May 22, 1913] 



NATURE 



307 



two widely separated objects are much more accurate 

 than micrometer (excluding double image or helio- 

 meter) observations with the same instrument, be- 

 cause the former are affected alike, the same time 

 element being common to each. 



PROPERTIES AND STRUCTURE OF ICE. 



AN interesting account of a number of experiments 

 by Prof. R. S. Tarr and Dr. J. L. Rich, of Cornell 

 University, appears in the Zeitschrift fiir Gletscher- 

 kunde (Band vi., p. 225). The results agree mainly 

 with those obtained by Miigge and MacConnell, and 

 show that, as urged in 1S69 by VV. Mathews, those of 

 Prof. Tyndall and Canon Moseley were inconclusive, 

 through not taking sufficient account of the time- 

 element in the problem. These recent experiments, 

 which were both numerous and designed to test the 

 various properties of ice, show that it welds readily 

 at a temperature of o° C. ;~that when a block of ice 

 has been cut through by a wire and regelation has 

 occurred, optical continuity is re-established, the new- 

 forming crystals being controlled by those previously 

 in existence, and that the welding, at temperatures 

 well below the freezing point, to some extent re- 

 sembles what has been observed in marble after being 

 crushed. 



The authors tentatively advance four proposi- 

 tions ; the first, that the observed deformation is 

 of the nature of plasticity, i.e. it is not initiated 

 until a certain strain is reached, the plastic yield-point 

 lying near the breaking point of the ice ; the second, 

 that the ease with which deformation may be pro- 

 duced varies with the direction in the crystal ; the 

 third, that the optical properties of a crystal are 

 affected by such deformation, the effect being de- 

 pendent upon the direction in the crystal in which the 

 deformation takes place ; and the fourth that granular 

 ice, composed of interlocking crystals, is subject to 

 deformation equally with a single ice-crystal. Pond- 

 ice was mostly used in the experiments, but granular 

 snow- and glacier-ice were also employed. The 

 authors notice a suggestive fact in regard to the first, 

 that in a cake 30 cm. thick, about 10 cm. at the top 

 consisted of finely granular ice; the next 15 cm. of 

 coarse prismatic crystals of ice, standing perpendicular 

 to the water surface, and the remainder of finely 

 granular ice with diversely oriented crystals. 



THE WINDS IN THE FREE AIR* 

 TT was noticed in very early times that the wind 

 A in the upper air may be very different from what 

 it is on the surface. Lucretius says: "See you not 

 too that clouds from contrary winds pass in contrary 

 directions; the upper in contrary way to the lower." 

 Bacon advocated the use of kites in studying the 

 winds; but it is only in quite recent years that any 

 systematic attempt has been made to investigate the 

 free air above the surface of the earth. Kites have 

 been flown to a height of four miles, but it is a 

 matter of some delicacy to get even so high as two 

 miles. 



The temperature of the free air may be recorded 

 by a meteorograph attached to a small rubber balloon, 

 which continues to ascend until the pressure of the 

 gas inside bursts the envelope, and the instrument 

 descends again to the surface. The beautiful instru- 

 ment constructed by Mr. W. H. Dines, F.R.S., the 

 pioneer of upper air research in this country, is so 

 light that the torn fabric of the balloon is sufficient 

 to act as a parachute and check the speed of descent. 



1 Discourse delivered at the Royal Institution on Friday, April n, by Mr. 

 Charles J. P. Cave. 



NO. 2273, VOL. 91] 



The general result of the observations has been to 

 show that the temperature of the air decreases with 

 height up to a certain point, above which the tem- 

 perature distribution is nearly isothermal ; however 

 much higher the balloon may ascend, there is little 

 further change of temperature. This upper layer, 

 discovered by M. Teisserenc de Bort, whose recent 

 death meteorologists of every country lament, is called 

 the stratosphere; the lower part of the atmosphere is 

 the part that is churned up by ascending and descend- 

 ing convection currents, and is called the troposphere. 

 The height at which the stratosphere is reached, as 

 well as the temperature of the layer, varies from day 

 to day and from place to place. In these latitudes 

 it is met with at heights varying from about 8 to 

 14 km., with temperatures varying from — 40 to 

 -8o° C. 



It is not, however, with temperatures that I am 

 chiefly concerned to-night, but with the wind currents 

 in the different layers of the atmosphere. If one of 

 the balloons carrying instruments, or a smaller pilot- 

 balloon, is observed with a theodolite, its position 

 from minute to minute can be determined, and from 

 its trajectory, or its path, as it ascends, the winds 

 that it encounters can be calculated. 



The theodolite used is constructed specially for the 

 purpose ; a prism in the telescope reflects the light at 

 right angles, so that the observer is always looking in 

 a horizontal direction, even if the balloon is overhead.' 

 It is important that the observer should be in as 

 comfortable a position as possible, for an ascent some- 

 times lasts more than an hour and a half, during which 

 time the observer can only take his eye from the 

 telescope for a few seconds at a time, otherwise he 

 may lose sight of the balloon and be unable to find it 

 again. 



The balloon having been started from one end of the 

 base, observations are taken from both ends at exactly 

 the same times, usually every minute. From the 

 positions of the balloon at each successive minute, 

 which are plotted on a diagram, the run of the balloon 

 during the minute can be measured, and hence the 

 wind velocity during that minute can be obtained. 

 After the wind velocities have been measured off, and 

 the wind directions obtained from the directions of 

 the lines on the diagram, another diagram is con- 

 structed showing the relation of the wind velocity and 

 direction to the height. 



It is not necessary, however, to have two observers 

 if the rate of ascent of the balloon is known ; in such 

 a case, the complete path of the balloon can be cal- 

 culated from the observations of one theodolite. It is 

 not, however, possible to know the rate of ascent with 

 complete accuracy, as up and down currents in the 

 air will affect the normal rate. In practice, especially 

 in clear weather, the method is fairly satisfactory. 

 The method of one theodolite requires less prepara- 

 tion, and the subsequent calculations of the path of 

 the balloon are less laborious, than in the case of 

 observations taken with two theodolites from opposite 

 ends of a base line. 



The best time for observations is towards sunset, so 

 that the balloon reaches its greatest height after 

 the sun has set on the surface of the earth ; at such 

 times the balloon, still illuminated bv the sun, shines 

 like a planet, and on one occasion I should have 

 found it impossible to tell which was the balloon and 

 which was Venus, except for the movement of the 

 balloon. The distances at which balloons may be 

 seen through the telescope of the theodolite are re- 

 markable. A striking instance was when the flash 

 of the sun on the small meteorograph was seen, not 

 once, but repeatedly, when the balloon was about 

 nine miles above the sea and at a horizontal distance 

 of about thirtv miles. 



