Study of the New York Obelisk as a Decayed Boulder. 157 



emission of heat of definite amount, at a constant rate, from a point, 

 through a homogeneous medium; even thus, the rates of increment, 

 at successive points along a radius of the spherical heat-wave pro- 

 jected through the medium, would evidently decrease, at a rapidly 

 augmenting geometrical ratio with the distances from the center. 

 In our experiment, however, the following sources of variation and 

 disturbance must have accompanied these theoretical conditions: — 



(a). Irregular distribution of temperature through the rock, 

 before the experiment. 



(6). Irregular source of heat: an indefinitely large number of 

 points, yielding heat in varying amount and intensity. The fuel 

 had to be re-adjusted in the stove, twice during the afternoon, with 

 distinct influence in cooling the surface of the stone; and farther 

 variation must have been produced by the slight breeze which 

 sprang up in the latter part of the afternoon. 



(c). Heterogeneous medium : an aggregate of several minerals 

 of difi"erent conductivity of heat, chiefly quartz, feldspars, biotite, 

 and hornblende: the occurrence of these minerals in crystals of 

 varying size, lying in all positions, with interstices of irregular size 

 intervening: separation of the aggregate into laminae of varying 

 thickness (mostly 2 to 3 centimeters) and texture, with the biotite- 

 plates mostly arranged in parallelism with the lamination-planes 

 and in part along those planes. 



{d). Presence of moisture in the interstices, probably in varying 

 quantity in diff'erent layers of the rock, and producing irregular 

 conversion of sensible into latent heat, during the production and 

 the escape of vapor. 



(e). Radiation of heat and vapor, both from the heated surface, 

 on the east face of the boulder, and laterally from the north face. 



In considering the figures in the table, the influence of these, and 

 probably other conditions of variation, is strongly marked. Taking 

 as a standard the average number of seconds in time required for an 

 increment of one degree of temperature (Centigrade) to a depth of 

 one centimeter, we find great oscillation along any line, either of depth, 

 as marked by a particular thermometer, or of period of time, par- 

 ticularly of the latter. At any depth, within about 8 or 9 centi- 

 meters from the surface, the average increment of 1° per cm. varies 

 from 25 to over 50 seconds, say about 36 seconds ; while at any 

 periods, passing across the columns of depth, the average increment 

 varies up to more than 100 seconds. At the extreme depths of 25 



