March 6. 1913] 



NAT URL 



The receipt of these time signals is so easy a 

 matter that every observatory, and every other 

 institution or person needing accurate time, ought 

 to take advantage of them. 



(11 School Gardening, with a Guide io Horticul- 

 ture. By A. Hosking. Pp. xi + 326. (London: 

 W. B. Clive, 1912.) Price 3s. 6c/. 

 2) Plant Geography. By Prof. G. S. Boulger. 

 Pp. viii+ 136. (London : J. M. Dent and Sons, 

 Ltd., 1912.) Price is. net. (The Temple 

 Primers.) 

 (1) Mr. Hosking has produced a useful book, 

 or rather three small books, under the title of 

 "School Gardens." The second part deals with 

 soils, manures, and the cultivation of garden 

 crops; while part iii. is devoted to garden pests 

 and miscellaneous information. 



Part i., which gives the title to the book, is to 

 us the section of most interest and value, and we 

 would gladly have seen it expanded at the expense 

 of the other portions of the book which require 

 treatment on a more generous scale. On the 

 subject of school gardens the author can speak 

 with a full experience, and his practical details 

 throughout are concise and thoroughly to the 

 point. 



The school garden must not be considered in 

 the light of a paying venture. Its value will only 

 appear when the pupils have become settled in 

 life ; then the stimulus to observation and method 

 and the interest in outdoor pursuits they received 

 will be fully appreciated, and the experiment will 

 reap sufficient reward. 



(2) In the small compass of 136 pages Mr. 

 Boulger has succeeded in compiling a very read- 

 able account of plant geography. The four 

 divisions of the book deal with the evolution of 

 the plant world, the factors of distribution, floristic 

 regions, and botanical ecology or topography. He 

 has wisely devoted the larger part of the book 

 to the consideration of factors of distribution 

 rather than to detailed accounts of the floras of 

 different regions, since the science of plant geo- 

 graphy is so fundamentally bound up with the 

 proper understanding of the ways and means of 

 plant dispersal. 



Mendel's Principles of Heredity. By W. Bate- 

 son, F.R.S. Pp. xiv + 413. (Cambridge Uni- 

 versity Press, 1913.) Price 12s. net. 

 A review of the first edition of Dr. Bateson's 

 valuable conspectus of discoveries in regard to 

 heredity made by the application of Mendel's 

 methods of research, appeared in Nature of May 

 25, 191 1 (vol. lxxxvi., p. 407). Since then a vast 

 amount of work has been done upon various sub- 

 jects of Mendelian analysis ; and Dr. Bateson has 

 endeavoured to take account of this by a series 

 of appendices g'iving descriptive references to 

 papers representing advances upon the state of 

 knowledge when the original volume was pub- 

 lished. Short of rewriting the book, this was 

 probably the best means of giving a new lease of 

 life to a standard work upon Mendelism by a 

 leading exponent of its principles. 

 NO. 2262, VOL. 91] 



LETTERS TO THE EDITOR. 

 [The Editor does not hold himself responsible for 

 opinions expressed by his correspondents. Neither 

 can he undertake to return, or to correspond with 

 the writers of, rejected manuscripts intended for 

 this or any other part of Nature. No notice is 

 taken of anonymous communications.'] 



The Spectra of Neon, Hydrogen, and Helium. 



In the issue of Nature for February 27 (p. 699), 

 Prof. Collie and Mr. Patterson have directed atten- 

 tion to numerous approximate coincidences between 

 lines of neon and hydrogen, from which it is presum- 

 ably intended to be inferred that certain lines of neon 

 are ordinarily present in the vacuum tube spectrum of 

 hydrogen. A further examination of the observa- 

 tional data, however, seems to be desirable. 



Messrs. Collie and Patterson have omitted to state 

 that in the region considered, ^6507 to ^3472, Wat- 

 son's list of the secondary spectrum of hydrogen con- 

 tains more than 700 lines, while that of neon con- 

 tains 260 lines, of which nearly 100 are of intensity 4 

 or greater. With spectra of this complexity there is 

 nothing at all remarkable in the occurrence of a 

 considerable number of approximate coincidences 

 between lines belonging to the two spectra. As 

 stated by Messrs. Collie and Patterson, there are, in 

 fact, twenty neon lines of intensity 4 and upwards 

 which fall within a quarter of an Angstrom unit of 

 lines of hydrogen ; while, if all the neon lines are 

 included in the comparison, and differences of wave- 

 length amounting to a whole Angstrom unit be 

 allowed, the number is brought up to no. 



Messrs. Collie and Patterson, however, do not seem 

 to have realised the accuracy of modern spectroscopic 

 tables, such as they have utilised in the present com- 

 parison. A difference of more than a few hundredths 

 of an Angstrom unit in the tabulated wave-lengths 

 of two lines should now suffice to prove that they 

 have different origins, unless other evidence of prob- 

 able identity is forthcoming. If the permissible dis- 

 crepancy be reduced to one-twentieth of an Angstrom 

 unit, there remain only six lines which might be 

 regarded as possibly common to the two spectra, 

 namelv : — 



Neon Hydrogen 



^nsity Wave-length 



o ■■■ 6175-14 



o ... 6143-30 



o ••• 5343-43 



••• 4537-91 



1 ... 3520-60 

 o ... 347 2 ' 6 5 



Thus, of the hundred brightest lines of neon, only 

 six are found in hydrogen within the probable limits 

 of error, and only one of the six brightest is among 

 them. There is no evidence that the six "coincident" 

 neon lines have special properties which would favour 

 their survival, and the coincidences cannot, therefore, 

 be properly regarded as significant. Even twenty such 

 coincidences would not prove a relation between the 

 two spectra, unless it could be shown that the lines 

 in question were the most persistent of the neon 

 spectrum. 



A verv similar result is indeed obtained when a 

 comparison is made between neon and iron. Over 

 the same range of spectrum there are thirteen of the 

 hundred brighter neon lines which differ by no more, 

 than one-twentieth of an Angstrom unit from iron 

 lines, but this would scarcely be accepted as evidence 

 of any relation between the two spectra. 



As regards the comparison of neon with helium, the 

 mean deviation of the three lines noted is 0-16, which 



