September i, 1923] 



NA TURE 



325 



the holes so frequently present in the blocks — which 

 I was told, when a student at the Royal School of 

 Mines, some forty years ago, might be due to the 

 presence of roots and rootlets in the sand before 

 consolidation — but to a special case in which all the 

 details suggested, from my previous knowledge of 

 such things, the work of marine or esturine annelids. 

 Without having seen what I saw, Mr. Chapman 

 questions the validity of the grounds for the 

 suggestion. 



There is no evidence that the blocks to which I 

 referred originated in the Bagshot Sands. They may 

 have been associated with the Reading beds. 



Assuming that all the tubular cavities in sarsens 

 were caused through the presence of roots and 

 rootlets in the original sand, what evidence is there 

 that such roots grew in situ ? It may have been 

 driftwood. Plenty of such wood is to be found, in 

 a lignitic and pyritised condition, in some of the 

 Bagshot beds. I have seen some sarsen rock passing 

 into conglomerate, indicating the proximity of littoral 

 conditions. 



It would be of interest to know if Mr. Chapman has 

 found any grains of comminuted land shells, burrows 

 and bones of animals, and burrows and remains of 

 insects in the consolidated dune-rock he describes. 



C. Carus-Wilson. 



Strawberry Hill, Middlesex, 

 August 18. 



Barometric Pressure in High Latitudes. 



Mr. R. M. Deeley's reply in Nature of August i8 

 my letter in the issue of June 21 does not meet my 

 objection, and since he repeats the misleading state- 



lent that surface pressure is low at the poles it seems 

 lesirable to come to a closer definition of terms. In 



ly letter I made it clear that surface pressure was 

 be regarded " high " at the poles, not so much in 

 Relation to the absolute value as with respect to the 

 )elt of minimum pressure — the theatre of maximum 

 pyclonic activity — about latitude 60° N. or S. ; but 

 "Ir. Deeley under the general term " Arctic regions " 

 Ices not distinguish between the sub-polar regions 

 Ibout 60° N. or S. and the true polar regions about 

 N. or S. 

 In maps produced by the late Prof. H. Mohn in his 

 masterly discussion of the scientific results of the 

 Fram expedition of 1893-96, which confirm in a 

 remarkable way previous work of the late Dr. A. 

 Buchan (see, for example, " Encycl. Britannica," 

 191 1 edn.. Polar Regions), it is shown that in winter a 

 ridge of high pressure (over 762 mm.) is located across 

 the North Polar basin connecting the great Canadian 

 and Siberian high pressure areas, and separating the 

 deep barometric minima of Bering Sea and Iceland 

 (748 mm.), and that this is the season when the 

 pressure gradient is steepest on the north side of these 

 minima, just as it is on the south side. Dr. G. C. 

 Simpson's maps embodied in his famous Antarctic 

 volume are no less emphatic about relatively high 

 surface pressure around the South Pole, even on that 

 part of tne area which is high plateau, and the fact that 

 the expression " Antarctic Anticyclone " is nowadays 

 a household word among meteorologists, geographers, 

 and geologists alike ought to prevent unqualified 

 statements to the effect that the South Pole is a 

 centre of low surface pressure. It has been found 

 that towards both poles pressure decreases from lat. 

 40° N. or S. at the height of about 6000 feet, but at 

 the surface the total effect of all layers is to pro- 

 duce that slight excess of pressure which permits the 

 polar outflow of air which Mr. Deeley admits does 

 take place. 



NO. 2809, VOL. I 12] 



Mr. Deeley then goes on to say that he has 

 attempted to explain why these outflowing polar 

 winds do not reach low latitudes. But any one who 

 thinks in terms of daily weather changes instead of 

 in the cast-iron terms of average wind and pressure 

 charts must realise that polar currents do have 

 abundant opportunity of reaching low latitudes at 

 irregular intervals, this, furthermore, being a foremost 

 point in the theory of Prof. Bjerknes. There is 

 nothing which so paralyses meteorological thought as 

 the habit of regarding rhean charts as though they 

 represented actual unchanging conditions rather than 

 merely the generalised expression of conditions which 

 are perpetually varying to such an extent that the 

 average type depicted is comparatively seldom 

 realised, is generally to a greater or lesser extent 

 distorted, and occasionally altogether subverted or 

 inverted, as during our spells of east wind in the belt 

 of " westerlies." 



Finally, whatever effect the stratosphere may have 

 on pressure at sea-level, Mr. Deeley appears to forget 

 that the broad facts of low winter pressure over the 

 oceans and high continental pressure are mutually com- 

 plementary, as also the reverse distribution of high 

 summer oceanic pressure and low continental pressure, 

 and are well known to be due to seasonal contrasts 

 of surface temperature, the difference of both pressure 

 and temperature being greater in winter. 



L. C. W. BONACINA. 



27 Tanza Road, Hampstead, N.W.3, 

 August 19. 



Is there a Change Of Wave-length on Reflection 

 of X-rays from Crystals ? 



A. H. CoMPTON [Phys. Rev., 21, 207) has recently 

 shown that there is a change of wave-length when 

 X-rays are scattered by an amorphous substance. If 

 reflection of X-rays from crystals is a special case of 

 scattering it would seem that there might be also a 

 change of wave-length on reflection. Assuming such 

 a wave-length change, we have for reflection from a 

 single plane of atoms 



(I) 



cos^i 

 X, 



cos 6, 



where Xj and Xg are the incident and reflected wave- 

 lengths and ^j and 0^ are the grazing angles of 

 incidence and reflection respectively. For reflection 

 from successive planes of atoms we have 



(2) 



d sin 6^ d s in 6^ _ 

 Xi Xj 



where d is the grating space of the crystal and n is 

 the number of vibrations (an integer) difference be- 

 tween the waves reflected from two consecutive 

 planes. Also we have Compton's change of wave- 

 length formula 



(3) . . . X2=Xi+27Sin2(^i+(^2)/2, 



where 7 =A/mc =0-024 A. U. 



From these three relations the formula for the in- 

 cident wave-length X^ can be found in terms of d and 

 ^1, which is the angle measured experimentally. Let 

 X' be the apparent wave-length obtained from Bragg's 

 law wX' = 2fi sin ('i. The relation between X' and Xj 

 is found to be 



(4) 



7»sinM, 



1 ' Xj +7 



From this it appears that X' is greater than \, the 

 true wave-length, by about 0-024 A.U. Also it appears 

 that X' is less for higher orders of reflection, a result 



I 2 



