632 [June 18, 



distances at which the cross of stars begins and ceases to be visible, 

 those at which the series of parallel lines is seen, the velocity in 

 miles per hour, the weight of each engine, and also the length and 

 weight of each train (when it could be identified) . 



This Table proves that in all cases but one (which in fact is scarcely 

 an exception) there is sufficient vibration to excite the cross at 670 

 yards, and that in 24 per cent, of the number it is seen beyond 

 1000, its maximum being 1176. At the southern end such dis- 

 tances reach far beyond the tunnel, while at the north they fall 

 within it. From comparing them in the two cases, the author infers 

 that the train's agitation extends laterally as far when it is in the 

 tunnel as when in the open cutting. The amount of disturbance 

 does not depend solely on the velocity and weight of the train, but 

 also on other circumstances, of which prolonged action and length of 

 train are the chief. In one instance, with only a velocity of 1 1 '4 miles, 

 the cross was seen at 1110 yards a proof that no regulation of the 

 speed in passing an Observatory at a distance of 300 or 400 yards 

 would be of any avail. 



The system of parallel lines is only seen between lines making 

 angles of 45 with the perpendicular to the rails, that is, at distances 

 under 427 yards; it scarcely ever is produced unless the cross be 

 visible beyond 1000 yards. 



These forms are also produced by the reports of cannon of twelve 

 ounces calibre, at distances from 300 to 3000 yards ; in the last case 

 there is but a faint trace of the cross. In all, the appearance is mo- 

 mentary, not lasting in any case more than a second and a half. They 

 are not produced by the roar of a two-pound rocket fired 82 feet from 

 the mercury, though very loud. When the cannon were fired in the 

 tunnel, where the perpendicular meets it, two sets of tremors were 

 seen one, he believes, propagated through the ground, the other 

 through the air about a second later, the sound escaping probably 

 through the shafts. Attempts were made to substantiate or refute 

 this hypothesis ; but the difficulties of rapidly shifting and unshifting 

 the coverings prepared for the purpose were such as to compel him 

 to relinquish them. 



These observations were reduced in 184 7 ; but conceiving all danger 

 to the Royal Observatory was past, the author did not think it 

 necessary then to proceed with them. As, however, no Observatory 



