March 3, 1905.] 



SCIENCE. 



353 



nouns this is extremely j.rbiti'ary; and, more- 

 over, in all inflected languages, words very 

 often undergo transformation in gender dur- 

 ing course of time. Le Peloponnese, for in- 

 stance, has a feminine termination, but is of 

 the masculine gender; and Galilee may be of 

 either gender. Val, feminine in the Latin 

 vallem, and still feminine in French proper 

 nouns, has become masculine by common 

 usage, taking the plural vaux by analogy with 

 mal, cheval, etc. Some words, like sang, are 

 masculine in certain combinations, and femi- 

 nine in others. Finally it can hardly be 

 claimed that the form ' Mont Pelee ' does 

 violence to a language which authorizes us to 

 place the feminine article before hon-hec, and 

 the masculine before a variety of words like 

 rouge-gorge, rouge-queue, cent-garde, grand' 

 croix, patte-pelu, etc. 



C. R. Eastman. 



Harvard Uxiversity. 



the metric fallacy. 



To THE Editor of Science : In a recent 

 article in Science on the discussion of the 

 metric question in the House of Lords, Dr. 

 Seaman repeats with approval the claim as 

 to the great saving of time that would be 

 accomplished in school by the use of the 

 metric system. This claim has been one of 

 the chief supports of the metric cause for 

 generations, and has remained practically un- 

 challenged except by a general denial. The 

 forms in which is was presented in the House 

 of Lords last February and in the report of 

 our House Committee on Coinage, Weights 

 and Measures in 1902, are so typical of this 

 claim that both are given here: 



Lord Belhaven, in House of Lords, February 23 : 

 " There is a great waste of time in tlie education 

 of children, t.hrougli the learning of the arithmet- 

 ical tables and their application. Out of 221 

 school-masters consulted, 212 replied. One hun- 

 dred and ninety-seven stated that there would be 

 a considerable saving of time if the metric system 

 were introduced; of these, Ifil estimated the sav- 

 ing at one year; 30 estimated it at two years; and 

 six went so far as to estimate it at three years." 



Report of the Committee on Coinage, Weights 

 and Measures to the House of Representatives, 

 April 21, 1902: " Estimates made by the Depart- 



ment of Education and others show that tiie work 

 of at least two thirds of a year in the life of 

 every child would be saved by the adoption of the 

 metric arithmetic. " * * Teachers and pupils 

 alike imanimously testify as to the ease with 

 wiiica the system is taught and learned and iiie 

 faculty with which it is applied to the problems 

 which in ordinary arithmetic are complex and 

 difficult to solve. When we consider that there 

 are over 15,000,000 school children *n the United 

 States being educated at a public cost of not less 

 than $200,000,000 per year, the enormity of the 

 waste will be appreciated. In the lifetime of a 

 single generation nearly $1,000,000,000 and 40,- 

 000,000 school years are consumed in teaching a 

 system which is in harmony with that of no other 

 nation of the world." 



This argument has been reiterated with so 

 much emphasis and with such a show of au- 

 thority that it has unquestionably carried 

 conviction to the minds of thousands. The 

 opinions of experts regarding their own trade 

 are ordinarily accepted by others. If edu- 

 cators say the metric system would efFect a 

 saving of one to three years in the school life 

 of a child, why should it not be accepted as 

 true ? 



Within a few weeks Frederick A. Halsey 

 has applied the scientitic method to the scliool 

 children argument, and, in view of its gen- 

 eral acceptance, with startling results. It is 

 to him that I am indebted for the data on 

 this point. In the report of the course of 

 study for elementary schools, dated May 27, 

 1903, the board of education of New York 

 city gave a time schedule for each study for 

 the eight years. This schedule is based on 

 1,500 minutes per week, and the time allotted 

 for all branches of mathematics amounts to 

 34J weeks for the eight years. No reliable 

 data is available as to the proportion of this 

 time occupied in the study of weights and 

 measures; 20 per cent, of the text-book on 

 arithmetic, however, is occupied by the chap- 

 ters on compound numbers, weights and meas- 

 ures. In order to be liberal to the metric 

 cause we will apply this rate, 20 per cent., to 

 the whole time, including that occupied with 

 algebra and geometry. The total time de- 

 voted to the study of compound numbers, 

 weights and measures during the eight years 



