3IO 



NA TURE 



[January 14, 1909 



strata is ascertained, this will be precisely that quantity 

 of water which has evaporated from the surface of the 

 land. 

 The only weakness that I have been able to perceive 



in the conclusions 



vhich 



'Waler 



I drew in 1907 regarding this 

 matter is the exact depth which 

 was affected. This I concluded 

 to be 7 feet in the Pusa soil. 

 .\s a matter of fact, the ascer- 

 tainment of this depth with 

 exactitude is not possible. In 

 the marginally shown diagram 

 are three curves. .'\ repre- 

 sents the water immediately 

 after rain ceases (in an ideal 

 physically uniform soil). B re- 

 presents the water as ascer- 

 tained at the conclusion of the 

 dry period, and cuts A at the 

 point X, below which no de- 

 crease of water has been 

 perceptible by the methods 

 employed. Owing, however, to 

 errors] due principally to differ- 

 ence of physical character of 

 soil, this point cannot be determined exactly, and in 

 realitv the curve might be BC, that is, it might cut A at, 

 say, y, and not at x ; but the difference between the 

 ascertained loss and the real loss can only be trifling. It 

 Is perhaps needless to add that this error does not affect 

 the principles involved. It is naturally assumed that the 

 poiiils X and y are above the stratum of soil which is 

 maintained in " a saturated condition by the underground 

 water. J. Walter Leather. 



I'usa, Bengal, November 23, 1908. 



Dr. Leather is no doubt correct in supposing that the 

 upward movement of water through the soil is gradual, 

 and in his further deduction that there must, for a time, 

 be some strata in \vhich the water has not yet begun to 

 move ; but we do not know the velocity at which water 

 travels upwards in the soil, and consequently one cannot 

 sav whether the time during which any particular stratum 

 remains unaffected is to be measured in days or months. 

 Dr. Leather's results do not give the velocity of upward 

 movement, but the difference between the loss and the 

 gain of water at different depths. It is no more possible 

 to calculate the amount of w^ater that has passed through 

 a particular stratum by determining the amounts present 

 at two different times than- it would be to calculate the 

 quantity of heat passing along a rod of unknown thermal 

 properties bv measuring the temperature change at a par- 

 ticular point. 



The great value of Dr. Leather's results lies in the fact 

 that thev are the most complete set of moisture determina- 

 tions yet made under conditions of drought. If only .some 

 physicist could be induced to turn his attention to soil 

 problems he would find these da\a very useful. 



E. J. RissEi.i.. 



The Rothamsted Laboratories, Harpenden, Herts. 



The Correlation of Teaching. 



The valuable summary of Prof. Perry's address to the 

 " correlation " conference given in Nature of December 

 3, igo8 (p. 143). contains the following statement : — " If 

 a boy wrote a description of anything he had done in a 

 l.iboratory or elsewhere, it should be an exercise in 

 English." This is, unfortunately, accepted by educa- 

 tionists at the present time. Can Prof. Perry not aid in 

 breaking down this barrier to progress rather than in fix- 

 ing its joints more firmly? He has done so much, cannot 

 he do more? 



1 would suggest, whether in class-room exercises or in 

 examinations, that the boy's resultant essay should be 

 examined and corrected by the examiner without reference 

 to spelling, writing, grammar, &c. The object should 

 surely be to put on paper what has been learnt about the 

 subject in hand. Spelling being the result of the use 

 of accurate vision, writing being the result of ohysical 

 and nervous stability, need in no circumstances affect the 



value of a scientific production. .As the boy gets older 

 and as his view of life extends, the value of his produc- 

 tion, from the point of view of English, will gradually 

 improve. 



Besides, it is nothing short of an absurdity to look at 

 certain exercise books, where the red-ink corrections, that 

 have taken so much of the teacher's time, are chiefly 

 connected with something that in no way affects the value 

 of the exercise itself. Charlie Woods. 



December 31, 1908. 



I AGREE with Mr. \\'oods in his condemnation 01 a 

 system in which every exercise is treated as if grammar 

 and composition were as important as the subject-matter ; 

 but surely he is going too far when he says that in 

 writing a description of what he has done or seen, a boy 

 need not fear that his grammar or spelling or composition 

 will be criticised at all. I ask the science master, in 

 teaching science, to teach also mathematics and English 

 and sketching incidentally, and to take some pleasure in 

 doing it. If he insists that these subjects are the absolute 

 preserves of the mathematical, the English, and the 

 drawing masters, he must not be astonished when a 

 classical master openly expresses pride in an ignorance of 

 " stinks." 



I do not think that the system which Mr. ^^'oods con- 

 demns is very much in vogue, whereas the system of which 

 he approves has done an immense amount of harm. We 

 all of us know men holding the highest science degrees 

 whose spelling, grammar, and composition are beneath 

 the contempt of a board-school boy. In nine cases out of 

 ten it will be found that when a student cannot give a 

 clear account of what he has seen or done, he has no 

 clear ideas about the matter, and every examiner knows 

 that it is only the very exceptional man who has clear 

 knowledge yet cannot express himself clearly. If the 

 examiner has a keen sense of justice this candidate gives 

 him more trouble llian any five others. John Perrv. 



An Electromagnetic Problem. 



The problem noticed by Mr. Comstock in Naturk of 

 November 19, 1908, is an interesting one, but I do not 

 see how the " laws of electricity and conservation of 

 energy require in themselves the discrete structure of 

 electricity or the association of electricity with matter." 

 The electromagnetic field produced by a uniform spherical 

 sheet of electricity, unassociated with matter expanding 

 under its own repulsion, is not zero, but indeterminate. 

 The total energy of the system remains finite and constant, 

 while the velocity of expansion is that of light. Thus 

 perfect uniformity of electricity, together with isolation, is 

 not incompatible with the laws of electricity and con- 

 servation of energy. The indetcrminatencss of the electro- 

 magnetic field will, of course, surprise no one who is 

 willing to start with a distribution of electricity differing 

 infinitely little from that of perfect uniformity, arranged 

 as a .sheet differing infinitely little from spherical, and 

 expanding in surroundings departing infinitely little from 

 the symmetrical. A. Core. 



Mr. Core's objection would apply in many problems 

 where certain functions appear to vanish because of 

 symmetry, but in the present case I think it does not apply. 



In physical problems it is well to avoid mathematical 

 " sheets " except in unusually simple circumstances, and 

 in the present case the spherical shell of electricity which 

 is expanding under the mutual repulsion of its parts is to 

 be considered of finite tliickness and of constant volume 

 density of electricity. 



In these circumstances the displacement current is 

 evidently .equal and opposite to the convection current when 

 the sphere is expanding, and hence the curl of the mag- 

 netic force is zero everywhere. This requires the magnetic 

 force to be zero everywhere, since such a vector vanishing 

 at infinity and having its curl and divergence both zero 

 must itself vanish. It is not then immediately evident 

 what becomes of the electrical energy lost on expansion. 



D. F. Comstock. 



Institute of Technology, Boston, December 17, iqoS. 



NO. 2046, VOL. 79] 



