13 
developed system of stolons beneath the surface of the soil, 
again indicating the importance of io underground system in 
drought resistance. 
IlI.—SALT IN THE KEW WATER SUPPLY. 
W. B. TuRRILL. 
At the request of the Director a series of analyses to determine 
the salt-content of the Kew Water Supply was commenced by 
the writer in September 1921, and continued intermittently till 
the end of the year. Since some of the results obtained were 
striking, at first unexpected, and have a practical bearing, it 
has been thought advisable to publish a short summary of them. 
Copies of the reports, giving full details of the collection of the 
water and mud samples, and also of the analytical methods 
employed, are preserved at the Director’s Office and in the Kew 
Library. 
An account of the Kew water system is given in the Kew 
Bulletin, 1897, p. 334. Briefly it may be stated that water is 
allowed to enter the Lake, as required, from the river imme- 
diately opposite, at certain high tides. From the Lake it passes 
to filter beds in the Gardens, and is then pumped up to a reservoir 
in Richmond Park, from whence it comes back to the Gardens 
under considerable pressure. 
Preliminary experiments showed that the water in use at 
Kew during August and September contained considerable 
quantities of chloride, mostly in the form of sodium chloride. 
On this basis it was decided to analyse samples taken from the 
Lake and from the River Thames at different states of the tide. 
The results were calculated both in terms of chlorine and of 
chlorides, considered as sodium chloride. For the purposes of 
™ this note we may quote the results throughout as grams of 
sodium chloride per 1000 cc. of water. Altogether over 20 
distinct samples have been analysed in the course of the 
investigations. 
ix samples of water were collected on September 15 and 
16, and their analysis brought to light the fact that the lake 
water, at this time, was 10 times as salt as the river water, 
collected immediately opposite the inlet to the Lake at high tide. 
It was obvious that the immediate source of salt water at Kew 
was the Lake. The river water contained between 0-15 and 
0-18 grams of NaCl per thousand, varying slightly at different 
levels, and the Lake water 1-5. It was suggested that the 
concentration of salt in the Lake might be due to one or both 
of two causes :—(1) Considerable, and abnormal, evaporation 
during the hot summer. (2) Tidal or river conditions other than 
those occurring when these samples were taken. Finally it has 
