March 9, 1883. 



SCIENCE. 



12T 



METEORIC AND TERRESTRIAL ROCKS. 



An unavoidable delay in the completion of 

 the plates of a work to be published in the 

 memoirs of the Museum of comparative zoology 

 has rendered it advisable to publish in advance 

 a brief abstract of some of the results thus far 

 obtained. The vcork will contain descriptions 

 of the microscopic characters of meteorites 

 and the allied rocks ; their classification ; col- 

 lected and arranged chemical analyses ; a dis- 

 cussion of the principles of classification ; the 

 origin of rocks ; the present and past state of 

 the earth in its bearings upon petrographj', 

 etc. 



The previous delaj-s in the publication of 

 this work have been owing to other labors, a 

 change of plan greatlj- extending its scope, 

 and the fact that work of the proposed char- 

 acter is vastly more difficult than simply 

 ' pigeon-holing ' rocks in different species, 

 according to the minerals they happen to con- 

 tain. 



The results which it is desired to present 

 here are as follows : — 



1 . Petrographical research demands a former 

 liquid globe, and one whose interior portions 

 are either now liquid, or in such a condition 

 that the}' can readily become so. 



2. That the interior of the earth is now 

 probablj' liquid, is shown not only from petro- 

 graphical a,nA geological research, but also bj' 

 the fact that the best and more recent observa- 

 tions either prove or render it probable that 

 iron and such rock materials as are believed to 

 compose the in/ra-sedimentarj- portion of the 

 earth are lighter when hot-solid, at or near the 

 melting point, than thej- are at about the same 

 temperature when liquefied. Hence, according 

 to Thomson's law, pressure lowers their fusing 

 point, instead of raising it. 



3. No sinking of the earth's crust to the 

 centre could take place ; for, since the interior 

 is heterogeneous, the crust on sinking would 

 meet with material of higher specific gravity, 

 the heat imparted to the sinking matter would 

 cause it to grow lighter, and the viscositj' of 

 the material still liquid would retard its descent. 



4. All so-called physical and mathematical 

 demonstrations of the earth's soliditj' have 

 been based on certain hypothetical globes of 

 unlike constitution with the earth ; and hence 

 have not the slightest application to it, but to 

 the hypothetical globes only. 



5. All rocks originally came from the cool- 

 ing molten material of the globe, and the 

 chemical and sedimentary rocks have resulted 

 from the disintegration of that material. 



6. All eruptive or volcanic (including plu- 

 tonie) rocks were derived from material which 

 either had never solidified, or had been re- 

 liquefied ; but they were not derived from sedi- 

 mentary or chemical deposits. 



7. In the shrinkage and fracturing of the 

 earth's crust, the depression of any portion 

 into the still molten interior would naturally 

 displace and cause the hea-\'ier liquid to over- 

 flow, just as the fracturing and depression of 

 ice causes the heavier water to overflow it. 



8. Water is the accident of an eruption, and 

 is not the cause. It is met hy the lava on its 

 waj' to the surface, but is not the cause of the 

 advance towards that surface. Hence it is 

 probable that explosive volcanic action has be- 

 come more common in recent times, while 

 quiet outflows were more abundant in past 

 ages. 



9. Regions of crystalline rocks are, as a 

 rule, regions in which eruptive, or mixed 

 eruptive and sedimentarj', agencies have pre- 

 vailed, and are of every geological age, — 

 meaning, bj' eruptive agencies, the original 

 and secondary results of a cooling globe, in- 

 cluding thermal waters. Metamorphism is even 

 more common in eruptive than in sedimentary 

 rocks. 



10. The original rock-materials of the uni- 

 verse are the same, from whatsoever region 

 they come, and the same principles should be 

 emploj'ed in classifying them ; while the classi- 

 fication, to be natural, ought to express their 

 relationships. 



11. A natural classification of rocks should 

 be based on all their characters taken as a 

 whole. It must be an empirical one, as in 

 zoology and botany ; and ascertained b}' 

 studying all known forms, and arranging them 

 according to their petrological, lithological, 

 and chemical characters, — taking the rocks as 

 a whole, and considering all their relations. 



12. The present received classifications of 

 rocks are artificial, based on part of the 

 characters to the exclusion of others ; they 

 correspond to the Linnean artificial botanical 

 classification, and hold about the same rela- 

 tions to a natural classification of rocks as that 

 does to the natural classification of plants. 



13. The great mass of rocks separated from 

 one another as distinct species in these classi- 

 fications are mere varietal forms of certain 

 definite natural species, — the variation owing 

 to alteration, or to some little change in con- 

 ditions. 



Distinction should be made between superfi- 

 cial weathering and the chemical and molecular 

 changes that go on in all eruptive rocks after 



