24 



NATURE 



[March 8, 191 7 



suggested : " The bacterium can live on my roots 

 and supply me with nitrogen, and I furnish him 

 with phosphoric acid and potash." ' Happily such 

 passages, which are neither good science nor 

 eood fiction, are rare in the handy little volume. 



G. H. C. 



LETTERS TO THE EDITOR. 



[The Editor does 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 anonymous communications.^ 



The Horizontal Temperature Gradient and the Increase 

 of Wind with Height. 



It has been known for some time past that from 

 heights of about i to 9 km. the temperature is higher 

 in the high-pressure than in the low-pressure area, 

 and also that in general the wind, and especially the 

 west wind, increases with height. The following 

 simple proof shows that these two observational facts 

 are not independent of each other, but that, one being 

 given, the other follows as a logical consequence. 



A wind in the northern hemisphere exerts an 

 acceleration towards its right-hand side equal to 

 2a)Vsin0, where 10 is the earth's rotational velocity, 

 V the velocity of the wind, and ^ the latitude. Also, if 

 the path of the air particles is curved, there is a 

 further acceleration equal to v- (r, where r is the radius 



B 



LOW 



HIGH 



D 



of curvature, and the acceleration is away from the 

 centre of curvature. The total acceleration to the 

 right is 2iav sin ^ ±v- 1 r , and the sign of the term 

 in\Hjlving v- is positive in regions where the isobars 

 are concave to the low pressure. However, in these 

 latitudes the v^ term is not as a rule important, but 

 appears as a correction, generally positive, to the 

 term 210V sin (f>. 



Let ABCD be a vertical section at right angles to 

 the gradient wind, AB and CD being sections of the 

 isobaric surfaces, and AC and BD vertical straight 

 lines. If V be the gradient wind — i.e. the wind at 

 right angles to the paper— then the tangent of the 

 slope of AB is 2(oZ'sin ^-i-i'-/r :^, for 2u>v sm f -{■ v^ I r 

 is the horizontal acceleration and g the vertical. Simi- 

 larly, the slopeof CD is 2(uV sin + V^/r :^. If, then, 

 V is greater than V, BD must be greater than AC. 

 Now the pressure difference between A and C is equal 

 to the pressure difference between B and D, since .AB 

 and CD are isobaric lines ; and since the corresponding 

 elements in the two air columns AC and BD are of 

 equal pressure, and the dcnsitv in BD less, the tem- 

 perature in BD must be higher'than that in AC. That 

 is, if -y be greater than V, then t is greater thin T. 



Thus where the wind is increasing with height with- 

 out much change in direction, anyone with his back 

 to the wind will, if he follows an isobaric surface from 



NO. 2471, VOL. 99] 



left to right— that is, from cyclone to anticyclone — 

 find an increasing temperature. 



If we neglect the curvature it is easy to calculate 

 the numerical values. An increase of 1 metre per 

 second over a horizontal range of 100 km. in latitude 

 53° makes BD-AC=i-28 m. ; therefore, taking AC as 



1 km. and CD as 100 km., an increase of i m./s. per 

 km. height makes t-T = oooi28t, or, giving to t a 

 mean value of 250a, f-T = o-32°. Thus an increase 

 of I m./s. per km. gives in these latitudes an ap- 

 proximate rise of 1° C. per 300 km. along an isobaric 

 line at right angles to the wind. This is fully in 

 accordance with such observations as are available. 



As a corollary it follows that the strongest winds 

 have a high temperature on their right-hand side 

 below their own level and a low temperature above, 

 while on their left the converse holds, and it is cold 

 above and warm below^ cold and warm being used 

 relatively to the mean for the height. Since the 

 strongest winds are foUnd near the upper Hmit of the 

 troposphere in regions where the barometric surface 

 gradient is steep, this again agrees with the usual 

 distribution of temperature in cyclones and anti- 

 cyclones. 



The special tendency of west winds rather than east 

 to increase with height agrees with the natural rise 

 of temperature in the lower strata from north to south. 



W. H. Dines. 



Benson, February 23. 



Ten Per Cent. Agar-agar Jelly. 



It may be of use to put on record a method of 

 making a jelly containing ten or more parts by weight 

 of agar-agar to 100 parts by volume of solvent. _ 



Agar-agar powder is apt to form lumps when mixed 

 with water or with a mixture of water and glycerine. 

 If this difficulty is obviated by vigorous stirring 

 bubbles are formed. In the case of jellies of i-^ or 



2 per cent, strength this does not matter, as the 

 bubbles readilv come to the surface. With thicker 

 jellies this is 'not the case. These difficulties are 

 avoided by the following procedure. 



Powdered agar-agar is washed with ether, dried, and 

 passed through a sieve. This treatment removes a 

 fatty acid. 



Twenty grams of the purified agar-agar are placed 

 in a round-bottomed flask. The flask is provided 

 with a cork having two holes. Through one of the 

 holes passes a tube leading to a vacuum pump. The 

 other tube accommodates the stem of a separating 

 funnel. The air is exhausted, 140 c.c. of glycerine 

 are placed in the funnel and rapidly run into the flask. 

 The flask is shaken for a few seconds, by which time 

 the agar-agar powder will be found to be completely 

 and uniformly suspended in the glycerine. Sixty c.c. 

 of water that has previously been boiled and com- 

 pletelv cooled are now placed in the funnel, run into 

 the flask, and mixed w'th its contents by a few- 

 seconds' shaking. Air is allowed to enter the flask, 

 and the mixture is at once run out into a series of 

 glass svringes from which the pistons have been re- 

 moved and the nozzles of which are closed with rubber 

 caps. Each syrinere is filled about two-thirds full of 

 the mixture, the pistons are replaced, and the syringes 

 are then heated in a water-bath. The jelly is now 

 ready for use. Cans for the nozzles may be made by 

 boring a hole nearly, but not quite, through a rubber 

 cork. A bent strip of tin is required for each syringe 

 to hold the cap in place. The jelly when melted is 

 too stiff to pour out of a test-tube. It can be readily 

 squirted from the nozzle of the svringe. 



If the proportion of glycerine is increased the jelly 

 is weaker, but more transparent. With less glycerine 



