NATURE 



[March 6. 19 13 



is probably considerably greater than can be attri- 

 buted to errors of measurement. Moreover, helium 

 lines occur in connected series, and there is no justifi- 

 cation for supposing that one of them would be repre- 

 sented in the absence of other members of the same 

 series. The oxygen line 5330-84, which, it was pointed 

 out, is nearly coincident with neon 5330-90, is one 

 component of a triplet forming part of a series, and 

 would not appear in the absence of the associated 

 lines. 



To my mind the proper conclusion to be drawn 

 from the comparisons is that the respective spectra 

 are quite distinct, and that the approximate coin- 

 cidences are entirely accidental. A. Fowler. 



South Kensington, March 3. 



The Influence of Icebergs on the Temperature of the 

 Sea. 



Prof. Barnes, in Nature of February 20, gives an 

 important piece of information which seems to me to 

 enable us to clear up the confusion at present sur- 

 rounding this subject, as it explains the reason for the 

 different results obtained by Prof. Barnes in his earlier 

 and later observations, and why his results differ from 

 those of previous observers ; and it also helps us to 

 an explanation of the puzzle of the rising temperature 

 of the sea on approaching icebergs, found by Prof. 

 Barnes. The earlier observers made their tests in 

 the cold but weak sea-water floating on the surface. 

 Prof. Barnes's first tests were made at a depth of 

 5 ft. The first part of his curve, Fig. 1 (Nature, 

 June 20, 1912), gives the temperatures of the 

 sea as the thermometer passed under the outer edge 

 of the cold surface water, and was thus made in the 

 ordinary sea-water, and gave the temperatures below 

 the cold surface water, until the ship arrived within 

 a mile of the iceberg, where the increasing depth of 

 the cold surface water began to affect the thermometer, 

 and from that distance, the thermometer being now in 

 the cold surface water, the temperature fell rapidly 

 as the ice was approached. The thermometer in Prof. 

 Barnes's second ship, he tells us in his last letter, was 

 placed at a depth of 18 ft. below the surface, and 

 seems to have been always too deep to get into the 

 cold surface water. 



We now come to the question as to why these last 

 observations of Prof. Barnes show a constant rise in 

 the temperature of the water as icebergs were ap- 

 proached. We can scarcely imagine ice to have any 

 heating effect, and solar radiation does not seem to 

 meet the case. It would, however, appear that we 

 do not require to call in the aid of sunshine, or other 

 outside source of heat, to explain this rise in tempera- 

 ture, as it can be more simply accounted for by the 

 indraught current near the surface having to dip 

 below the cold surface water, its upper warmer water 

 being thus carried downwards towards the thermo- 

 meter. By this explanation there is no heating of the 

 water as it approaches the iceberg, but the warmer 

 surface water coming from outside the cold surface 

 area is carried underneath the cold water to lower 

 levels, so giving a rise of temperature at: these levels. 



If the above explanation be correct it would appear 

 that the surface cold current is the one to be mainly 

 depended on for indicating the presence of ice, because, 

 unless there is some depth of cold surface water, there 

 will be no depression of the inflowing current, and 

 therefore no rise of temperature on approaching the 

 iceberg. Perhaps the best method of observing would 

 be to have two thermometers, one near the surface 

 and the other at a depth of, say, tS ft., writing on 

 the same paper. Under ordinary conditions these two 

 would show nearly a constant difference, but would 



NO. 2262, VOL. qi] 



tend to diverge on the approach of ice, so checking 

 each other, and magnifying the indications. 



John Aitken. 

 Ardenlea, Falkirk, February 22. 



Systems of Lines obtained by Reflection of X-Rays. 



In continuation of the experiments of Mr. W. L. 

 Bragg (Nature, December 12, 1912, p. 410), we have 

 investigated the reflection of X-rays by mica and rock 

 salt. In these experiments we found that in general 

 two dark spots are obtained in consequence of the 

 reflection, one of which is crossed by equally-spaced 

 lines, which run at right angles to the plane of re- 

 flection. The distance between the different lines in- 

 creased with increasing distance of the photographic 

 plate from the crystal, and appeared greater with 

 rock-salt than with mica. In some photographs the 

 second spot was also striated. 



The plates cut from the crystals were fastened 

 down to aluminium foil 02 mm. thick. Successful 

 photographs were only obtained with rays of grazing 

 ini idence, an angle of about 80° being used in most 

 cases. 



The regularity in which the fringes were distributed 

 suggests that the phenomenon is due to interference. 

 Further experiments are,, however, --equired before 

 this question can be definitely settled. Since Prof. 

 Barkla and Mr. Martyn (Nature, February 13, 1913, 

 p. 647) have recently described similar results, it may 

 be desirable to publish our preliminary results, of 

 which a more complete description will soon be com- 

 municated to the German Physical Society. 



E. HUPKA. 



W. Steinhaus. 

 Physikalisch-technische Reichsanstalt, 



Charlottenburg, February 23. 



Four-horned Sheep in Scotland. 



So little seems to be known regarding the early 

 occurrence of Scottish four-horned sheep that the 

 following record will bear repetition. It occurs, 

 almost as an aside, in the account of the parish of 

 Moffat, in the lowland counties of Dumfries and 

 Lanark, published in Sir John Sinclair's "Statistical 

 Account of Scotland," vol. ii., p. 292, 1792. The 

 writer of the account, Rev. Mr. Alex. Brown, says : — 

 " It is not long since the sheep in this part of the 

 country, were of the four-horned kind; a few of 

 which, it is said, remain still in some parts of Niths- 

 dale. Their body is smaller, but their w 7 ool finer than 

 those of the present breed. Their want of weight 

 for the butcher, and greater difficulty and danger in 

 lambing have banished them from this place." 



This lowland four-horned race agrees with the 

 Hebridean in the characters of fineness of wool and 

 smallness of body. It also appears to agree 

 in the less tangible character of maternal in- 

 efficiency, for of an experiment carried out in a 

 small Western Islands' flock in the Isle of Man 

 a few years ago Prof. Wallace says ("Farm Live 

 Stock," p. 521, 1007): — "The animals weighed only 

 5 lb. to 6 lb. per quarter, and they proved to be such 

 indifferent nurses that they were eventually put away " 

 — causes remarkably similar to those which 

 "banished them" from south Scotland. At any rate, 

 it would seem clear that the four-horne'd breed of 

 sheep, the last remnants of which in Scotland wen: 

 isolated on the Hebridean and Western Islands, had' 

 at a comparatively recent date considerable outposts- 

 on the mainland. James Ritchie. 



The Royal Scottish Museum, Edinburgh. 

 February 26, 



