PROCEEDINGS OF THE POLYTECHNIC ASSOCIATION. 485 



the eflect on the cousumptiou of the tallow is too small to be 

 recognized. 



The observations of the later experimenters agree in throwing 

 doubt upon the iiiterpreation which Dr, McKeever gave of his 

 own experiments. 



Prof. Horsford then proceeded to detail a series of experiments 

 he had made, showing the probable source of error in Dr. McKee- 

 ver's investigation. He ascribes it to the incidental greater flaring 

 of the candle in the dark. The experiments with the lanterns he 

 explained by the well known effect of dark paint in absorbing 

 radiant heat, and converting it into heat of conduction, by which 

 the air in the painted glass lantern was more heated than in the 

 lantern that was not painted. 



Prof. Horsford then gave an account of the diminished draft in 

 the range flue of his dwelling house during the recent hot term, 

 which rendered it impossible to bake meats or bread in the oven 

 of his range. This continues from eleven o'clock to about three, 

 within which hours bread could not be baked. With the decline 

 of the sun in the afternoon, as in the early morning, the oveu 

 performed its oflice better. 



The chimney was fifty-four feet high. The roof of the house 

 •wiis of dark slate. It was all exposed to heat at about eleven. 

 Some of it began to pass into shade at about three. 



In the eflect of this greater exposure to the sun, during the 

 hours when the sun was highest, Prof. Horsford found the expla- 

 nation of the observed phenomanon. The heated top and sides 

 of the house warmed the air in contact, giving rise to an upmoving 

 column from the top of the house, and to an endless shroud of 

 air sweeping up the sides of the house. This ascending shroud 

 by friction draws the air from the cracks, doors and windows, ot 

 the house, lessening the pressure of the air in the interior, and of 

 course diminishing the draft. 



After showing the applicability of this explanation to the various 

 cases that occur with dwellings, iron furnaces, <fec., the paper con- 

 cluded with the following summary : 



First. That sunshine falling on the flame only of a burning body 

 does not afiect its rate of combustion. 



Second. That other things being equal, neither light nor dark- 

 ness exert appreciable influence on the rate of combustion. 



Third. That other things being equal, two samples of the same 



