June 29, 1883.] 



SCIENCE. 



591 



and therefore, according to him, true schists ; 

 hence he would argue they are the remains of 

 an extensive land area — an Atlantis — of 

 which, owing to denudation, onlj- these few 

 remains are left. H^s arguments are based on 

 the apparent microscopic schistose structure 

 of the St. Paul's rock, and on the fact that 

 certain olivine rocks have been found asso- 

 ciated with crystalline schists. 



In examining the first evidence, the writer 

 maj" state, that he has in his possession two 

 specimens of these rocks, sent him by Mr. 

 John Murray' ; and therefore the evidence that 

 their sections aflbrd will be given. The sec- 

 tions of one — the least altered — ■ are com- 

 posed of olivine, enstatite, diallage, picotite 

 or chromite, magnetite, pj'rite, actinolite, and 

 serpentine. 



While M. Eenard remarks that all the min- 

 erals have their principal axis parallel with the 

 supposed schistose or fluidal structure, thej^ 

 are found by me to stand in every direction 

 regarding that structure, — even at right an- 

 gles to one another. Indeed, no structure has 

 been seen by me that I can regard as truly 

 schistose or fluidal. 



A structure does exist somewhat resembling 

 * a schistose one, which appears to be the result 

 of secondary alteration of the rock-mass. M. 

 Renard states that the rock is fresh and unal- 

 tered in certain of the specimens, and one of 

 mine answers to his description of his sup- 

 posed unaltered rock. He further states, that 

 the structure of this rock is peculiar, and un- 

 like that of other olivine rocks. M-X sections 

 lead me to a somewhat different conclusion. 

 In them, portions were found that I regard as 

 the original, unaltered rock. These showed 

 the same structure and characters that other 

 unaltered olivine rocks show, and do not ap- 

 pear to be of anj' abnormal type. 



The main portion of the rock which M. Re- 

 nard regarded as groundmass, and held to be 

 unchanged, is, in my opinion, greatlj- altered, 

 and contains onlj- remnants of the original 

 minerals of the rock. He regards this ground- 

 mass as composed entirely of olivine grains, 

 but of this I have grave doubts. The micro- 

 scopic characters of this groundmass do not 

 appear to me to be those of ordinary olivine, 

 but rather those of minerals of secondar}- ori- 

 gin. That this groundmass is of secondar}' 

 origin, for the most part, is shown by its oc- 

 currence along the fissures in the unaltered 

 olivine ; hy its relations to the minerals that 

 it surrounds, — which relations are the same 

 as those existing between the original minerals 

 and their alteration products in other rocks ; 



and by the so-called schistose structure. I 

 do not design to call in question the work of 

 M. Renard, who is a thoroughly competent 

 observer, and whose sections have not been 

 seen, but rather to show that the characters 

 in my sections do not, in my judgment, bear 

 out the conclusion derived by him from his. 

 Questions of this kind are largely dependent 

 upon the methods of work and the kind of 

 study to which the observer has devoted him- 

 self. M. Renard has given much time to the 

 stud}' of crystalline schists and the older erup- 

 tive rocks ; while the present writer has, for a 

 number of years, devoted much of his time 

 to the study of unaltered rocks, and to the 

 tracing of their various types through to the 

 extreme phases of alteration, studying both 

 modern and ancient forms, with especial ref- 

 erence to their origin and development. It is 

 therefore natural that we should both look at 

 the St. Paul's rocks from a somewhat different 

 stand-point. 



But to continue. It is contrary to the laws 

 of phj-sics and chemistry, that a mineral in 

 the process of alteration should produce itself 

 again. Alteration is rather a passage from 

 an unstable to a more stable compound in the 

 conditions to which the rock is theu exposed : 

 hence the resulting mineral in this case must 

 belong either to another varietj- of olivine, 

 or to a distinct mineral sjDecies. This would 

 hold good if more than one mineral should be , 

 formed. 



The actinolite, picotite or chromite, magne- 

 tite, pyrite, and serpentine, are regarded by 

 me as secondar}- products in this case, and 

 not original minerals. 



In the places showing the unaltered condi- 

 tion of the rock, the granular structure is the 

 same as that believed to be due to crystalli- 

 zation from an igneous magma, and not owing 

 to detrital action. 



The specimens sent me show that they are 

 surface and weathered specimens, to which 

 cause is probably due much of the difficulty 

 met in their study. It is to be hoped, that, 

 should St. Paul rocks be visited again, great 

 pains will be taken to procure specimens as 

 far in the solid rock as possible. 



Microscopicall}-, then, the writer holds that 

 these rocks afford evidence in their structure 

 and composition favouing the view that they 

 are eruptive, while in his sections he can find 

 nothing supporting the theorj- of M. Renard. 



It now remains to look at the question of 

 the association of olivine rocks with schists as 

 proving that they are both of a common origin. 



This line of argument the writer had occa- 



