August 4, 1892] 



NA TURE 



OJD 



and active. The engineering use of the term ohm is mainly in 

 connection with insulation and other high resistaaces ; for large 

 conductors the equivalent "volt per ampere" is perhaps more 

 often used. It is the drop of potential which a given conductor 

 entails for a given current that is of real interest to an engineer, 

 and it is this of which in large leads he consciously thinks. 



A 6 ohm conductor means one that drops 6 volts for every 

 ampere that is sent along it. If you send 3 amperes along 

 such a line, the potential at the far end is 18 volts below that 

 at the near end. The clear realization of this fact would be al- 

 most aided by the complete title, 6 volts per ampere, instead of 

 the abbreviation, 6 ohms. Nevertheless, the name ohm is in 

 common use and hence may be assume I useful. 



A still more useful name, however, for good conductors would 

 really be the reciprocal of an ohm — the ampere per volt. Su -- 

 pose this called a mo, as Sir W, Thomson once suggested, then 

 a cable of 20 mos would be one that conveyed 20 amperes with 

 a drop of i volt. A thousand-mo cable would convey 500 

 amperes with a drop of half a volt, and so on. It is more 

 directly practical to think of the amperes conveyed per drop of 

 voltage, than of the drop of voltage per ampere. I believe that 

 some authorized name for unit conductance would be welcomed. 



Units of Inconvenient Size. 



The authorized name "coulomb" for unit quantity is barely 

 used by engineers, who are content with ampere-hour ; thus 

 proving that what is needed in practical units is not so much a 

 consistent decimal system, as a set of units each of practicable 

 magnitude. 



Farad. 



The effort after consistency has resulted in the useless " Fa- 

 rad " ; and this should be a lesson not to try and fix units of 

 unreasonable size. The c.g.s. units already exist as a consistent 

 system ; the only objection to them is that they are of unpracti- 

 cal size. The whole object of devising a practical system of 

 units was to have things of every-day size to deal with. The 

 volt, the ampere, and the ohm satisfy this condition. The 

 coulomb, the farad, and the watt do not. Already they have 

 practically given place to the ampere-hour, the micro-farad, and 

 the kilowatt. 



Considerably more progress would have been made in know- 

 ledge of ordinary capacities if the microfarad had been called 

 the farad, so that easy submultiples of it would have been 

 available to express the capacity of Leyden jars, and such 

 like things. The capacity of an ordinary jar would then have 

 been a few millifarads, and a microfarad would have been the 

 capacity of a short bit of connecting wire. I ask whether this 

 change would introduce serious confusion even now. I think 

 not. Nobody cares the least about "coulombs per volt," and so 

 there is no sense or use in the present farad. Telegraphists 

 would surely soon consent to drop the useless prefix micro ; and 

 the factor of a million is too great. to render doubt possible as to 

 what was intended, even in the transition stage. It ought to be 

 regarded as essential to have the practical unit somewhere not 

 hopelessly away from the middle of the range of probable mul- 

 tiples and submultiples. 



Coulomb. 



A coulomb again is almost useless as a synonym for the ampere- 

 second ; it is so easy to speak of ampere-minutes or ampere-hours. 

 If the name coulomb could be set free from its present superfluous 

 meaning it could usefully be applied to the electrostatic unit of 

 quantity, which wants a name. Teachers would find it con- 

 venient at once, and in the apparently imminent line of develop- 

 ment engineers might find it useful before long. It is the charge 

 on a two centimetre sphere at a potential 300 volts (or on a 

 one-foot sphere at 20 volts). The capacity of the two-centi- 

 metre sphere would be /^ of a (new) microfarad. 



Watt. 



Lastly with regard to the watt. The name volt-ampere is 

 almost as good as the name watt, especially since the watt is also 

 one joule per second. 



Both names, watt and joule, are not really wanted by elec- 

 tricians, to whom their coexistence is rather confusing. I believe 

 it would be more convenient to use the term watt in the sense it 

 gets so frequently used now, viz., energy, say a volt-ampere- 

 hour ; in which case a kilowatt would be synonymous with the 

 present Board of Trade unit. 



