594 



NA TURE 



[April 20, 189; 



bamic acid is present in the urine of a normal animal, and 

 increases after Eck's operation. It would be interesting there- 

 fore to compare these facts with what we know of the increase 

 of ammonia in pathological states of the liver in man. The 

 liver, however, is not the only place where urea is formed, for 

 the urea never completely disappeared in any of these experi- 

 ments ; and it is well known that in sharks which live seventy 

 hours after the removal of the liver, the urea in the muscles does 

 not diminish after the operation. 



Such are the chief new facts we have met with in this in- 

 teresting memoir, and it is certain that these investigations open 

 up a new field for further researches. The other papers ^ con- 

 tained in this volume call for little comment ; they relate chiefly 

 to the digestive and putrefactive processes taking place in the 

 human intestinal tract. 



It will be seen, however, that this fourth number sustains 

 the well-earned reputation of the three first ones, and that 

 the archives deserve to take their place among the chief scien- 

 tific journals which made their first appearance in the year 



STEAM ENGINE TRIALS. 



A PAPER on the last series of steam engine trials undertaken 

 ^^ by the late Mr. P. VV. Willans was read at the meeting of 

 the Institution of Civil Engineers on April ii. 



The paper dealt with an extensive series of condensing 

 trials made with a 40 I.H.P. Willans Central- Valve Engine. 

 These were intended to form a sequel to the investigations 

 described in the author's papers, entitled " Economy Trials of 

 a Non-condensing Steam Engine, Simple, Compound, and 

 Triple," read before the Institution in 1888 and 1889. The 

 principal objects in undertaking these trials were — (i) To 

 ascertain the initial condensation in the first cylinder, and to 

 trace the behaviour of the steam in the succeeding cylinders, 

 when working as a compound or triple-expansion engine ; (2) 

 To observe the effect of speed of rotation, area of exposed sur- 

 face, and range of temperature, upon the initial condensation, 

 and upon economy generally ; {3) To ascertain the percentage 

 of the theoretical mean pressure actually obtained ; (4) To 

 ascertain the ratio of the work done by each pound of steam to 

 the theoretical work due from it ; (5) To determine the con- 

 sumption of steam at all loads, and under various conditions. 



The consumption of steam was determined by discharging the 

 condensed water from the exhaust into a tank carried by a weigh- 

 bridge, and observing the intervals of time required for fixed 

 weights of water to run in. By this method, a continual watch 

 was kept on the performance of the engine during the whole 

 trial, and any disturbing cause was immediately detected ; leaky 

 steam-pipe joints did not affect the result, and the length of the 

 trial might be much reduced. Special experiments, made to 

 ascertain whether any addition was necessary to cover leakage 

 in the engine and exhau.-t-pipe, showed that this leakage was 

 slight. 



The method of determining the theoretical work due from 

 one pound of saturated steam when discharging into a condenser 

 was next considered, and it was shown that the thermal 

 eflSciency of a condensing engine must of necessity be less than 

 that of a non-condensing engine, owing to the greater pro- 

 portionate size of the "toe " of the diagram cut off for practical 

 reasons. In the non-condensing trials the best number of 

 expansions was computed from the approximate formula 

 p* v'' = constant ; but for the condensing trials the error in this 

 could not be neglected. The best ratio of expansion and mean 

 pressure were therefore calculated for adiabatic expansion, by 

 Mr. Macfarlane Gray's d <p diagram, combined with a volume 

 curve. Altogether sixty-two trials were made under various 

 conditions of speed, steam-pressure, load, and ratio of ex- 

 pansion, as well as with the engine working simple, compound, 

 and triple, and the results were embodied in ihq tables accom- 

 panying the paper. 



One of the principal deductions from these experiments was 

 the "straight-line" law of steam-consumption; and it was 

 shown by diagrams that the total water for the horse-power 



I "On the Putrefactive Processes in the Large Intestine of Man and on 

 the Microbes Causing Them," by iM. Lumft. "On the Micro-organisms 

 in the Organs of Choleraic Patients," by M. L. de Rekowski. "Contri- 

 butions to the Study of Chemical Processes in the Intestines of Man," by 

 M. Jakowski. 



corresponding to any mean pressure P, was W -i- KP., where 

 W was the water which would be used by the engine at 

 zero mean pressure (through initial condensation, radiation and 

 conduction), supposing it were frictionless, and K was the water 

 per hour required to produce each pound of mean pressure. 

