138 HISTORY OF SCIENCE. 



the fundamental principle of the art of experiment, which art was, in 

 Bacon's time, in its infancy, and he seems to have had no adequate 

 conception of its power in the investigation of nature. Hence, in his 

 elaborate enumeration of instances, he refers in general only to facts 

 known by observation. Under one head, Instantitz Polychiesta, or 

 things generally useful, he refers to experiments and apparatus, and 

 proposes to treat particularly of these in a subsequent part of his 

 work. 



Two correlative classes of instances are the Glaring (instantice os- 

 tmsivce, eluscentice, or predominantes] and the Obscure (instant ice clan- 

 destines , also fancifully called instantia crepusculi, twilight instances). 

 The former class are those which exhibit the quality in its greatest 

 intensity, predominating over all others which usually obscure or coun- 

 teract it. The latter includes the cases in which the quality showy 

 itself in the faintest manner and most imperfect state. These two 

 classes of instances come under what are now called the extreme cases, 

 and the importance of instances of this kind in verifying an induction 

 has been fully recognized ; for, " by placing our conclusions, as it were, 

 in violent circumstances, they try their temper and bring their vigour 

 to the test." An example not drawn, however, from Bacon himself 

 will enable the reader to perceive the force of an extreme case. When 

 Galileo had concluded, from the comparison of a great number of facts, 

 that the accelerating power of gravity is the same in every kind of sub- 

 stance, he endeavoured to demonstrate the fact by letting light and 

 heavy bodies of different kinds fall from a tower at the same instant, 

 when they were observed to strike the ground nearly at the same time, 

 the heavier being, however, somewhat in advance of the others. But 

 when the bodies were extremely light substances, such as cork or 

 feathers, the experiment appeared to contradict the induction instead 

 of confirming it ; but Galileo properly attributed the difference to an 

 interfering course, namely, the greater . proportional resistance of the 

 air. And as there was no means then known by which this interfering 

 cause might be removed, the experiment could not be tried at all in 

 what would have been the most convincing case of all, the extreme case. 

 But after the invention of the air-pump, the beautiful experiment, now 

 called that of the guinea and feather, was devised. This experiment 

 may be easily understood, even by one who has not witnessed it, by an 

 inspection of Figs. 52 and 53. A gold coin and a feather the ex- 

 tremes of heaviness and lightness are placed on little stages, from 

 which they are simultaneously let fall in the tall glass from which the 

 air has been exhausted. It is veiy curious and convincing to observe 

 the feather falling like lead, and striking the bottom at the same in- 

 stant as the coin. 



A very important class of cases are the Crucial Instances (instanticz 

 crucis), which Bacon quaintly names after the crosses or guide-posts 

 used to point out roads. " These instances are of such a kind, that 



