April 19, 1877] 



NATURE 



529 



ground, the vegetable covering of the soil, and openness of situa- 

 tion- Alexander Buchan 

 Edinburgh, April 9 



There is one consideration which your correspondents, the 

 secretary of the Scottish, and the president of the British Meteo- 

 rological Societies have equally overlooked, and which may 

 seriously affect the conclusions at which they arrive as to the 

 suitability of Greenwich for a first-class meteorological station. 

 Since the year 1846 the temperature observat-ons at Greenwich 

 liave been made under conditions of exposure of thermometers 

 which, whatever their merits or demerits, are not those usually 

 adopted. In a paper published in the Quarterly Journal 0/ the 

 Meteorological Society (October, 1873) I have shown from the 

 average of five years' daily ccmpaiisors that the effect of the 

 method adopted at Greenwich upon the mean annual tempera'ure 

 is to obtain a result o°"475 warmer than is obtained by the usual 

 method. This quantity is almost identical with the excess which 

 Mr. Eaton attributes to the local consumption of fuel, an explana- 

 tion surely most inadequate. Thus the discrepancy pointed out 

 by Mr. Eaton only serves to establish his opponent's case. Mr. 

 Buchan on the other hand must be unaware that it is the eye 

 observations alone, made from the revolving stand, that are relied 

 upon for temperature results at Greenwich, and though his con- 

 clusion would seem to be correct it does not seem possible to 

 sustain the argument by which he has arrived at it. 



Orwell Dene, Nacton John I. Plummer 



Cast-iron 



I HAVE been stiuck by the statement I found in several books 

 on physics thst cast-iron expands when it gets cool. As some of 

 these books are used as text-books in schools in this country, 

 and as this statement is contrary to the experience of all practical 

 men with whom I have conversed upon the subject, I think the 

 following translation of an article which appeared in Der Civil 

 Ingenieur, edited by K. R. Bornemann, in Freiberg, 1863, 

 ix. Band, iv. Heft, p. 219, may not be uninteresting, explaining, 

 as it seems to me fully, at least one of the facts on which the 

 statement mentioned above appears to be based, viz., the fact 

 that cold ircn swims in liquid iron : — H. M, 



" At a meeting of the Association of Saxon Engineers which 

 took place in Freiberg in August last year (1862) Mr. H. Giiison, 

 of Buckau, near Magdeburg, called the attention of the members 

 to a phenomenon which had frequently been observed by him, 

 but of which no proper explanation could be given at the time, 

 viz., that pieces of cold cast-iron swim perfectly in molten iron. 

 The question was, raised, to what causes this may be due, and as 

 from the physi.;al point of view it was thought an interesting one, 

 it was suggested that experiments should be made in order to 

 obtain a proper explanation thereof. 



"In consequence of this, M. Centner, Inspector of the Jacoli 

 Iron-works, near Meisen, made such experiments, the result of 

 which he communicated to the Association at a meeting on 

 May 17, 1863. The following is an extract of Mr. Centner's 

 report :— (Signed) W. Tauberth " 



"Before anwering this question I made some experiments in 

 order to ascertain whether this swimming is not caused by the 

 specific weight of the body ; by there I found confirmed that 

 cold cast-iron weighs ^V niore than an equal volume of molten 

 iron, for if a piece of cold cast-iron of 28 lbs. be used to form a 

 mould, and if this mould be filled with molten iron, the new 

 piece of metal thus obtained will only weigh 27 lbs. This weight, 

 of course, is also that of the liquid metal which was required to 

 fill the mould, formed from a piece weighing 28 lbs. 



"For this reason, in making moulds for cast-iron, a measure 

 is used which is -jV longer than the ordinary measure, if the 

 piece of iron to be formed is to have the full size of the ordinary 

 measure. 



"Repeated experimetits with a mass of molten iron of 

 2,000 lbs. gave me further proofs that the causes of this 

 swimming must be other than the specific weight. 



" For my experiments I used four bodies of cast-iron of diffe- 

 rent shapes, but of the same volume, viz. a plate of 6" inches in 

 the square and 1" thick, a cube of 3^", a cylinder of 4" diameter, 

 and 3" height, and a ball of 4" diameter. Each of these four 

 boti ies measured 36" cube and weighed 7 lbs. 