 These factors were shown to vary with the conditions under 

 which the engine was working. 



Eighteen of the trials were planned to assist in determining 

 the law connecting initial condensation with revolutions ; and it 

 was found that in the high-pressure cylinder at high mean 

 pressures the total condensation per unit of time was directly 

 proportional to the square root of the number of revolutions 

 per unit of time. As the mean pressure was diminished, the 

 condensation became more and more nearly constant at all 

 speeds ; and finally, at low mean pressures, the law appeared 

 to be reversed. For the low-pressure cylinder, the law was 

 modified. 



The important question of the changing proportions of steam 

 and water present during the expansive part of the stroke was 

 investigated by the Q <p diagram. The matter was first examined 

 theoretically by considering the effect of a thin liner of infinitely 

 conducting matter, and a curve was drawn on the (p diagram 

 showing the rate at which the steam initially condensed in 

 warming up the liner from the exhaust to the initial tempera- 

 ature was re-evaporated as the expansion proceeded. The 

 actual re-evaporation, as obtained by measurement of the 

 indicator cards was compared with this theoretical re-evapora- 

 tion, the difference measuring th^ delay in the return of the 

 heat from the liner to the steam. The losses due to con- 

 duction and radiation, to passage through ports, and to 

 incomplete expansion, could also be shown on the 6 <p dia- 

 gram. 



The question of the economical advantage of reducing the 

 power by automatic cut-oft versus throttling was discussed. 

 Broadley, the result was that the gain by varying the expansion 

 was large for a simple engine, moderate for a compound engine, 

 and, for a triple engine, almost inappreciable. It further 

 appeared that the gain at high speeds was greater than at low 

 speeds. 



A few trials made with the cylinders steam-jacketed showed 

 a slight gain, but further experiments were required to show 

 whether the gain was likely to be worth the extra trouble and 

 expense involved. 



The missing steam at cut.-oft varied in the trials to even a 

 greater extent than it did in the non-condensing trials — the 

 amount being much affected by the range of temperature, the 

 density of the steam, and by other conditions. 



It appeared that, under all circumstances, the triple-condens- 

 ing engine showed an advantage over the compound in regard 

 to steam-consumption ; but that, except for very large engines, 

 the compound-transfer engine was probably the best for pres- 

 sures below 150 lbs. (absolute) pressure per square inch. 



ETHNOLOGICAL OBSERVATIONS IN 

 AUSTRALIA. 



COME time ago Mr. R. Etheridge, Jan., carried on a series of 

 ""-^ geological and ethnological investigations in the valley of 

 the Wollondilly River, at its junction with the Nattai River, 

 New South Wales ; and in the latest number of the " Records 

 of the Australian Museum" (vol. ii. No. 4) he gives an inte- 

 resting account of the various facts he had occasion to study. 

 The following is the greater part of the passage in which he 

 records his ethnological observations : — 



The aborigines of the Wollondilly and Nattai Valleys, must, 

 from local accounts, have existed in considerable numbers, 

 and are now only represented by interments, carved trees, 

 wizards' hands, and charcoal drawings in rock shelters along 

 the precipitous escarpments. 



The first objects investigated under this head were the 

 " Hands-on-the-Rock." The " rock " consists of a huge mass 

 of Hawkesbury Sandstone, about seventeen feet in breadth and 

 length, hollowed out on the side overlooking the river to the 

 extent of six feet. It is perched on the side of a gentle rise 

 from the Wollondilly, having rolled from the higher ground 

 above, and alongside the track from the Nattai Junction to 

 Cox's River, in the immediate south-west corner of the Parish 

 Werriberri. The cavernous front of the rock is fifteen feet 



NO. 1225, VOL. 47] 



