October ,23, 1919] 



NATURE 



153 



LETTERS TO THE EDITOR. 



[The Editor docs 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 anonytnous communications.'] 



The Doubly Refracting Structure of Silica Glass. 



I HAVE recently observed that "silica glass" pos- 

 sesses a remarkable crystalline or quasi-crystalline 

 structure when examined in the polariscope. 



The double refraction is extremely weak, the 

 retardation being probably of the order of i/ioooth 

 of a wave. The structure cannot be advantageously 

 examined with a polarising microscope as used by 

 geologists, for the large number of lenses between 

 the Nicols show enough double refraction to spoil the 

 contrast between the dark and light parts. It is 

 advisable to use a Nicol, and not a glass reflector, as 

 polariser, and to dispense with lenses between the 

 Nicols. A magnifier of 2 in. focus is placed above 

 the analyser. An extremely bright light is necessary ; 

 I have used sunlight reflected straight into the 

 apparatus, thus obtaining an intrinsic brightness com- 

 parable with that of the sun's disc. (No doubt an 

 arc with a suitable condenser would do equally well.) 

 The Nicols are to be accurately crossed so that the 

 sun is invisible. 



If a circular plate of fused silica of "ordinary" 

 quality with polished faces is examined in this ar- 

 rangement, it shows a striking mosaic of dark and 

 bright parts without regular arrangement. The size 

 of this structure is of the order of half a millimetre. 

 It is seen superposed on the ordinary "dark cross" 

 due to_ strain, which extends across the whole disc, 

 ■6 cm. in diameter. 



A rectangular plate of the same class of material 

 showed the dark parts elongated into bands set in a 

 •definite direction, and suggestive of a flow structure. 



A _ circular disc of optical quality silica showed 

 a spiral structure. 



I have examined a large number of specimens of 

 sheet- and bottle-giass without meeting with anv trace 

 •of such a structure, which is evidently something quite 

 peculiar to silica glass. It may be' suggested tenta- 

 tively that silica' glass consists of a mass of "liquid 

 crystals " comparable with those described bv 

 Lehmann in the case of certain organic substances. 



It is intended to obtain photographs of these struc- 

 tures, and to study the effect of heat treatment on the 

 silica until it becomes visibly devitrified. 



Rayleigh. 



Imperial College, .South Kensington, 

 October 20. 



A Search for Fine Wool. 



Mr. Lydekker in his monograph on "Wild Oxen, 

 Sheep, and Goats," published in i8g8, stated that the 

 ancestral stock of sheep is not only extinct, "but 

 totally unk-nown." But in a book on' sheep published 

 in igi2 Lydekker admitted that the mouflon and 

 urial, had probably contributed to the making of 

 domestic breeds. .\s a matter of fact, it has been 

 proved _ beyond doubt (i) that the first domesticated 

 sheep in Europe (i.e. the sheep introduced by the 

 Alpine race about 7000 n.c.) were derived from a 

 urial (Ovis vignei) not unlike the one now inhabiting 



"the Kapet-Dagh, and (2) that nearly pure descendants 

 of the ancient Neolithic breed still survive on the 

 small uninhabited island of Soay (.Sheep Island) near 



"St. Kilda. Further, it is now realised that rams of 



NO. 2608, VOt. I04I 



at least three varieties of Ovis amnion have long been 

 used for maintaining the size and vigour of fat- 

 rumped and other breeds of Central Asia. More 

 important still, it has recently been ascertained that 

 the wool forming the inner coat of several of the 

 wild sheep of Asia is longer than in the Soav, and 

 decidedly finer than and quite as white as superfine 

 .Australian merino, usuallv said to be the finest and 

 whitest wool in the world. Crosses fcjetween Soay and 

 Southdown sheep yield excellent mutton, and bea'utiful 

 wool remarkable alike for its strength and quality. 



It is hence possible that, with the help of the urial 

 and other wild (ypes, new fine-woolled, vigorous 

 varieties of the merino might be introduced. In the 

 meantime, I am anxious to examine the wool of 

 crosses which include wild species amongst their 

 recent ancestors. Sir Joseph Banks, president of the 

 Royal Society when the attempt was made to estab- 

 lish the merino breed in England, was a keeper of 

 sheep, and was "well informed on all points relating 

 to the production and uses of wool." Some of the 

 readers of Nature who, like Banks, are interested in 

 sheep may be in a position to help in the new search 

 for the Golden Fleece. J. C. Ewart. 



The University, Edinburgh. 



Radiation Temperatures. 



I.\ a letter to Nature of October 9 (p. 113) Mr. Mal- 

 lock points out the uncertainties attaching to records 

 of "temperatures in the sun," due to the influence of 

 the nature of the thermometer bulb on the readings. 

 Similar uncertainties attach to readings of instru- 

 ments exposed to a clear sky at night, and with even 

 greater force, for there are two main radiation factors 

 here, one of them being the cold ground — such ther- 

 mometers being- generally placed only four inches 

 above the soil — and radiation to the ground is depen- 

 dent to a very large extent on the nature and condi- 

 tion of the soil, or of the vegetation growing on it. 

 It would appear that the so-called minimum earth 

 radiation temperatures have verv little value as 

 meteorological data. Both these and readings cf 

 " temperatures in the sun " are affected by a source of 

 error other than that noticed by Mr. Mall'ock, namely, 

 the size of the thermometer bulb. With very large 

 bulbs this may not obtain, but with bulbs of ordinary 

 dimensions, say from 1-2 to 02 c.c, the difference 

 caused by size is very noticeable, and is a curvilinear 

 function of that size; with still smaller bulbs the func- 

 tion becomes rectilinear, the apparent radiation effect 

 varying inversely with the linear dimensions of the 

 bulb. Within the limits of bulb-size above mentioned, 

 the differences observed mavbe 5 to 10 per cent, of the 

 total radiation effect ; and this, 'with readings " in the 

 sun," might represent differences of 3° to 6° F. These 

 figures apply to mercury thermometers ; I have no 

 observations yet with alcohol thermometers. 



Differences of radiation temperatures due to the size 

 of the object have an important bearing on subfects 

 other than meteorological records. Thus, it will be 

 impossible to cool a very small object by radiation 

 to a temperature appreciably below that of the sur- 

 rounding medium ; hence the damage done to the 

 nistils and stamens of flowers bv frost cannot be due 

 to radiation, but must be the result of the coldness 

 of the air about them ; therefore, methods of oro- 

 tection from frost dependent on preventing radiation 

 bv interposing a smoke cloud, or smudge, between 

 the tree and the skv, will he ineffective, unless, indeed, 

 the smoke cloud is sufficientlv extensive to cover a 

 larare tract of country, and thus ensure a material 

 reduction in the loss of heat from the ground by 



