November 3, 192 1] 



NATURE 



307 



considerable experimental difficulties, which so far 

 have not been completely overcome. It appears, 

 however, that one of the questions we had set our- 

 selves can be answered, partially at any rate, from 

 some data recently published by Richtmeyer (Phys. 

 Rev., July, 1921, p. 13), who has given the absorp- 

 tion curves of molybdenum and silver for homo- 

 geneous X-rays on both sides of their respective 

 critical (k) absorption frequencies, and for lead on the 

 longer wave-length side. Data on lead have been 

 previously given by Hull and Rice (Phys. Rev., vol. 8, 

 p. 326, 1916), who have also determined one point on 

 the shorter wave-length side. The values of Richt- 

 meyer for lead are proportionately higher than those 

 of Hull and Rice, apparently indicating that the latter 

 have inaccurately determined the thickness of their 

 thin absorbing screen. 



By plotting log (r/p) against log y\, the double values 

 of the mass absorption (fluorescent) coefficients of 

 molybdenum and silver at their respective critical fre- 

 quencies can be accurately determined. In the case 

 of lead the accuracy of the upper value (for the shorter 

 wave-length side) is governed by the accuracy of 

 the single determination of Hull and Rice as cor- 

 rected from the Richtmeyer data. The following 

 results are obtained for the ratio of the values 



/ i~ " when 1^ is the k critical absorption frequencv 



of an element and fiv is infinitely small. 



Lead 





Silver 



Molybdenum 



3"5 



676 



7 "06 



Critical frequency 2r58xio** 6'i86xio'® ;4-854Xio 



The relation between R and the critical frequency 

 is linear, and is expressed by the equation 



R = o-2i2 X io~'*(3S"i X loi^-v,). 



The relation between R and the atomic number 

 has been tested, but it does not appear to be nearly 

 as linear as the one given above. The data from 

 which this generalisation has been made are ad- 

 mittedly incomplete, and the experiments are being 

 continued in the hope of testing the extent of its 

 validity. W. Ewart Williams. 



B. L. WORSNOP. 



Wheatstone Laboratory, King's College, 

 Strand, W.C.2, October 15. 



The Film-photophone. 



It may be of interest to readers of the Note in 

 Nature of September 29, p. 161, to learn that the 

 photo-electrical equipment of the "speaking-film" is 

 quite new. It consists of an antimonite cell, and 

 was constructed especially for this purpose by the 

 present writer. A fragment of a single crystal of the 

 mineral antimonite (found in Japan and Borneo) is 

 connected to electrodes (of very large surface) in such 

 a manner that air and humidity are excluded. An 

 even flow of current is thus attained and the sudden, 

 unexpected jerks which formerly destroyed the 

 acoustic effect are avoided. 



The photo-electrical properties of antimonite have 

 been known for a comparatively short time. The dis- 

 cover}- was made by F. M. Jaeger (of Zaandam) in 



NO. 2714, VOL. 108] 



1907. The exceptionally high resistance of the first 

 cells was at that time, before the advent of the 

 amplifier, a serious obstacle to technical uses. In 

 191 1 I succeeded in constructing a cell of lower resist- 

 ance which enabled Prof. B. Glatzel in 1912 to 

 demonstrate graphically by the oscillograph the sur- 

 prising rapidity with w^hich the antimonite adjusts 

 itself to varying intensities of light. 



That synthetic antimonite, made by melting 

 together antimony and sulphur in suitable proportions, 

 is sometimes very sensitive was shown by Olie and 

 Kruyt in 19 12. 



The investigations of F. C. Brown seem to indicate 

 that single crystals of selenium might also be used 

 with advantage. But they are difficult to make, and 

 the problem of affixing the electrodes is not yet solved, 

 although solutions may be said to be in sight, 



W. S. Gripenberg. 



Helsingfors, Finland, October 13. 



The first two sentences in Dr. Gripenberg 's letter 

 are misleading in the sense that they suggest that the 

 onlv photo-electrical equipment capable of being used 

 with "speaking-films" is the antimonite cell. It is 

 well known, of course, that other substances besides 

 selenium respond to .fluctuations of light intensity, 

 and antimonite is, apparently, one of them. Another 

 is the "thalofide cell," which has recently beeri adver- 

 tised extensively, and was invented by T. W. Case. 

 I believe I am correct in saying that the sensitive 

 substance in this cell is thallium sulphide. The 

 comparative merits of these various substances will, 

 no doubt, ultimately decide which is best to use with 

 speaking-films. For the present, at any rate, selenium 

 has been bv no means completely ousted — a fact 

 which is made evident bv its adoption and use in 

 connection with the film-photophone of Mr. Bergland, 

 to the efficient performance of which attention has 

 been directed by the Times correspondent. 



The Writer of the Note. 



Rainfall Records ?t Rothamsted. 



The following rainfall figures from Rothamsted are 

 worth noting. The records date back to 1852 for the 

 large rain-gauge (i/iooo acre), but for the purpose 

 of comparison the figures for the last fiftv years are 

 taken, since the three percolation gauges (also i/iooo 

 acre) were not built until 1870. They relate to the 

 harvest-year, September i-August 31 : — 



Harvest Year. 



Percolation 



Throu<;h Through Throiieh 



20-in. 40-"n. 60-in. 



Rain. soil. soil. soil. 



Inche«. Inches. Inches. Inches. 



Average for last 50 years 29-500 I4"834 I5'482 I4'659 

 September i. 1920, to 



August 31, 1921... 16*282 6-921 7161 6-8l2 



For the past eight months (February i-September 

 30) the figures are : — 



1921 

 Average 



8-511 

 18-239 



1-121; 

 6525 



1-230 1-^76 

 6910 6-528 



The rainfall for the period September i, 1920, to 

 August 31, 1921, is the lowest since the records 

 started in i8-;2, the previous lowest being 19-504 in. 

 in 1897-98. The highest figures for the period are 

 41048 in. in 1878-79. • \y. D. Christmas. 



Rothamsted Experimental Station, 

 Harpenden, October 27. 