NO. II 88, VOL. 46] 



The rate of working, or power, could then be expressed in a 

 rational and unforced way as so many watts per hour or so many 

 volt-amperes. It is much more natural to give a name to a 

 definite thing like a quantity of energy, than it is to give it to a 

 mere rate of working. The latter is instinctively felt to need a 

 reference to time ; just as a velocity unit has not been practically 

 found to need a name, being quite simply expressible in feet per 

 second or miles per hour; and even when a name has been 

 given, like " knot," instinct constrains people to practically get 

 rid of it again by speaking of knots per hour, just as we find 

 "kilowatts per hour" already of en used in electrical work- 

 shops. I suggest, therefore, that the present watt is too small, 

 that it is sufficiently expressed by a joule per second, and that it 

 would be more useful if magnified 3,600 times, and turned into 

 a unit of energy. 



That we should thus have several energy units — the erg, the 

 joule, and the wait, all of quite different sizes, is no objection, 

 but an advantage, seeing the extreme importance of energy. 

 Such things as length, ma-s, time, and energy demand a fair 

 range of units. It would be tedious to express centuries in 

 seconds. 



(2) Mag."?etic Circuit. 



In speaking of the magnetic circuit I wish to refer back to my 

 opening remarks concerning the electric circuit, and the class of 

 things for which names should be found. In the magnetic 

 circuit the only thing at present seriously attempted to be 

 named is, in accordance with the historic parallel of the ohm, a 

 coefficient or characteristic of a coil of wire — its coefficient of 

 self-induction ; the unit of which has been called variously n 

 secohm, a quadrant, and a henry. 



Total Indue /ion. 



But the real active thing with which engineers are concerned 

 is total magnetic induction, total number of lines of force acro'>s 

 an airgap : as between the polepieces or through the armature 

 of a dynamo, or in the circuit of a transformer. It may be 

 called the electromagnetic momentum per turn of wire ; or the 

 surface integral of B. This total induction is in some respects 

 analogous to electric current, and has occasionally been called 

 magnetic current (a bad name), or "magnetic flux." It is, 

 however, more strictly analogous to the coulomb, and its time 

 rate of variation is the more proper representative of electric 

 current. 



Its common practical name at present is " total lines," or 

 "total induction," or "number of lines." 



Now " one line " is awkward as a unit, besides being (if a 

 c.g.s. line) inconveniently small. The earth, for instance, 

 sends 4,400 such lines through every horizontal square metre 

 about England ; through a square inch it only sends a frac- 

 tion of a line. A practically sized unit of induction badly 

 wants a name, and " henry" would have done for it very well, 

 and have been both more suitable and more useful for the 

 actual quantity than for a coefficient. But "henry" has 

 already been half appropriated to the secohm, so, for illustrative 

 purposes at any rate, I propose to use the name "weber" for 

 the unit magnetic flux. 



Concerning the most convenient size for the weber, there is 

 much to be said for making it 10^ c.g.s. lines, though that is 

 bigger than ordinarily occurs in practice ; because then a wire 

 which cuts one weber per second will have a volt difference of 

 potential between its ends. Or a coil of twenty turns through 

 which the magnetic induction changes at the rate of one weber 

 per second will have an E.M.F. of twenty volts induced in it. 

 The average E.M.F, in such a coil, spinning thirty turns a 

 second, and enclosing a maximum total-induction of one weber, 

 is 600 volts. 



This is the dynamo use of the unit ; the following is the 

 motor use. 



A wire carrying an ampere and cutting a weber per second, 

 does work at unit rate, viz., one joule per second. 



Probably the simplicity of all this compensates for the rather 

 unwieldy size of the unit. A strongly magnetized piece of iron 

 may have 20,000 lines to the square centimetre ; so a weber 

 could occur across a narrow airgap half a square metre in area. 



The earth gives an induction of about one weber through 

 every 23,000 square metres of England, or 100 webers per square 

 mile. The earth induction through a horizontal square metre is 

 44 micro- webers, so with micro- and milli-webers the range would 



