3^8 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY/ 
conditions of the period under review (the winter 191 6-17) and also 
for details as to the damage done to plants by the frost. 
" As regards the instrumental side of the data asked for, the result 
is not entirely satisfactory, either as regards quantity or quality. 
As regards the first point, they very inadequately represent the 
climate of the British Isles during the period. Their distribution 
was as follows : from Scotland, Ireland and North Wales two schedules 
were received from each ; from Cumberland, Northumberland, Yorks, 
Cheshire, and Derby one from each ; and from South Wales, and 
fourteen counties in the south of England, of which Suffolk is the 
most northern, a total of forty- four. 
" Besides being scanty, the instrumental data also revealed the 
existence of a great diversity of method, or perhaps a great want 
of method of any sort, in setting up the instruments, and in observing 
and recording their indications. In a few cases screens had been 
employed for the thermometers, and the information supplied was 
evidently based on observations carefully and regularly made and 
recorded ; in others a casual note revealed that the thermometers 
had been hung on a post, or affixed to a wall, without protection 
from sun or rain (?), and at heights of from one foot to five feet above 
the ground ; and whilst there was no evidence of regularity and method 
in reading them, there were not wanting indications that in the absence 
of proper records the memory alone had often been relied upon for 
answers to the questions asked in the schedules, both as to the season 
under discussion and also those previous seasons with which they 
were being compared. 
" It was evident therefore that the climatological data contained 
in the schedules were alone inadequate for the purpose in view ; 
and it became necessary to combine with them information derived 
from other sources in order to arrive at a just conception of the 
character of the weather during the winter of 191 6-17, the period under 
review, and wherever possible this has been done. 
" It may be permissible here to remark, with reference to this 
lack of reliable data, that it emphasizes the need of having in every 
important garden a simple set of climatological instruments, which 
should show the temperature and hygrometrical condition of the 
air (a dry and a wet bulb thermometer), and the amount of rainfall ; 
and if a record of the duration of sunshine could be added so much 
the better.* 
" A thermometer is an unemotional instrument that, provided it 
be properly made, properly set up and screened from sun and rain, 
and regularly read, can be relied upon to give unbiassed answers to 
the following three questions, asked in the ' Schedule of Queries ' 
upon which the present inquiry is based : (1) What were the lowest 
temperatures recorded during the winter ? (2) How do they compare 
* With the exception of the Sunshine Recorder these instruments need cost 
but little. It is possible to purchase thermometers quite good enough for the 
purpose suggested for one shilling each. 
