ON THE RAINFALL OF THE BRITISH ISLES. 193 



Wales, Scotland, and Ireland. (2) Those who are willing to help on the 

 completion of the collection will see at once what I have, and what I have not ; 

 and thereby they will be saved much needless copying and I much corre- 

 spondence. 



A barren index, however, throws little light on the history of rainfall-in- 

 vestigations. Before entering on this I desire to meet at the very outset an 

 objection sometimes raised, viz., that we cannot trust very old observations, 

 and therefore 1 may be busy collecting useless materials. I maintain that 

 we can trust them, and have the pleasure of knowing that my friend Mr. 

 Glaisher fully agrees with me on this point. I think them far more reliable 

 than many modern ones ; for in the 17th and early part of the 18th century, 

 to measure the fall of rain was esteemed a serious undertaking, only to be 

 accomplished by first-class men. The repeated reference to the height of 

 their gauges, their diameter, and the number of pounds and ounces troy cor- 

 responding to an inch of rain over the area of the " tunnel " of their gauges, 

 and the details frequently given, combine to render it certain that they took 

 every reasonable precaution to secure accuracy. The results they obtained, 

 and which they would hardly credit, we of the 19th century know to be just 

 what was due to the situation of the gauges. Another point to be remem- 

 bered is this : — it is not intended to use any old observations in determinino- the 

 absolute mean fall at any place, or, in other words, in determining the 'geo- 

 graphical distribution of rain ; that will be done from recent observations with 

 tested instruments. The old observations will only be used for determining 

 the existence or otherwise of secular variation, and for that purpose an in- 

 accurate gauge would do as well as a perfect one. 



The earliest observations I have yet obtained are those of Mr. Townley, of 

 Townley in Lancashire, extending (with two intervals) from a.d. 1677' to 

 1705, or twenty-eight years. Mr. Townley believed his to.be the first made 

 in England ; I do not feel sure that they were so. The- only other observer in 

 the 17th century whose observations have yet been obtained was the Rev. W. 

 Derham, of Upminster, near Romford in Essex, who began in March 1696^ 

 and has left a series of yearly totals (months also for some years) for nine- 

 teen years. 



The longest and most perfect record at present obtained was kept by 

 Thomas Barker, Esq., of Lyndon in Rutland; it extends from 1736 to 1794 

 or fifty-nine years. It is doubtful if so long a record will ever again be kept 

 by one person, with one gauge, and without interruption. 



Erom Table I. it will readily be seen how sudden and great has been the 

 increase of observers during the last thirty years, and especially durino- tho 

 last five, in which one finds the results of the recent grants of this Associa- 

 tion. From 1677 to 1800 the returns never number more than twenty-six 

 per annum, and averaged only four ; and in subsequent ten-yearly periods the 

 numbers average, up to 1810, 15 ; to 1820, 21 ; to 1830, 40 ; to 1840, 103 ; 

 to 18o0, 1/8; to 1860, 366; and at the present time the returns number 

 rather more than a thousand. It is right to mention that no records have vet 

 been obtained for 1687 and 1688, 1694 to 1696, 1717 to 1721, 1724 and 

 1 / 25, since which time there is no year without one or more complete registers • 

 that is to say, rain-records go back complete 140 years, and with intervals 

 about fatty years more, or nearly 200 years altogether. 



1865. 



