148 



NATURE 



[April i, 1920 



1 



rocks of Porto Rico divide themselves into an older 

 series, mainly volcanic, which is regarded as Cre- 

 taceous or a little earlier, and a sedimentary Cainozoic 

 group, determined by marine fossils to be of Eocene, 

 Oligocene, and Miocene age, Oligocene beds largely 

 predominating. Though it is not mentioned on the 

 cover of the part, a good map of the island, by Messrs. 

 Reeds and Briesemeister, faces p. 30. The scale is 

 I : 950,400, and red contours are sketched at 100 ft., 

 500 ft., 1000 ft., and 1500 ft. In his detailed account 

 of the geology of the San Juan district Mr. Douglas R. 

 Semmes describes the very interesting and very annoy- 

 ing topography of the Tertiary limestone belt, where 

 "pepinos"- — we prefer this term, meaning "cucum- 

 bers," to Mr. Berkey's "haystack hills" adopted 

 in the paper — give rise to a remarkably broken country. 

 This topography is due to the irregular falHng-in of 

 waterways in the Cainozoic limestone, complicated by 

 the occurrence of beds of shale. In the petrographic 

 section we welcome the appearance of Vogelsang's 

 term " vitrophyre " ; but the German spelling that is 

 retained, even in a plural, which is written " vitro- 

 phyrs," makes us fear that this useful word is here 

 limited as Rosenbusch desired. 



Aerial navigation has become of such vast import- 

 ance that any aid which meteorology can afford is 

 welcomed, while, on the other hand, the meteorologist 

 looks with much expectation to the airman for ob- 

 servations which may advance our knowledge of the 

 general movements of the atmosphere. The Meteoro- 

 logical Office has just issued "An Analysis of Cloud 

 Distribution at Aberdeen during the Years 1916-18" 

 (Professional Notes No. 9, price 4^. net). The 

 analysis is by Mr. G. A. Clarke, assistant at Aberdeen 

 Observatory. It is practically a first effort at averag- 

 ing the number of days in each month on which cer- 

 tain cloud characteristics are predominant, and from 

 this deducing- by the estimated average height of the 

 cloud the occasions when air was cloud-free 5elow 

 certain heights. The number of occasions upon which 

 flying would have been handicapped on account of the 

 lowness of the cloud is 31 per cent, of the total, and 

 of the remainder rather more than one-third show no 

 cloud below 7000 ft. The weakness of the analysis 

 is that the cloud-heights have been worked on average 

 results deduced from altogether different observations. 

 It is recognised by meteorologists not only that the 

 heights of clouds may vary at different stations, but 

 also that they are subject to diurnal and seasonal 

 variations. 



In a paper on operating a by-product producer-gas 

 plant for power and heating, read recently at the 

 Institution of Electrical Engineers, Mr. W. H. Patchell 

 gives particulars of the running of a plant belonging 

 to the Hoffmann Manufacturing Co., Ltd. The gas 

 plant is on the Lymn system, and the power units 

 consist of four-cylinder horizontal Premier engines of 

 about 500 brake-horse-power at 190 revs, per min. 

 Each gas engine is fitted with an exhaust boiler, and 

 the boilers were installed with water-heaters. The 

 dynamos were supplied by Messrs. Crompton, and 

 are open type direct-current shunt-wound interpole 

 360-kw. machines running at 190 revs, per min. ; the 

 NO. 2631, VOL. 105] 



first two machines work at no volts, and in the 

 second instalment of plant, machines working at 

 220 volts are used. The figures obtained for a period 

 of six months' running show a consumption of 1-51 lb. 

 of coal per kw.-hour, and a thermal efficiency of 

 19-9 per cent, on the units delivered to the feeders. 

 The best figure quoted by Mr. David Wilson 

 (Technical Adviser to the Controller of Coal Mines) for 

 electric power stations in the South of England is a 

 consumption of 2-32 lb. of coal per unit and a thermal 

 efficiency of 13-05 per cent. The best station in the 

 Northumberland district gave i-8o lb. of coal per unit 

 and a thermal efificiency of 18 per cent. Mr. Patchell 

 considers that the large-cylinder high-power gji- 

 engine will be developed in this country as it h;i-- 

 been abroad — an opinion in which he appears to differ 

 from some other gas-engine authorities. 



A PAPER read to the North-East Coast Institution 

 of Engineers and Shipbuilders on March 19 by Dr. 

 W. H. Hatfield, of the Brown-Firth Laboratories, and 

 Mr. H. M. Duncan, of Messrs. C. A. Parsons and 

 Co.'s Research Laboratory, deals with the mechanical 

 properties of turbine steels. Unfortunately, the 

 authors were unable to obtain specimens of turbine 

 steel which had done good service in severely stressed 

 parts, and a standard with which they could compare 

 other steels was therefore lacking, but the conclusion 

 is reached that design has probably more to do with 

 the life of turbine parts than the quality of the steel. 

 One disc which failed in practice, however, proved to- 

 be weak when tested in a radial direction, and the 

 defects of structure, are illustrated by means of photo- 

 micrographs. The paper contains a number of tests 

 by different methods, the conclusions as to the relative 

 value of impact, bending, hardness, and tensile tests 

 being, in the main, the same as those reached by 

 Dr. Hatfield in his paper read before the Institution 

 of Mechanical Engineers. An investigation of the 

 Sankey test is included, the relation between the 

 length, diameter, and resistance of the test piece being 

 examined. A formula is given which yields a rough 

 approximation to the values which would be obtained 

 under standard conditions. Formulae are also given 

 for the Stanton repeated impact test, and the data 

 collected should.be of interest to engineers who are 

 concerned with testing. 



Some interesting particulars regarding the use of 

 mechanical reduction gears between the turbines and 

 the propeller in the Royal Navy were given in a paper 

 read at the recent meeting of the Institution of Naval 

 Architects by Eng.-Comdr. H. B. Tostevin. By i9if> 

 it was considered that enough progress had been made 

 to warrant a complete change-over to this type of 

 driving, and at present there are installed or on 

 order 612 sets of gears of a total horse-power of 

 7,828,000. The largest set transmits 36,000 shaft- 

 horse-power, and there are four sets on this ship, 

 totalling 144,000 h.p. In all naval work the turbine 

 spindles, pinions, and gear-wheels are supported on 

 rigid bearings, and the alignment is determined by 

 accurate machine work in boring the gear housings- 

 and fitting the bearings. In general, a gearing ratio 



