262 



SCIENCE. 



[N. S. Vol. XXIV. No. 609. 



through which the end of the arm of No. 1 

 is moved. The same method may be used 

 wherever small accurate rotations of any 

 shaft are desired and the same apparatus 

 may be used on dynamos, engines and other 

 forms of machinery. 



A report was made by Mansfield Merri- 

 man, professor of civil engineering, Lehigh 

 University, South Bethlehem, Pa., on ' Con- 

 stant and Probable Errors in the Estima- 

 tion of Linear Distances and Vertical 

 Angles, as ascertained by 1,712 Observa- 

 tions on 128 Students.' 



The observations were made by asking 

 students to record their estimates of the 

 length, width and thickness of a board, 

 and also of the magnitude of several ver- 

 tical angles. The angles were estimated in 

 degrees and also by the ratio of horizontal 

 to vertical projection. The following con- 

 elusions were drawn: 



1. For lengths, about sixty per cent, of 

 the estimates were too large, and the av- 

 erage constant error was always positive, 



2. For angles, about eighty per cent of 

 the degree estimates, and about sixty per 

 cent, of the ratio estimates, were too large. 



3. The estimates of vertical angles by the 

 ratio method are more accurate than those 

 by degrees. 



4. The probable error of a single esti- 

 mate is larger than the average constant 

 error. 



5. The estimates of the freshmen class 

 had much larger errors than those of other 

 classes. 



This interesting example of the applica- 

 tion of laboratory methods to the class- 

 room created much interest on the part of 

 the engineering teachers present. 



The next paper was read by the secre- 

 tary, and described and discussed the re- 

 sults of 'Some Experiments on the Fric- 

 tionless Orifice' and was by Horace Judd 

 and Roy S. King, assistant professors of 

 experimental engineering at Ohio State 



University. The paper is in sequence 

 with one presented by one^ of the authors 

 at the St. Louis meeting, on 'Pitot Tubes; 

 with Experimental Determinations of the 

 Form and Velocity of the Jets,' and which 

 was published in the Engineering News of 

 March 31, 1904, page 318. 



The experiments described in the paper 

 were made on five frictionless orifices, 

 ranging in size from three fourths inch to 

 two and one half inches in diameter. The 

 term 'frictionless orifice' is one which has 

 been commonly used in the mechanical 

 laboratory at the Ohio State University 

 for many years. It is the same as what 

 some of the text-books call 'an orifice in a 

 thin plate,' and indicates that there is no 

 appreciable loss by friction by a stream of 

 water flowing through such an orifice. So 

 far as known, the term was first employed 

 as early as 1873 by Professor S. W= Robin- 

 son, emeritus professor of mechanical en- 

 gineering at the Ohio State University, and 

 will be found on page 552 of Vol. VI. of 

 the Geological Reports of Ohio, 1886, in his 

 report on the 'Measurement of Gas Wells 

 and other Gas Streams and the Piping of 

 Natural Gas.' 



The facilities available consisted of a 

 closed standpipe into which several pumps 

 could force water. A horizontal drum 

 was connected near the bottom of the 

 standpipe, and in the end of the drum 

 the orifice to be tested was secured; the 

 coordinates of the issuing jets were meas- 

 ured by the hydraulic micrometer caliper, 

 described in a paper presented by Pro- 

 fessor Wm. T. Magruder at the St. Louis 

 meeting of the association." The water 

 was caught and measured in a calibrated 

 cistern. 



The experiments were made to find: 



^Science, N. S., Vol. XIX., No. 479, Marcli 4, 

 1904, p. 364. 



= Science, Vol. XIX., No. 479, March 4, 1904, 

 p. 364. 



