202 Prof. T. G. Bonnet/ — On Dana's Classification of Hocks. 



and schistose rocks not necessarily metamorphic clastic. Within 

 limits there is truth in this statement ; bat the qualification is all- 

 important. I venture then to inquire whether Professor Dana has 

 carefully studied the microscopic structure of these two groups and 

 found that there is no difference in this respect between massive 

 igneous and massive metamorphic clastic rocks, or between schistose 

 igneous and schistose clastic rocks. This is a question to the study 

 of which I have devoted much time both in the field and with the 

 microscope ; and mostly with this result — that in the former case 

 not seldom the distinction was plain enough when sufficient pains 

 were taken in looking for it ; in the latter it was usually very marked. 

 Hence, notwithstanding the weight of Professor Dana's authority, I 

 venture to protest against any system of classification which sets at 

 naught the distinction between a sedimentary and an igneous rock. 

 There may be cases where metamorphism has been carried so far that 

 the distinctive characters of the former rock are lost, but the 

 evidence in favour of this is certainly far less satisfactory than it is 

 commonly asserted to be, and (in the present state of our know- 

 ledge) to assume that such cases exist will, I think, encourage 

 sloth in research and be a backward rather than a forward step in 

 science. 



There is also, from his own point of view, another serious objection 

 to Professor Dana's classification. He draws a very marked dis- 

 tinction between the potash-felspar series (including leucite) and the 

 soda-felspar (including nepheline). But while not denying the 

 utility of this, in part, we must remember that there are few potash- 

 felspar rocks in which there is not some soda-felspar (including 

 oligoclase), and that cases are not uncommon in which the one about 

 equals the other in quantity. 1 Even Professor Dana, by his remark 

 on the nepheline rocks (phonolites, etc.), virtually admits that the 

 distinction can hardly be maintained ; and further, what are we to say 

 of the association of leucite and nepheline, and of the occurrence of 

 these minerals with both orthoclastic and plagioclastic felspars ? 

 Again in Professor Dana's classification we have lherzolite grouped 

 with eclogite, and serpentine with chlorite schist; simply because 

 they contain no felspar, notwithstanding their great mineralogical and 

 chemical differences. 2 



1 It is not seldom hard to say (even after chemical analysis) whether a rock 

 should be called a syenite or a diorite, a minette or a kersantite. 



2 Analysis of eclogite (garnet and omphacite, with quartz, disthene and mica) from 

 Eppenreuth. Si O 2 = 57-10 Al 2 3 = ll-66 Fe 2 3 = 2-84 FeO = 3-22 MnO = 0-31 

 Mg 0=637 Ca 0=1380 K 2 0=0-81 Na 2 0=2-21 H 2 0=0-54. Analysis of 

 lherzolite (olivine,enstatite and diopside with picotite) from Kalohelmen. Si 2 = 37'42 

 Al 2 O 3 = 0-10 Mg 0=48-22 Fe = 8-88 MnO=0-17 M 0=0-23 H 2 0=0-71. 

 (Yon Lasaulx, Elem. derPetrog.) Analysis of Serpentine (from Cornwall). Si 2 = 

 38-50 Alo0 3 = 1-02 MgO=36-40 Ca =1-97 Fe 2 3 = 4-66 Fe 0=3-31 NiO=0 : 59 

 H 2 0=12-35 Fe S = 0-41. Undecomposed residue =1-37. (Q.J.G.S. xxxiii. 925). 

 Analyses of Chlorite schist variable ; these are two given by Zirkel (Petrog. 

 i. 311) (l)Si0 2 = 31-54 Al,0 3 = 544 Fe,O 3 = 10-18 Mg 0=41-54 H,0=9-32 

 (2)SiO 2 = 42 08 Al 2 3 = 3-57 Fe 0=26-85 Mn 0=0-59 Ca = 1-04 MgO = 

 17-10 H 2 = 11-24. In such a grouping, even chemistry, as it seems to me, is 

 fairly thrown overboard. 



