May 6, 1897] 



NATURE 



19 



tractor, although differing from the latter form to a certain 

 extent. With the discharge orifice of this form a fair sjieed 

 was obtained, whereas with ordinary discharges the boat could 

 hardly be moved along. 



The Rev. Mr. Capell, the originator of this discharge nozzle, 

 gave particulars of the experiment made with the vessel, and 

 concluded that the success of jet propulsion depended on the 

 form of the discharge nozzle. The particulars given were not 

 sufficiently detailed to- enable the problem to be adequately 

 discussed, and it would be requisite to know, before arriving 

 at any conclusion, whether the observations taken were 

 properly verified, and the recording instruments were sufficiently 

 trustworthy for implicit reliance to be placed upon them. 



Mr. Walker's paper was on an interesting subject, and some 

 of the experiments which he showed were of a practical nature. 

 They will be doubtless useful to those not acquaiiited with the 

 details of this field of research. The question of relation of 

 speed, power absorbed, and air discharged with propeller 

 ventilating fans was discussed. Seventeen three-bladed fans 

 were tried, being driven by a continuous-current series- 

 wound electrical motor of about one-third electrical horse- 

 |X)wer. The current was taken off the mains of the West- 

 minster Electric Supply Corporation. The fans were run 

 at a speed up to six hundred revolutions a minute ; the velocity 

 of the air was measured by an anemometer. The results, which 

 are too voluminous to quote in full, were contained in tables 

 attached to the paper. The effect of cross section of fan-blades 

 was discussed in the paper. The blades were of sheet-iron ; all, 

 excepting one, of 1/16 inch thick. Their cross sectional lines 

 were all composed of straight lines or arcs of circles. The fans 

 in each group differed from one another only in the cross section 

 of their blades, which were flat, plano-convex, or concavo- 

 convex of different degrees of curvature. A notable feature of 

 the experiments made by the author was that the effect of put- 

 ting a curved surface upon the back of a flat-bladed fan, thus 

 giving a plano-convex section, was to increase the mechanical 

 efficiency 28 per cent., the volumetric efficiency 54 per cent., 

 and the pressure efficiency i -4 per cent. The angle of the blades 

 was 1 7°. The most efficient fan of the group was one having 

 a blade concavo-convex in section with a hollow space between 

 the faces, when the mechanical, volumetric, and pressure effi- 

 ciencies were respectively 28, 65, and 21 per cent. The 

 efficiencies were thus increased by making the blades thicker in 

 the middle of their breadth. To test the effect of feeding the 

 fans from the tips of the blades the delivery tube through which 

 the air was passed was moved forward, the fan thus being out- 

 side the tube. This increased the mechanical, volumetric and 

 pressure efficiencies from 169, 62*0, and 2"0 to 294, 78*0, and 

 3 "I per cent, respectively. The velocity of air on entering and 

 leaving the fans was measured by the anemometer. Experi- 

 ments were made to test the effect of a contracted outlet and 

 inlet. The fan worked partly in a delivery tube, the outer end 

 of which was partially closed by plates with holes of varying 

 sizes. The efficiency was naturally much reduced. It was 

 anticipated that the slow speed of the blades near the centre 

 partly accounted for this, and a circular disc was therefore fixed 

 in front of the fan on the delivery side. This prevented the air 

 passing back again through the centre of the fan, which it 

 might do owing to the slow speed, and the efficiency was raised. 

 The more the delivery orifice was closed, the larger had to be 

 the disc. 



Without entering into theoretical views as to the action of the 

 blades, the author stated that, having regard to the stream-line 

 principle, the section of the blades should be as ship-shape as 

 possible. The two losses in an air-propeller are rotary motion 

 imparted -to the air, and skin friction of the blades. The loss 

 from the latter cause was found to be comparatively small by 

 means of experimenting with flat thin blades set at a plane 

 coinciding with the plane of rotation. 



The summer meeting of this Institution will be held this year 

 in Birmingham, during the last week in July. 



ANNUAL MEETING OF THE U.S. NATIONAL 



ACADEMY OF SCIENCES. 



T^HE National Academy of Sciences held its annual meeting at 



Washington, April 20-22, with about the usual attendance 



of members, but a marked paucity of papers, only fourteen 



having been read, of which number five were biographies. 



These were of Dr. G. Brown Goode, by Prof. S. P. Langley ; 

 of Prof. Thomas L. Casey, by Prof. H. L. Abbot ; of Prof. 

