136 report— 1878. 



rocks), from their real thermal capacities in the wet and dry states, by- 

 volume and by weight. 



Percentage corrections thus deducted from the specific heats by volume 

 of porous rocks in the table, must be added, however, (and vice versa) to 



the value of the ratio - given in its last column. The specific heats of 



dry and wet sand were found by enclosing them hermetically in a thin 

 flask, and are free from any such source of uncertainty as that described ; 

 while the amount of water present in the heated clay and in two speci- 

 mens of brick was known, and the real allowances having been made for 

 these, the specific heats assigned to them in the table need not be cor- 

 rected. The following examples of the correction when it seems to be 

 required will illustrate its use in other cases. 



The average specific heat by volume of three specimens of Calton Hill 

 trap rock (taken from the escarpment of rock on the west side of the 

 Observatory) is 052* ; the average percentage limit of correction to this 

 quantity is 3g- per cent. Taking a half as a probable fraction of it (or 

 0-016 X 0'52 = ■£$ X '52 = -009) and subtracting it from the provisional 

 value, the corrected specific heat by volume of the Calton Hill trap rock 

 is 0511 ; or 051 instead of - 52 as presented in the table. The average 



value of the quantity - for the same specimens is O0060, in the Table, and 



c 



the rate of correction which this requires will be the same as that just 

 used (^o)) for the specific heat, but to be added instead of subtracted 

 from it, with the result O'OOBl instead of O0060 taken from the table. The 

 value of the same ratio calculated by Sir W. Thomson from the under- 

 ground thermometer observations was O00786 ; and adopting for this 

 rock's specific heat by volume the value - 524, which agrees very closely 

 with the value 0-511 here arrived at, he deduced a value of the thermal 

 conductivity of - 00415, rather higher than the mean, - 00312, of the values 

 here concluded by the application of the direct method of experiments. 



The average specific heat by volume of six specimens of Craigleith 

 sandstone, four of which are from the site of the thermometer borehole 

 which was established there f by Professor Forbes in the year 1837, is 



* Two additional places of figures to which the observations and calculations 

 for the table originally extended are suppressed in this abridgment of its determi- 

 nations in order to present them together in an easily apprehended view, where it is 

 believed that very little material utility of the individual results is sacrificed by 

 rejecting the last two figures obtained by the reductions. 



•* f" The quarry was visited on March 17th, 1878, under the direction of Mr. Wedder- 

 burn from Messrs. Adie & Sons, and of Mr. Wallace from the Edinburgh Eoyal Ob- 

 servatory, by Professor Herschel ; the foreman of the present tenant (Mr. Hunter, of 

 Newcastle) soon communicating the object of his search to the best possible authority 

 at the quarry, one of the oldest workmen there, Robert Buchanan, who assisted in 

 sinking the hole, and in depositing the thermometers in it in January, 1837. The 

 face of the quarry in that neighbourhood has not been altered, although the field 

 between it and the tenant's house of Craigleith-Hill, in which the hole was sunk to a 

 depth of 24 feet, was opened over nearly the whole area near the house to a depth of 

 about 20 feet to seek for " Liver rock " (the finest, stone of the quarry) in the imme- 

 dia e neighbourhood of the tenant's house. But the search was unavailing. Robert 

 Buchanan was Mr. Johnson's foreman when the ground was thus turned over and the 

 thermometers were taken up ; and he directed the excavations. He selected for Pro- 

 fessor Herschel from the face of the quarry near their site several specimens of the 

 rock coinciding as nearly as possible with the top and bottom beds of the sandstone, 

 .through which the thermometer hole must have passed. Four plates cut from these 

 specimens are named T 1, 2, and B 1, 2 in the table ; and two more plates of ordinary 

 samples of the quarry stone, " Liver rock," a uniform stone without clefts or partings, 



