734 



SCIENCE. 



[N. S. Vol. VII. No. 178. 



third series, capillary depression was per- The effect of not equalizing the capillary 

 ceptibly though very slightly unequal, in depression is very apparent when these re- 

 direction to make readings too high." suits are compared with the earlier ones 



In the above quotation 'M' means mil- quoted. But on account of increased skill- 



lionths of atmospheric pressure. The cal- fulness of observation, due to long experi- 



culated probable error of the thirty read- ence, the individual readings of each set are 



ings taken together, is only ninety-two more uniform than before ; so that the net 



hundredths of a unit in the third decimal result is better. 



place ; that is to say, less than a thous- In this example, we have the measure- 



andth part of a millionth of atmospheric ment of about two millionths of atmos- 



pressure. The probable error of the three pherie pressure, with a probable error of 



mean results, considered as single readings, only one part in three thousand, of the 



is only eleven hundredths of a unit in the quantity measured. 



third decimal place of millionths. The net From the foregoing, we may safely con- 

 result may be expressed as follows, in terms elude that with the apparatus described, 

 of atmospheric pressure : Considered as small gaseous pressures may be easily meas- 

 thirty measurements : ured, with a probable error of less than a 

 0.000 000 434 60 ±0.000 000 000 92. thousandth part of a millionth of atmos- 



Considered as three measurements : pheric pressure. 



0.000 000 434 60 ± 0.000 000 000 11. Cleveland, O. ^^^^"'^^ ^^ ^^'"^- 



Here we have the measurement of a total 

 quantity of less than half a millionth of at- 

 mospheric pressure, with a probable error ^0^^^ THOUGSTS CONCERNINO THE TEACM- 



. , , . j-c^i . X fi-u ING OF CHEMISTRY. 

 of only about a fifth of one per cent, of the 



quantity measured. ^^^ ^^^ P^^^ce to a short set of ' Notes 

 To show how small is the effect of vari- ^P°^ Qualitative Analysis,' recently pub- 

 able capillary depression in the large mer- ^i^^^*^' ^ "^^^^ "«« «f ^^^ expression : 

 cuiy columns, the following measurements " ^^lere is small doubt that, were it not for 

 were made July 25, 1897. No correction ^he expense of printing, every teacher of 

 was made of accidental capillary differences, chemistry would use a text-book made by 

 but the columns were always observed with ^^^^'''^^ ^i*^ '^i*^^'- P*^'^ ^^ scissors." 

 a falling meniscus. The zero of the microm- ^^^ ^ ^^^'i^^ "^ ^^^ "t^le book which 

 eter was freshly adjusted for each reading, afterwards appeared in one of the foreign 

 and before each of the six sets of readings Joi^rnals, the critic referred to the above 

 the mercury was lowered and then read- sentence, with the added remark : ' Sad, in- 

 justed to the proper height in the gauge *^*^ed, if true !' He who wrote the criticism 

 in is a distinguished chemist, for otherwise his 

 M. M. M. M. M. M. opinions could not find place in so eminent 

 2.210 2.203 2.209 2.198 2.198 2.202 a journal ; but the thought crosses me": Is 

 .204 .195 .202 .203 .204 .198 he a teacher ? There is a tremendous "dif- 

 .209 .198 .204 .208 .200 .196 j. , , ,, • i- j. i, 

 .203 .204 .210 .200 .196 .208 ^^''^^^^ h^i^^^n the Specialist who never 

 203 192 202 .198 .196 .203 enters the class-room and the trained in- 

 Means 2.2058 2.1984 2.2054 2.2014 2.1988 2.2014 structor who but rarely leaves it. 



Calculating the probable errors we have : -A- man may rank in the highest grade as 



Six mean readings 2.20187 M. ± 0.00073 M. ^ scientist, and yet be nothing of a teacher ; 



All readings 2.20187 " ± 0.00059 " he may be skillful to the last degree in map- 



