficiency of rainfall is shown for the 

 years 1921-25; the previous period of de- 

 ficiency was 1891-95. The next period of excess 

 should occur in 1936-40. The present deficiency 

 of rain seems decidedly a fulfilment of Prof. 

 Bruckner's cycle. 



An absolute drought is reckoned as more than 

 fourteen consecutive days wholly without rain, 

 and a partial drought is a period of more than 

 twenty-eight consecutive days the aggregate rain- 

 fall of which does not exceed 001 in. per diem. 

 No absolute drought has occurred at Greenwich 

 this year, and the only partial drought was from 

 February i to March 5, a period of thirty-three 

 days during which the total rainfall was 0-24 in. 

 The spring drought of 1893 is probably the most 

 severe of recent years ; the absolute drought con- 

 tinued for forty-four days, whilst the partial 

 drought at Dungeness lasted for 127 days, and 

 at North Ockenden, Romford, Essex, for 128 

 days. The abnormal summer of 191 1 experienced 

 three absolute droughts at Greenwich — April 1 1 

 to 24, fourteen days; July i to 23, twenty-three 

 days; and August 2 to 18, seventeen days. There 

 was an exceptionally long partial drought con- 

 tinuing for fifty days, from June 30 to August 18 ; 

 the aggregate measurement of rain during the 

 period was 0-33 in. As many as three absolute 

 droughts occurred in London in the years 1868 

 and 1887, and four in the year 1858. In 1880 

 there was an absolute drought for twenty-eight 

 days — from August 9 to September 5. In the 

 year 17 16 it is recorded that, in consequence of 

 a long drought and a south-west wind, the River 

 Thames became so low that thousands of persons 

 passed across on foot under the arches of London 

 Bridge. 



There is a great diversity in the periodicity of 

 rainfall, and two consecutive summers often 

 differ widely from each other, as shown by the 

 rains in 1920 and 192 1. In 1903, a remarkably 

 wet year, the aggregate measurement of rain at 



