596 



NA TURE 



{April 2 2, i( 



was composed of thirty-five plots of one chain square, or of one- 

 tenth of an acre each, and comprised forty drills or roi\s of 

 plants. The plots were arranged to form five rows of seven 

 plots each, and a rectangle of three and a half acres. They 

 were so disposed within this area that every manured plot was 

 adjacent to an unmanured plot, with the object of obtaining 

 repeated confirmations of any differences which might be indi- 

 cated in favour of the dressings. Every trial was made in 

 duplicate in both series, and the results are graphically shown 

 by tinted plans, on which the number of the plot, the manurial 

 dressing used, and the number and weight of ro3ts grown are 

 printed. 



Not only do these experiments prove ground basic cinder to 

 be a valuable fertiliser, but they assign it a higher position than 

 ground coprolite, and place it only slightly below "super- 

 phosphate " in value. This remarkable result is, we find, sup- 

 ported by statements resting on the authority of Dr. Biedermann 

 Cttitralhlatt (vol. xiv. part 2), in which the phosphoric acifl in 

 basic cinder is asserted to be more readily appropriated liy 

 growing plants than is the phosphoric acid contained in copro- 

 lites. 



The subject is full of interest as bearing upon the positive 

 profits of steel manufacture, and also upon the manufacture of 

 superphosphate and upon agricultural practice. It has attracted 

 the attention of M. Grandeau, of the Faculty of Science of liie 

 French University, who contributed a review of the pamphlet 

 before us occupying over three columns of Le Tempi newspaper. 

 In the course of his remarks he says : — " Les resultats obtenus 

 en Angleterre confirment pleinement, on le voit, ceux que les 

 agronome> allemands ont publies et que j'ai precedemment 

 analyses. Les scories de dephosphoration sont appelees a jouer 

 un role considerable dans la fumure du sol. Des negociants 

 ont deja traite avec quelques-unes des importantes usines de 

 I'Est (Alsace-Lorraine notamment) pour I'achat de loute leur 

 production de scories." 



The experiments conducted a* Downton and Ferry Hill were 

 not only instituted to ascertain the positive value of ground basic 

 cinder in comparison with unmanured plots. In them the 

 ground cinder was compared with ground coprolite, with ordin- 

 ary superphosphate, with a rich superphosphate, with a super- 

 phosphate made direct from the basic slag, and with a super- 

 phosphate to which green vitriol was added. The subject is 

 likely to arouse a very considerable amount of attention. 



AN IMPRO VED FORM OF TEMPERA TURE 

 REGULATOR 



T N 1882, at the request of the Board of Trade, the Royal 

 Society appointed a Committee, consisting of Sir G. Airy, 

 Col. Clark, and Prof. Stokes, to advise on the question of im- 

 proving the existing means of the comparison of standards of 

 length at the Standard Office. In their report the Committee 

 suggested improvements in the arrangement for securing greater 

 uniformity and steadiness of temperature in the bars under com- 

 parison. As a first step in this direction, the Cambridge Scien- 

 tific Instrument Company were requested to investigate the 

 subject of temperature regulators, and to consider the general 

 design of a comparing apparatus. They proposed that the 

 standards, some hours before they were observed, should be 

 placed in metal cases and immersed in a bath containing water ; 

 and that they should not be disturbed during the process of 

 observation. If the uniformity of temperature of the water 

 could be ensured, it would secure the equality of temperature of 

 the standards and remove one of the greatest difficulties in the 

 construction of a satisfactory comparing apparatus. 



After some preliminary trials a regulator was constructed 

 similar to that used at the Bureau International des Poids et 

 Mesures at Sevres.^ Its action depends on the variation of 

 pressure of a saturated vapour, caused by a change of tempera- 

 ture. The pressure on the volatile liquid and vapour is due to 

 the atmosphere, as well as to a column of mercury ; consequently 

 the regulated temperature will vary with any change in the 

 atmospheric pressure. In the following experiments the corre- 

 sponding change of temperature for one inch alteration of the 

 barometer was about o°'37 C. The accuracy of the regulator 

 therefore dejiends on the constancy of the atmospheric pressure, 



' See "Travaux et Miimoires du Bureau International des Poids et 



and to overcome this serious disadvantage an arrangement was 

 devised for maintaining a constant pressure on the volatile 

 liquid. 



An iron bath, D (Fig. i), containing water, was placed in a 

 wooden box. The intermediate space was filled with sawdust ; 

 this was done to minimise the unequal cooling due to the vary- 

 ing temperature of the room. The two ends of a U "Shaped 

 tube were fixed into the bottom of the iron bath and passed 

 through holes in the side of the wooden b ix. The water was 

 kept warm by means of two gas-flames placed under a part of 

 this tube. One of the gas-flames was connected to the regulator 

 and the other direct to the gas-main. The object of the second 

 flame was to re-light the regulated gas-jet in case it should have 

 been extinguished by the regulator. 



The water in the bath was kept thoroughly stirred by air 

 forced through it by means of bellows. These were placed 

 inside the box in order to keep them warm, and also for the 

 more important reason of avoiding the currents of air which 

 would otheruise be produced at each stroke. The air used for 



F.ol 



stirring was thus saturated with aqueous vapour, and did not 

 cool the water in the bath by absorbing moisture from it. 

 The nozzle of the bellows was connected to the (J 'Shaped tube 

 by a branch inserted just above the point where the gas-flame 

 was applied. The air thus pumped through the upper part of 

 the U'ti'be caused a rapid circulation of water through it. This 

 method has the advantage of applying the heat in a manner 

 which does not tend to make the water in one part of the bath 

 perceptibly hotter than the rest. 



The volatile liquid in the regulator was a mixture of 

 methyl-chloride and ethyl-chloride, boiling at about 2i° C. 

 under the normal atmospheric pressure. It was contained 

 in a flat bulb, G, blown at the end of a glass tube, and 

 was under a head of mercury. The glass lubes containing 

 the mercury were connected by short lengths of canvas- 

 lined india-rubber tube. A double brass tube was secured to 

 the open end of the regulator, E ; this is shown enlarged in 



