114 



EXPERIMENT STATION. 



[Jan. 



Amount of Glassware tested. 



Year. 



Total Number 

 of Pieces. 



Number 

 Inaccurate. 



Per Cent. 

 Inaccurate. 



I90I, 

 190-2, 

 1903, 

 1904, 

 1905, 

 1906, 

 1907, 



Totals, 



.5,041 

 2,344 

 2,240 

 2,026 

 1,665 

 2,457 

 3,082 



18,855 



291 

 56 

 59 

 200 

 197 

 763 

 204 



1,770 



5.77 

 2.39 

 2.63 

 9.87 

 11.83 

 31.05 

 6.62 



9.391 



The grand totals may be further subdivided into the several 

 classes of glassware of which they were composed : — 



Character of Glassware tested. 



Kind. 



Total 

 Number. 



Number 

 Inaccurate. 



Per Cent. 

 Inaccurate. 



Cream bottles, . 

 Milk bottles. 

 Skim milk bottles. 

 Pipettes, . 

 Acid measures. 

 Totals, 



7,714 

 6,826 



675 

 2,834 



806 



710 

 784 

 106 

 69 

 101 



18,855 



1,770 



9.20 

 11.49 

 15.70 



2.43 

 12.53 



9.391 



The manufacturers repeatedly protested against the refusal 

 of their glassware, and asserted that similar shipments were 

 passed in other States. Such might easily have been the case, 

 where the error was small, due to differences in method of test- 

 ing or in allowable limit of error, possibly both. In some 

 instances the condemned pieces were forwarded to another sta- 

 tion and re tested, but even this apparently failed to satisfy the 

 manufacturers. It became evident that further investigation 

 was necessary in order to bring the matter to an uncjuestion- 

 able basis and to remove all reasonable grounds for com})laint. 



1 Average. 



