Mechanical Science, 409 



meter connected with the interior of the bpttleio shetr th? presstj^re 

 exerted. When a bottle is burst iu this way by the hydraulic press, 

 no violent dispersion of its parts takes place, unless indeed, in 

 plaice of heiwr Jilled with water, a portion of air is left in ; then, 

 when it breaks it flies to pieces, and would cause danger if exposed. 



Bottles intended for the manufacture of brisk Champag-ne or 

 Burgundy, being tried, were found to break with a force between 

 12 and 15 atmospheres, exerted from within outwards: a few rose 

 to 18 atmospheres. Bottles which had contained Champagne of 

 the finest quality, broke at the same pressures. Bottles which 

 resisted the pressure of 12 atmospheres, usually broke with one or 

 two atmospheres more, but the number of these was small. The 

 fracture of bottles in the manufacture of brisk Champagne is from 

 10 to 20 per cent. ; and in certain cases, which, however, are rare, 

 almost the whole have been broken. It appears quite certain, that 

 during the fermentation of the wine, the pressure rises above 

 12 atmospheres, but the full extent can only be ascertained by 

 careful experiments made by the wine proprietors. 



The commissioners then remark, that the best bottles intended for 

 brisk wines are too weak ; the general fault is want of strength 

 and uniformity in the belly of the bottle, especially at the junctions 

 with the neck and with the bottom. 



As the greater number of bottles for brisk wines are of the same 

 quality, it becomes a question why some should break and others 

 not. This difference is supposed to depend upon the form of the 

 neck and quality of the cork, allowing a little gas to escape in some 

 cases and not in others. If the bottles and corks were all alike, 

 all those which contained the same liquor at the same temperature 

 would probably break at the same pressure. The only means of 

 avoiding fractures is either to make the bottles sufficiently strong, 

 or to allow a little escape of gas by the cork. The least thickness 

 of glass in the belly of the bottle should be 2 millimetres ( .079 of 

 an inch), but generally it is only 1 millimetre at the part next to 

 the bottom. — Bull. Univ. E. xiv. 80. 



4^ On jtke Optical Influence of two coloured Objects on each 

 tiher. — The production of complementary colours in the eye, when 

 coloured objects are observed either alone or in association, has fre- 

 quently been an object of investigation with philosophers. Lately 

 M. Cheyreuil has had his attention drawn to it by his situation, and 

 coneeiyes that he has been able to generalize beyond what others 

 hj\a done before him. He has ascertained this general fact, that 

 when two differently coloured bodies are iu juxtaposition, their colours 

 are n|Pditie(i, by the effect of vicinity. If one is lighter than the. 

 othen ,it becbme^ more light whilst the other becomes darkened.^ 

 By experiment, the influence of the seven primitive colours, wit^.^ 

 black and wiiite^, upon. each other, has been ascertained ; the law^ 

 has be^n sought for. which governs these modifications, and is 

 given as follows : — When two colours A and B, are viewed simul- 



