I So 



TsNATURE 



[0CT0BER7, 1920 



length i6±4 A. lonj^er than the lint's which they 

 accompany, are readily accounted for as due to the 

 heavier of the two isotopes, atomic weights 35 and 37, 

 of which .\ston {Phil. Mag., vol. xxxix., p. 611, 1920) 

 has shown ordinary chlorine to consist. An approxi- 

 mate theory shows the wave-length of the band 

 centre to vary as the square root of the effective mass, 



m=-^^'~^, where w, is the mass of the hvdrogen 



nucleus and Wj that of the chlorine atom. Taking 

 TO = 35/36 for the lighter and 37/38 for the heavier iso. 

 tope, the calculated difference between the wave- 

 lengths of corresponding lines for the two isotopes 

 comes out 13 A. This is much larger than the differ- 

 ences of about 0004 A. which have been found between 

 lines of the isotopes of lead (Aronberg, Astrophysical 

 Journal, vol. xlvii., p. 96, 1918, and Merton, Roy. 

 Soc. Proc, A, vol. xcvi., p. 388, 1920). 



I hope soon to publish a more detailed account of 

 the theory and measurements of these lines, probably 

 in the Astrophysical Journal. F. W. LooMls. 



New York University, University Heights, 

 New York City, U.S.A. 



A New Visual illusion. 



A VISUAL illusion which I have never seen referred 

 to may be of interest. If the gaze is steadily fixed 

 for a few minutes on a spot in the descent of a water- 

 fall which has a fairly long unbroken fall, and after- 

 wards quickly transferred to the adjacent hillside, 

 the hill itself appears to rise slowly as a whole, some- 

 what as though it were an elevator. The same result 

 mav be obtained by looking fixediv at the broken 

 surface of a rapid and fairly wide stream ; on direct- 

 ing the eyes suddenly to the opposite bank this appears 

 to move slowly up-stream. 



The illusion seems to me to be due to the rapidly 

 moving water tending to carry the vision along in its 

 own direction, as occurs when any moving object is 

 unreflectively observed. But while the eyes are kept 

 fixed on the selected point, this tendency becomes 

 counteracted by a series of slight and rapid but un- 

 conscious muscular efforts which prevent the eyes from 

 following the motion, .\fter the gaze is removed to the 

 adjacent stationary ground, these muscular efforts 

 automatically f>ersist for a short time, thus causing 

 the ground to appear to move in the opposite direc- 

 tion to that of the water. But possibly a more 

 accurate explanation can be advanced. 



J. E. Turner. 



55 .Mlerton Road, Mossley Hill, Liverpool. 



Plant-life in the Cheddar Caves. 



SiNCE reading the letter relating to plant-growths 

 in the Cheddar caves by Mr. L. Pendred which 

 appeared in N.4TURE of August 5, I have been able, 

 by the courtesy of Messrs. Gough, to examine the 

 plants in situ, and to secure a quantity of material 

 for fuller investigation. 



The green patches on the cave-walls were found to 

 consist of a small green unicellular alga. The loose 

 cave-earth on the sides of the cave-paths yielded a 

 few specimens of fern prothallia. The plant patches 

 in the neighbourhood of the electric lights were found 

 to consist of the following species of mosses : Plagio- 

 thecium denticulatuni, Amblystegium serpens, and 

 Fissidends bryoides, all of which are fairly common. 

 My determinations of these species have been con- 

 firmed by Mr. A. Gepp, of the botanical department 

 of the British Museum (Natural History). 



I think Mr. Pendred 's suggestion that the spores 



NO. 2658, VOL. 106] 



were carried into the caves by the spades or on 

 the clothes of the workmen is highly probable, 

 or that the excellent ventilation maintained in the 

 caves may have resulted in the spores being carried 

 in by air-currents. In any event, the dampness and 

 the air-currents would be factors assisting in the 

 subsequent germination of the spores. 



Edith Bolto.s. 

 Armstrong College, Newcastle-upon-Tyne. 



Old Maps. 



Referring to the kind notice of my presidential 

 address to the Conference of Delegates at the Cardiff 

 meeting of the British Association in Nature of 

 September 16, p. 90, time did not permit details to 

 be given of the evolution of Scottish maps, or those 

 of Faden, etc., would certainly have been referred 

 to. The large map of Gary's which I mentioned 

 was not the one you surmise, " with the coach roads 

 coloured in blue, which is on the scale of five miles 

 to an inch," but the map on the scale of two miles 

 to an inch. The typist or printer, quite pardonably, 

 has apparently mistaken my "two" for "ten." With 

 regard to Griffith's map of Ireland, I still contend 

 that the date of the map " to be included as a classic " 

 is that of 1853. I know there was a slightly improved 

 edition in 1855 ; there is one in the library of the 

 Geological Society of London, but that was the third 

 edition, and not the second, as stated in your 

 columns. T. Sheppard. 



The Museum, Hull. 



I CONFESS that I was not aware of a large map by 

 J. Gary on the scale of one inch to two miles. The 

 one that I mentioned, with mail-coach routes in blue, 

 is of constant service to me. Great confusion has 

 been caused in regard to Griffith's geological maps of 

 Ireland by references to them as successive editions, 

 as if all were published and on the same plan and scale. 

 Maxwell H. Close (Journ. R. Geol. Soc. Ireland, 

 vol. v., p. 136, 1879) is, I think, responsible for 

 calling a geological map exhibited in 1815 "the first 

 edition," but he carefully added that it was never 

 printed, and he evidently meant only " the first form." 

 He emphasised the fact, by underlining the word, 

 that W. Smith's map of England was published in 

 the same year. Griffith's map of 1835 was, according 

 to John Phillips {op. cit. above, p. 138), large, but 

 also unpublished. Phillips utilised its details in 1838. 

 In 1838 a coloured geological map by Griffith was 

 issued in connection with the Report of the Railway 

 Commissioners, scale one inch to ten miles. A few 

 months later in the same year his large map (one 

 inch to four miles) appeared under the same auspices, 

 and was, as Close tells us, sold to the public from 

 March 28, 1839. These two maps of 1838 can 

 scarcely be called two editions of the same ground- 

 work, since they were both simultaneously in pre- 

 paration. The large map of 183S (published, with 

 date, in 1839) was completely revised and re- 

 engraved, with the addition of mineral localities, and 

 issued in six sheets in .\pril, 1855, the date being 

 engraved on it against Griffith's signature. The map 

 of 1853 was a small one (one inch to sixteen miles), 

 and was issued in a guide to land-valuers. 



The plates of the "classic" map of 1855 are pre- 

 served in the Ordnance Survey Office, Dublin. From 

 inquiries that I have recently made of this office and 

 of the original publishers, no coloured copies seem 

 now to be available. 



It is rarely that one has a chance of correcting Mr. 

 Sheppard. The Writer of the Note. 