" If the cause of the swimming were the specific weight, an 



enual narl- r-f tlif> vnlnmf nf all th«»sp hr>dip<; onfrhl- to rpmain 



above the surface of the liquid iron, but such is not the case at 

 all. The volume above the surface of the liquid iron is different 

 in each of the four bodies ; it is greatest with the plate and 

 smallest with the ball. Thus it is dependent on the shape and 

 position of the surface which rests on the liquid iron, 



" In order to come to the real causes of this swimming, I 

 must first remind the reader that in every hot liqu'd in an open 

 vessel, in consequence of the more rapid cooling at the surface, a 

 continuous current is originated, the interior hottest parts a=-cend- 

 ing and the exterior colder ones descending ; and thus a more or 

 less visible movement or agitation is produced in the mass. 

 Such currents occur in every mass of molten iron, and are there 

 especially remarkable in consequence of a contemporaneous 

 ascension and separation of slags, which, when they have 

 arrived at the surface, are generally pushed towards the edges, 



" If a solid piece of iron be put on the liquid mass the former 

 gets at once heated at the expense of the latter ; the portions of 

 water and of air which are contained in the solid piece get ex- 

 panded and expelled with considerable force, thus forming a 

 current in opposition to the ascending one above mentioned. 



" This expulsion of air and water may even cause dangerous 

 explosions, it the usual precaution is neglected to warm the solid 

 piece somewhat, before it is brought into contact with the liquid 

 mass. 



" Now there is ro doubt that these opposite forces alone are 

 able to raise the heavier sdid piece more or less according to the 

 more or less favourable surface it presents. 



" But besides this there is to be taken into consideration that 

 the overweight of the solid piece of iron is diminished by the 

 previous heating which when the solid piece comes into contact 

 with the liquid, is at once augmented and that the proportion of 

 heat of the molten iron to that of the solid piece must to some 

 extent have an influence on the more or less deep immersion of 

 the latter. 



" A further cause, although a slight one, of this swimming of the 

 solid piece is the cohesion of the liquid iron ; but at any rate this 

 becomes of some importance in conjunction with the above-men- 

 tioned continuous ascending of slags which collect under the 

 swimming body and retain partly the air expelled by the latter, 

 helping in this way to keep it afloat 



" Solid cast-iron being ^j^ heavier than an equal volume ot 

 molten iron, the overweight of each of the four bodies used for 

 my experiments is consequently only j lb., and in the present 

 case it is therefore only this one quarter of a pound which the 

 above-mentioned opposite forces have to lift within areas of 12" 

 to 36" square in order to keep the body swimming, 



" On increasing the size of the solid bodies, however, it will 

 be easily understood that a limit to this swimming will soon be 

 reached, and indeed I accidentally found this limit on my first 

 trial in quadrupling the sizes of the four bodies so that each of 

 them weighed 28 lbs, ; for all the bodies with the exception of 

 the plate which was 12" in the square and i" thick went to the 

 bottom. The plat« on being put gently on the liquid mass was 

 just kept afloat, but its surface was a little below the surface of 

 the molten iron, i lb. overweight therefore with a surface of 

 12" in the square could scarcely be kept swimming. 



" The behaviour of the other three bodies at the bottom was 

 remarkable in consequence of a continuous vehement ebullition 

 accompanied by the shooting out of long white brilliant flames, 

 and these phenomena can only be attributed to the water and 

 air expelled by the heat. 



"These experiments with bodies of 28 lbs. weight show therefore 

 that above thisweight, without giving to the bodya mere favourable 

 surface than 12" in the square, these bodies do not what is prc- 

 perly called swim in the sense that part of the solid body is kept 

 above the level of the liquid. For if the bodies, for instance the 

 ball, the cube, and the cylinder on being moved, rise and fall a 

 little alternately, this can no longer be called swimming, for it is 

 just the transition from oveiweiglit to equilibnum. 



" That the greater or lesser degree of density of the different 

 sorts of iron will also e-xercise an influence cannot be doubted, 



* * Less fortunate but still interesting was an experiment which 

 I made with four pieces of zinc of the same shapes as the pieces 

 of iron on 200 lbs. of liquid zinc, when with the pieces of 7 lbs. 

 each the same thing took place as with the cast-iron pieces of 

 28 lbs. each on liquid iron, viz., the plate was just kept afloat 

 and the three other bodies went to the bottom. 



" With zinc, therefore, this phenomenon of swimming does not 

 occur with such heavy bodies as with iron, and this may be 

 explained by the fact that with zinc, in consequence of the much 



smaller differfnw of tpmr>*»ratiirf> Vwtwppn f-h<» linnirl r,r\A flii. 



