?oo 



NATURE 



[January 24, ic,oi 



years ; thus averaging about 27 days annually. (The annual 

 number of frost days is about 55.) 

 The year distribution is as follows : — 



1841 7 ... 1856 2 .. 1871 3 ... 1886 4 



1842 — ... 1857 — ... 1872 — ... 1887 4 



1843 — ... 1858 — ... 1873 — ... 1888 3 



1844 I .. 1859 3 ... 1874 3 ... 1889 4 



1845 4 ••• i8^° 3 ••■ '^75 I ••■ '^9° 7 



1846 3 ... 1861 5 ... 1876 2 ... 1891 6 



1847 6 ... 1862 — ... 1877 — ... 1892 2 



1848 I ... 1863 — ••• 1878 3 ... 1893 5 



1849 2 ... 1864 4 ... 1879 5 ... 1894 5 



1850 — ... 1865 4 ... 1880 5 ... 1895 II 



1851 — ... 1866 — .. 1881 10 ... 1896 — 



1852 — ... 1867 7 ... 1882 — ... 1897 — 



1853 I ... 1868 — ... 1883 — ... 1898 — 



1854 2 ... 1869 — ... 1884 — ... 1899 I 



1855 14 ... 1870 7 ... 1885 — ... 1900 2 



162 



The greatest number in any one year is 14, in 1855. Next 

 come 1895 with 11, 1881 with 10, four with 7 each, &c. Con- 

 sidering winters instead ol years, the highest number is 12, in 

 1854-55. On the other hand there are 22 years with none of 

 these days, i.e. more than a third of the whole. We do not find 

 more than y^«r such years in succession ; such a group is pre- 

 sented in 1882-85. 



The distribution in months is as follows : — 



Jan. Feb. Mar. Nov. Dec. 



68 42 5 2 45 = 162 



Thus, January \% facile princeps. February and December are 

 nearly equal. The days are rare in March, and most rare in 

 November. Of the two in November, one was on the 30th, in 

 1856 (i9°'4), the other on the 28th, in 1890 (i8"-3). The latest 

 in March was on the 14th, in 1845 (i3°*i). 



Speaking roughly, we seeni to have had an increase in those 

 very cold days. Grouping in decades we find this :-^ 



1841-50 1851-60 1861-70 1871-80 1881-90 1891-1900 

 24 25 27 22 32 32 



The first three total 76 ; the last three 86. To put it otherr 

 wise, the thirty consecutive years ending 1895 had more of 

 those days than any other thirty-year group. 



Do these days present any definite relation to the sun-spot 

 cycle? I think we may discern (as in the case of frost days) a 

 tendency to greater cold before a maximum of sun-spots than 

 after. This may be variously shown ; here e.g. is one way : 



Compare the group of years from the seventh after a maximum 

 year to the next maximum year (inclusive) with the six years 

 after the latter maximum. We may construct a table as follows : 



a b 



Annual Annual Relation 



Average. Average. a to b. 



1844-48 ... 3-0 ... 1849-54 ... 0-8 ... -t-2;2 



1855-60 ... 37 ... 1861-66 ... 2-2 ... +1-5 



1867-70 ... 3-5 ... 1871-76 ... 1-5 ... -t-20 



1877-83 ... 3-3 ... 1884-89 ... 2'5 ... +0-8 



1890-93 ... 50 ... 1894-99 ••• 2-8 ... -1-2 2 



Av. -f- 1 7 



Thus the group of years ending with a maximum year Shows 

 an average which is always in excess of that of the six-year gioup 

 following. 



These very cold days are often found in groups. Among the 

 longest are February 16-22, 1855, and February 5-10, 1895. 



I may close with a list of the ten coldest days : — 



Min. 



NO. 1630, VOL. 63I 



NATIONAL PHYSICAL LABORATORY.^ 



■"P HE first annual report of the Executive Committee 

 *• of the laboratory, which was laid before the Royal 

 Society at its annual meeting, is in some respects dis- 

 appointing. It contains a record of much valuable 

 labour, rendered useless by the opposition to the site on 

 the Old Deer Park at Richmond, which had been selected 

 for the laboratory ; while the delays caused by that 

 opposition have made the progress of the scheme very 

 slow. 



The Richinond site was chosen by Lord Rayleigh's 

 committee, and approved by the Treasury ; the director's 

 first task, after taking up his duties, was to visit the 

 Reichsanstalt and the Bureau International at Sevres. 

 The courtesy of the authorities of these two institu- 

 tions is suitably acknowledged in the report, and is 

 another evidence of the international character of science. 

 Meanwhile, in order that the new laboratory might, 

 from the beginning, adapt itself to real wants, various 

 committees had been considering the questions which 

 seemed to press most urgently for solution. With their 

 reports before them, the executive committee prepared 

 plans, and authority was given in June last to obtain 

 tenders for the work. Then followed a delay of some 

 four months. During the summer a deputation from the 

 Royal Society waited on Mr. Hanbury, urging that the 

 original scheme should go on ; but towards the end of 

 October it was announced that Her Majesty had been 

 graciously pleased to assign Bushy House and grounds 

 for the purposes of the laboratory, and that in order to 

 meet the additional capital expenditure involved, the 

 Government were prepared to ask Pnrhament to raise 

 the grant of 12,000/. for building to 14,000/. This was 

 accepted by the Council of the Royal Society, but it was 

 pointed out that, to quote the words of the report : — 



The executive committee "cannot, however, conceal 

 from themselves that it will be very difficult for them to 

 maintain and administer a national physical laboratory 

 on the Bushy site for the amount annually allowed by 

 the Treasury, and they fear that it may be necessary for 

 them to press, in the near future, for an addition to that 

 allowance." 



Meanwhile, plan-making had to begin again. With 

 the very cordial assistance of the Office of Works a new 

 scheme was prepared and approved, and now the 

 workmen are on the ground and the alterations have 

 commenced. 



Fortunately, the structural changes necessary will be 

 remarkably small. 



Bushy House is in many respects well suited, as it 

 stands, for a physical laboratory ; the basement, how- 

 ever, was dark and damp, and the whole sanitary arrange- 

 ments needed reconstruction. The basement is to be 

 improved by the construction of a dry area round the 

 house, and the insertion of larger windows ; the present 

 flagged floor is to be removed, and to be replaced by 

 concrete and cement. Modern drainage is to be intro- 

 duced everywhere, and in place of the cess-pools now in 

 use, connection is to be made with the public sewer ; this 

 necessitates a main drain some 300 yards long. 



The house itself consists of a central block about 70 

 feet square, containing a basement and ground floor 

 with two floors over. The ground floor rests on brick 

 groining, forming the roof of the basement ; it is thus 

 possible to secure steady supports for apparatus at 

 almost any point ; the building is very substantial, and 

 it will be easy to maintain a uniform temperature 

 throughout the basement. 



The front of the house faces east approximately ; 

 unfortunately, the two main rooms of the central block 

 on the first floor look south and west respectively ; in 



1 Report of the Executive Committee for 1900, and Programme of Work 

 for 1901. . . 



