628 



NATURE 



[May 13, 1922 



to demonstrate them hitherto in meteorological 

 records. For most of them some meteorological 

 parallel can in fact be found. 



(3) Periods for which there is good, but not first- 

 rate, evidence both in wheat prices and in meteor- 

 ology. There are four of these with lengths of 

 3-415 years, 4-415 years (traced by Mr. Baxendell in 

 rainfall), 5-960 years (traced by various writers in 

 barometric pressure), and 8-050 years (no doubt the 

 same as the period to which Prof. H. L. Moore and 

 others have directed attention). 



(4) The fourth group includes periods, some of 

 which no doubt have reality, but all of which present 

 inconstancy of action, changes of phase or other 

 puzzling features. These include periods of 2-735, 

 5-423, 7-417, 12-050 and 17-400 years in addition to 

 the well-known eleven-year period of the sunspots, 

 which reappears in wheat prices with much of its 

 normal instability of character. 



The large number of periods found is striking. On 

 a priori grounds, however, there is nothing really 

 surprising in the suggestion of so many separate 

 periodicities in the weather. Comparison between 

 the weather cycles indicated by analysis of wheat 

 prices from 1550 to 1850 and the actual rainfall in 

 the ensuing 72 years, 1850 to 192 1, confirms the 

 view that these periodicities are real and important. 

 For the purpose of this comparison, 11 of the 13 

 cycles in the first three groups are combined by a 

 simple graphic method, the lengths and phases taken 

 being exactly' those determined by harmonic analysis ; 

 the combined result is shown in a single " synthetic 

 curve" and is compared with the rainfall at 24 

 stations in Western and Central Europe {i.e. roughly 

 the same area as that covered by the wheat prices 

 records) . A very high measure of agreement appears 

 between the synthetic curve derived from wheat 

 prices before 1850 and the rainfall as actually recorded 

 after 1850. 



In particular, the synthetic curve shows de- 

 pressions foretelling lack of rain (which, in the area 

 under review, is generally beneficial to wheat) at 

 each of the markedly dry years in the past 70, namely 



1857, 1864, 1870, 1874, 1883-4, 1887, 1898, 1904, 

 1908-9, 1 92 1. It shows peaks foretelling heavy 

 rain in the rainy years 1852, 1866, 1872, 1876 (for 

 1877), 1906, 1912, 1916. The only important dis- 

 crepancies are the failure of the synthetic curve to 

 show peaks for the rainy years, i860 and 1903, and 

 a depression for the somewhat local, though severe, 

 drought of 1893. In other words, the synthetic 

 curve which, subject to certain reservations, could 

 have been drawn in 1850, if then drawn, would have 

 foretold nearly all the important droughts and rainy 

 seasons of the next 70 years. The drought of 192 1 

 stands out quite remarkably well. 



In view of the inevitable errors in the simple 

 hypothesis upon which the synthetic curve has been 

 constructed, the very large measure of agreement 

 obtained in spite of these errors is all the more 

 convincing as to the substantial correctness of the 

 results of the analysis. The question whether definite 

 periodicities in the weather exist and can be dis- 

 covered, must now be answered in the afi&rmative. 



The two most interesting questions of all must 

 still be left unanswered. The first is, what will be 

 the weather next year and in the following years. 

 As to that no prophecy is made at all ; rather, 

 the opportunity is taken of definitely withdrawing 

 anything which might have appeared like a prophecy 

 in the earlier paper. The author is not prepared to 

 say anything as to whether 1923 will be wet or dry. 

 The full mathematical analysis in many ways has 

 confirmed the earlier conclusions, but in more im- 

 portant ways goes beyond them; the general result 

 might be quite different. Trustworthy prophecy of 

 general weather conditions on the basis of periodicities 

 now demonstrated should become possible in the near 

 future, but only after a far more elaborate investiga- 

 tion than it has yet been possible to make. 



The second question is, as to the seat and the 

 physical cause of the periods appearing in the weather. 

 A further analysis of the economic data may be 

 helpful in some directions, but for the most part, 

 that is a question for astronomers and physicists 

 and should be left for them to answer. 



The Teaching of Natural History in Schools.^ 



pROF. HICKSON and the other distinguished 

 ■*■ zoologists who have drafted the memorandum 

 referred to below are deeply and properly concerned 

 at the general neglect of zoology as a school subject. 

 They maintain that this science should serve as a 

 means of introducing youth to many of the greatest 

 problems of life, and they therefore express astonish- 

 ment, which will be shared by many people, at the 

 findings of the investigators of the Secondary School 

 Examinaton Council on the subject of school natural 

 history. This body, in a recently issued report, com- 

 mitted itself to the remarks that " very few of the 

 candidates (for certain important school examinations) 

 offer this subject (Natural History), and it seems very 

 doubtful whether it is worth while to maintain it as 

 qualifying for a Pass with Credit in Science." 



The members proceeded to express the opinion that 

 the principles of biological science can be better 

 illustrated by means of botany. From this latter 

 view the authors of the memorandum dissent vigor- 

 ously, and although the arguments they use will not 

 perhaps appeal to botanists, there can, we tliink, be 



' ";Memorandum on the Teaching of Natural History in Schools," 

 prepared by the Zoology Organisation Committee at the request of the 

 Committee of Section D, British Association, Edinburgh, 1921. 



NO. 2741, VOL. 109] 



no doubt that all fair-minded botanists will support 

 their conclusion that the principles of biology cannot 

 be taught without reference to the animal kingdom. 

 More is wanted, however, than vigorous protest 

 against examiners' opinions. Zoology is a science of 

 vast range and school time is already taken up with 

 many subjects. It devolves on the teachers of zoology 

 to show in detail the kind of zoological syllabus that 

 can be put into operation in schools as a basis for 

 zoological teaching. In the drafting of such a 

 syllabus they would naturally consult those teachers 

 who have succeeded in maintaining and popularis- 

 ing zoology in their schools. As a preliminary 

 to such a useful piece of work it would be necessary 

 for zoologists to make up their minds whether it 

 would be better to give the school zoology mainly or 

 exclusively a natural history bias, or to attempt 

 to treat the subject experimentally. There are argu- 

 ments in favour of either method, and it might be 

 found possible to draft alternative courses. With 

 carefully drawn detailed syllabuses to guide instructors, 

 and with occasional University courses for teachers, 

 it should not prove impossible to secure for natural 

 history a place in the curriculum of every school. 



F. K. 