 Charles E. Brown-Sequard, by Prof. H. P. Bowditch (by title) ; 

 of Prof. Hubert A. Newton, by I'rof J. W. Gibbs ; and of 

 Mr. George H. Cook, by Prof. G. K. Gilbert. 



An experimental study on the influence of environment upon 

 the biological processes of the various members of the colon- 

 group of bacilli, by Dr. Adelaide Ward Peckham, was presented 

 by Prof J. S. Billings. 



Prof T. C. Mendenhall read a paper on the energy involved in 

 recent earthquakes. He also read a paper on a ring pendulum 

 for absolute determinations of gravity, giving results of a sug- 

 gestion of Mr. A. S. Kimball that a disc of metal vibrating in 

 its own plane would constitute an improved apparatus for such 

 determinations. This gives the equivalent of a pendulum of 

 any length from infinity to that of the diameter of the outer 

 circumference of the ring. The ring is suspended from its 

 inner circumference ; and the length of the equivalent pendulum 

 is computed by the following formula, in which / is the length 

 required, R is radius of the outer, and R„ is that of the inner 

 circle : — 



^^R3-f-3R^ 



With a crudely prepared disc of this description, results were 

 obtained correct to one part in 10,000. 



Prof S. C. Chandler read a paper on variation of latitude, a 

 full abstract of which will appear in Nature. He also pre- 

 sented another paper on variation of latitude and constant of 

 aberration from observations at Columbia University, by 

 Messrs. J. K. Rees, H. Jacoby, and H. S. Davis. These 

 observers report a series of observations extending from 

 May 9, 1893, 'ill June 14, 1894, divided into groups of from 

 30 to 100. They confirm Chandler's period of about 427 days. 

 They also fix accurately the latitude of the observatory of 

 Columbia University, which is 40° 48' 27"'I95. 



Prof A. A. Michelson gave a description of a new harmonic 

 analyser, an apparatus devised by him, which enables him to 

 integrate in a few minutes long and difficult problems such as 

 would require weeks for mathematical solution. 



In his paper on the position of the Tarsiids and relationship 

 to the phylogeny of man. Prof Theodore Gill maintained that 

 man is more nearly allied to the chimpanzee and the gorilla 

 than to the orang-outang ; the abbreviation of arms and loss of 

 cranial ridges having been caused by disuse of arms for tree 

 climbing, and of teeth for crushing branches, &c., so that 

 powerful facial muscles were no longer required, nor the ridges 

 to which they were attached. The teeth also approached more 

 closely together, filling up the gaps in jaw of apes. Children 

 still show ancestral type in disproportionate length of arm. 



Prof. A. Agassiz read a paper on some recent borings in coral 

 reefs, in which he maintains that the old Darwinian theory of 

 subsidence is no longer tenable, as that would require a thick- 

 ness of 2000 feet in such reefs, but in most cases examined the 

 thickness was within 130 feet. Observations include the 

 Yucatan atoll, about 30 fathoms ; Solomon Islands, 125 to 

 130 feet ; Florida elevated reef, 60 feet, but this has been denuded 

 and may have been originally of twice this thickness ; along the 

 coast of Cuba, 145 feet. Prof. Agassiz attempted to measure 

 the thickness of the great coral reef near Australia, which is 1500 

 miles long, and 50 to 75 miles wide ; but could not yet obtain 

 accurate results. He is confident, however, that the thickness 

 of it is only 25 to 30 fathoms. Prof. Agassiz concludes, 

 however, that barrier, fringing and atoll reefs are none of 

 them thick. 



Prof. A. W. Wright read a paper on some recent experiments 

 in Rontgen rays. By using plane glass he obviates the misleading 

 action of a prism in which the thick part absorbs rays, and indi- 

 cates an apparent negative index of refraction. No indication of 

 refraction was found, however, in using plane glass arranged at 

 an angle so that it would refract rays of light. A thin beam of 

 X-rays was also passed between the poles of a powerful magnet. 

 The poles were then reversed, but no change in the direction oi 

 the rays could be detected. Some very recent experiments, 

 however, which he has not yet fully verified, seem to show that 

 perhaps these rays may be diffracted, even if not capable of being 

 refracted. The conjecture is due to the fact that, on passing 

 the beam through a platinum net-work in the manner described, 

 faint interference lines seemed to be produced. 



NO. 1436, VOL. 56] 



