39$ 



ALFRED HARKER 



Fig. 3. 



cut off by straight lines joining the points P, Q, etc., to X. If 

 we now draw curves such as Pp, Qq, etc., to indicate in a general 

 way the usual behavior of the several bases in average igneous 

 rocks, we obtain some idea of the respects in which the mixed 

 rock is peculiar. In the diagram it is shown as being unusually 



poor in the first base 

 J> and rich in the second. 



Next suppose a rock- 

 magma to become en- 

 riched in lime by dis- 

 solving limestone, in 

 the proportion of a 

 parts of the magma to 

 b parts of lime. The 

 result is illustrated by the diagram, Fig. 4. The line OM is 

 divided at m in the ratio a to b. If MP represents the lime in the 

 original magma, mp' will represent that in the mixture, this 

 ordinate being cut off by the straight line joining P to Y. If 

 MQ and MR represent two other bases in the original magma, 

 mq' and mr' will represent them in the mixture, these ordinates 

 being cut off by straight lines joining the points Q and R to 0. 

 It is noteworthy that the mixture, as compared with ordinary 

 igneous rocks, may be unduly rich in other bases besides lime : 

 in the figure this is the case with the second of the two bases 

 represented (curve passing through R). 



The foregoing general considerations lead us to anticipate 

 that a rock resulting from admixture may be, and in many cases 

 must be, of peculiar chemical composition. A rock-series, for 

 example, may consist of basalt, pyroxene-andesite, dacite, and 

 rhyolite; but it does not follow that a mixture of basalt and rhy- 

 olite will produce an andesite or a dacite. Still less will a basalt 

 be converted into an andesite by addition of silica, or a rhyolite 

 into a dacite by addition of lime. The processes by which differ- 

 ent igneous rocks have been evolved from a common stock are 

 too complex and subtle to be reversible, at least by so crude a 

 method as that of admixture. To illustrate these remarks from 



